summaryrefslogtreecommitdiff
path: root/info/challenges/AroBend/AroundTheBend.tex
blob: ac01d7180751162a514a3dd54ca2067340493dac (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
4634
4635
4636
4637
4638
4639
4640
4641
4642
4643
4644
4645
4646
4647
4648
4649
4650
4651
4652
4653
4654
4655
4656
4657
4658
4659
4660
4661
4662
4663
4664
4665
4666
4667
4668
4669
4670
4671
4672
4673
4674
4675
4676
4677
4678
4679
4680
4681
4682
4683
4684
4685
4686
4687
4688
4689
4690
4691
4692
4693
4694
4695
4696
4697
4698
4699
4700
4701
4702
4703
4704
4705
4706
4707
4708
4709
4710
4711
4712
4713
4714
4715
4716
4717
4718
4719
4720
4721
4722
4723
4724
4725
4726
4727
4728
4729
4730
4731
4732
4733
4734
4735
4736
4737
4738
4739
4740
4741
4742
4743
4744
4745
4746
4747
4748
4749
4750
4751
4752
4753
4754
4755
4756
4757
4758
4759
4760
4761
4762
4763
4764
4765
4766
4767
4768
4769
4770
4771
4772
4773
4774
4775
4776
4777
4778
4779
4780
4781
4782
4783
4784
4785
4786
4787
4788
4789
4790
4791
4792
4793
4794
4795
4796
4797
4798
4799
4800
4801
4802
4803
4804
4805
4806
4807
4808
4809
4810
4811
4812
4813
4814
4815
4816
4817
4818
4819
4820
4821
4822
4823
4824
4825
4826
4827
4828
4829
4830
4831
4832
4833
4834
4835
4836
4837
4838
4839
4840
4841
4842
4843
4844
4845
4846
4847
4848
4849
4850
4851
4852
4853
4854
4855
4856
4857
4858
4859
4860
4861
4862
4863
4864
4865
4866
4867
4868
4869
4870
4871
4872
4873
4874
4875
4876
4877
4878
4879
4880
4881
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897
4898
4899
4900
4901
4902
4903
4904
4905
4906
4907
4908
4909
4910
4911
4912
4913
4914
4915
4916
4917
4918
4919
4920
4921
4922
4923
4924
4925
4926
4927
4928
4929
4930
4931
4932
4933
4934
4935
4936
4937
4938
4939
4940
4941
4942
4943
4944
4945
4946
4947
4948
4949
4950
4951
4952
4953
4954
4955
4956
4957
4958
4959
4960
4961
4962
4963
4964
4965
4966
4967
4968
4969
4970
4971
4972
4973
4974
4975
4976
4977
4978
4979
4980
4981
4982
4983
4984
4985
4986
4987
4988
4989
4990
4991
4992
4993
4994
4995
4996
4997
4998
4999
5000
5001
5002
5003
5004
5005
5006
5007
5008
5009
5010
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022
5023
5024
5025
5026
5027
5028
5029
5030
5031
5032
5033
5034
5035
5036
5037
5038
5039
5040
5041
5042
5043
5044
5045
5046
5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
5076
5077
5078
5079
5080
5081
5082
5083
5084
5085
5086
5087
5088
5089
5090
5091
5092
5093
5094
5095
5096
5097
5098
5099
5100
5101
5102
5103
5104
5105
5106
5107
5108
5109
5110
5111
5112
5113
5114
5115
5116
5117
5118
5119
5120
5121
5122
5123
5124
5125
5126
5127
5128
5129
5130
5131
5132
5133
5134
5135
5136
5137
5138
5139
5140
5141
5142
5143
5144
5145
5146
5147
5148
5149
5150
5151
5152
5153
5154
5155
5156
5157
5158
5159
5160
5161
5162
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
5179
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
5222
5223
5224
5225
5226
5227
5228
5229
5230
5231
5232
5233
5234
5235
5236
5237
5238
5239
5240
5241
5242
5243
5244
5245
5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
5264
5265
5266
5267
5268
5269
5270
5271
5272
5273
5274
5275
5276
5277
5278
5279
5280
5281
5282
5283
5284
5285
5286
5287
5288
5289
5290
5291
5292
5293
5294
5295
5296
5297
5298
5299
5300
5301
5302
5303
5304
5305
5306
5307
5308
5309
5310
5311
5312
5313
5314
5315
5316
5317
5318
5319
5320
5321
5322
5323
5324
5325
5326
5327
5328
5329
5330
5331
5332
5333
5334
5335
5336
5337
5338
5339
5340
5341
5342
5343
5344
5345
5346
5347
5348
5349
5350
5351
5352
5353
5354
5355
5356
5357
5358
5359
5360
5361
5362
5363
5364
5365
5366
5367
5368
5369
5370
5371
5372
5373
5374
5375
5376
5377
5378
5379
5380
5381
5382
5383
5384
5385
5386
5387
5388
5389
5390
5391
5392
5393
5394
5395
5396
5397
5398
5399
5400
5401
5402
5403
5404
5405
5406
5407
5408
5409
5410
5411
5412
5413
5414
5415
5416
5417
5418
5419
5420
5421
5422
5423
5424
5425
5426
5427
5428
5429
5430
5431
5432
5433
5434
5435
5436
5437
5438
5439
5440
5441
5442
5443
5444
5445
5446
5447
5448
5449
5450
5451
5452
5453
5454
5455
5456
5457
5458
5459
5460
5461
5462
5463
5464
5465
5466
5467
5468
5469
5470
5471
5472
5473
5474
5475
5476
5477
5478
5479
5480
5481
5482
5483
5484
5485
5486
5487
5488
5489
5490
5491
5492
5493
5494
5495
5496
5497
5498
5499
5500
5501
5502
5503
5504
5505
5506
5507
5508
5509
5510
5511
5512
5513
5514
5515
5516
5517
5518
5519
5520
5521
5522
5523
5524
5525
5526
5527
5528
5529
5530
5531
5532
5533
5534
5535
5536
5537
5538
5539
5540
5541
5542
5543
5544
5545
5546
5547
5548
5549
5550
5551
5552
5553
5554
5555
5556
5557
5558
5559
5560
5561
5562
5563
5564
5565
5566
5567
5568
5569
5570
5571
5572
5573
5574
5575
5576
5577
5578
5579
5580
5581
5582
5583
5584
5585
5586
5587
5588
5589
5590
5591
5592
5593
5594
5595
5596
5597
5598
5599
5600
5601
5602
5603
5604
5605
5606
5607
5608
5609
5610
5611
5612
5613
5614
5615
5616
5617
5618
5619
5620
5621
5622
5623
5624
5625
5626
5627
5628
5629
5630
5631
5632
5633
5634
5635
5636
5637
5638
5639
5640
5641
5642
5643
5644
5645
5646
5647
5648
5649
5650
5651
5652
5653
5654
5655
5656
5657
5658
5659
5660
5661
5662
5663
5664
5665
5666
5667
5668
5669
5670
5671
5672
5673
5674
5675
5676
5677
5678
5679
5680
5681
5682
5683
5684
5685
5686
5687
5688
5689
5690
5691
5692
5693
5694
5695
5696
5697
5698
5699
5700
5701
5702
5703
5704
5705
5706
5707
5708
5709
5710
5711
5712
5713
5714
5715
5716
5717
5718
5719
5720
5721
5722
5723
5724
5725
5726
5727
5728
5729
5730
5731
5732
5733
5734
5735
5736
5737
5738
5739
5740
5741
5742
5743
5744
5745
5746
5747
5748
5749
5750
5751
5752
5753
5754
5755
5756
5757
5758
5759
5760
5761
5762
5763
5764
5765
5766
5767
5768
5769
5770
5771
5772
5773
5774
5775
5776
5777
5778
5779
5780
5781
5782
5783
5784
5785
5786
5787
5788
5789
5790
5791
5792
5793
5794
5795
5796
5797
5798
5799
5800
5801
5802
5803
5804
5805
5806
5807
5808
5809
5810
5811
5812
5813
5814
5815
5816
5817
5818
5819
5820
5821
5822
5823
5824
5825
5826
5827
5828
5829
5830
5831
5832
5833
5834
5835
5836
5837
5838
5839
5840
5841
5842
5843
5844
5845
5846
5847
5848
5849
5850
5851
5852
5853
5854
5855
5856
5857
5858
5859
5860
5861
5862
5863
5864
5865
5866
5867
5868
5869
5870
5871
5872
5873
5874
5875
5876
5877
5878
5879
5880
5881
5882
5883
5884
5885
5886
5887
5888
5889
5890
5891
5892
5893
5894
5895
5896
5897
5898
5899
5900
5901
5902
5903
5904
5905
5906
5907
5908
5909
5910
5911
5912
5913
5914
5915
5916
5917
5918
5919
5920
5921
5922
5923
5924
5925
5926
5927
5928
5929
5930
5931
5932
5933
5934
5935
5936
5937
5938
5939
5940
5941
5942
5943
5944
5945
5946
5947
5948
5949
5950
5951
5952
5953
5954
5955
5956
5957
5958
5959
5960
5961
5962
5963
5964
5965
5966
5967
5968
5969
5970
5971
5972
5973
5974
5975
5976
5977
5978
5979
5980
5981
5982
5983
5984
5985
5986
5987
5988
5989
5990
5991
5992
5993
5994
5995
5996
5997
5998
5999
6000
6001
6002
6003
6004
6005
6006
6007
6008
6009
6010
6011
6012
6013
6014
6015
6016
6017
6018
6019
6020
6021
6022
6023
6024
6025
6026
6027
6028
6029
6030
6031
6032
6033
6034
6035
6036
6037
6038
6039
6040
6041
6042
6043
6044
6045
6046
6047
6048
6049
6050
6051
6052
6053
6054
6055
6056
6057
6058
6059
6060
6061
6062
6063
6064
6065
6066
6067
6068
6069
6070
6071
6072
6073
6074
6075
6076
6077
6078
6079
6080
6081
6082
6083
6084
6085
6086
6087
6088
6089
6090
6091
6092
6093
6094
6095
6096
6097
6098
6099
6100
6101
6102
6103
6104
6105
6106
6107
6108
6109
6110
6111
6112
6113
6114
6115
6116
6117
6118
6119
6120
6121
6122
6123
6124
6125
6126
6127
6128
6129
6130
6131
6132
6133
6134
6135
6136
6137
6138
6139
6140
6141
6142
6143
6144
6145
6146
6147
6148
6149
6150
6151
6152
6153
6154
6155
6156
6157
6158
6159
6160
6161
6162
6163
6164
6165
6166
6167
6168
6169
6170
6171
6172
6173
6174
6175
6176
6177
6178
6179
6180
6181
6182
6183
6184
6185
6186
6187
6188
6189
6190
6191
6192
6193
6194
6195
6196
6197
6198
6199
6200
6201
6202
6203
6204
6205
6206
6207
6208
6209
6210
6211
6212
6213
6214
6215
6216
6217
6218
6219
6220
6221
6222
6223
6224
6225
6226
6227
6228
6229
6230
6231
6232
6233
6234
6235
6236
6237
6238
6239
6240
6241
6242
6243
6244
6245
6246
6247
6248
6249
6250
6251
6252
6253
6254
6255
6256
6257
6258
6259
6260
6261
6262
6263
6264
6265
6266
6267
6268
6269
6270
6271
6272
6273
6274
6275
6276
6277
6278
6279
6280
6281
6282
6283
6284
6285
6286
6287
6288
6289
6290
6291
6292
6293
6294
6295
6296
6297
6298
6299
6300
6301
6302
6303
6304
6305
6306
6307
6308
6309
6310
6311
6312
6313
6314
6315
6316
6317
6318
6319
6320
6321
6322
6323
6324
6325
6326
6327
6328
6329
6330
6331
6332
6333
6334
6335
6336
6337
6338
6339
6340
6341
6342
6343
6344
6345
6346
6347
6348
6349
6350
6351
6352
6353
6354
6355
6356
6357
6358
6359
6360
6361
6362
6363
6364
6365
6366
6367
6368
6369
6370
6371
6372
6373
6374
6375
6376
6377
6378
6379
6380
6381
6382
6383
6384
6385
6386
6387
6388
6389
6390
6391
6392
6393
6394
6395
6396
6397
6398
6399
6400
6401
6402
6403
6404
6405
6406
6407
6408
6409
6410
6411
6412
6413
6414
6415
6416
6417
6418
6419
6420
6421
6422
6423
6424
6425
6426
6427
6428
6429
6430
6431
6432
6433
6434
6435
6436
6437
6438
6439
6440
6441
6442
6443
6444
6445
6446
6447
6448
6449
6450
6451
6452
6453
6454
6455
6456
6457
6458
6459
6460
6461
6462
6463
6464
6465
6466
6467
6468
6469
6470
6471
6472
6473
6474
6475
6476
6477
6478
6479
6480
6481
6482
6483
6484
6485
6486
6487
6488
6489
6490
6491
6492
6493
6494
6495
6496
6497
6498
6499
6500
6501
6502
6503
6504
6505
6506
6507
6508
6509
6510
6511
6512
6513
6514
6515
6516
6517
6518
6519
6520
6521
6522
6523
6524
6525
6526
6527
6528
6529
6530
6531
6532
6533
6534
6535
6536
6537
6538
6539
6540
6541
6542
6543
6544
6545
6546
6547
6548
6549
6550
6551
6552
6553
6554
6555
6556
6557
6558
6559
6560
6561
6562
6563
6564
6565
6566
6567
6568
6569
6570
6571
6572
6573
6574
6575
6576
6577
6578
6579
6580
6581
6582
6583
6584
6585
6586
6587
6588
6589
6590
6591
6592
6593
6594
6595
6596
6597
6598
6599
6600
6601
6602
6603
6604
6605
6606
6607
6608
6609
6610
6611
6612
6613
6614
6615
6616
6617
6618
6619
6620
6621
6622
6623
6624
6625
6626
6627
6628
6629
6630
6631
6632
6633
6634
6635
6636
6637
6638
6639
6640
6641
6642
6643
6644
6645
6646
6647
6648
6649
6650
6651
6652
6653
6654
6655
6656
6657
6658
6659
6660
6661
6662
6663
6664
6665
6666
6667
6668
6669
6670
6671
6672
6673
6674
6675
6676
6677
6678
6679
6680
6681
6682
6683
6684
6685
6686
6687
6688
6689
6690
6691
6692
6693
6694
6695
6696
6697
6698
6699
6700
6701
6702
6703
6704
6705
6706
6707
6708
6709
6710
6711
6712
6713
6714
6715
6716
6717
6718
6719
6720
6721
6722
6723
6724
6725
6726
6727
6728
6729
6730
6731
6732
6733
6734
6735
6736
6737
6738
6739
6740
6741
6742
6743
6744
6745
6746
6747
6748
6749
6750
6751
6752
6753
6754
6755
6756
6757
6758
6759
6760
6761
6762
6763
6764
6765
6766
6767
6768
6769
6770
6771
6772
6773
6774
6775
6776
6777
6778
6779
6780
6781
6782
6783
6784
6785
6786
6787
6788
6789
6790
6791
6792
6793
6794
6795
6796
6797
6798
6799
6800
6801
6802
6803
6804
6805
6806
6807
6808
6809
6810
6811
6812
6813
6814
6815
6816
6817
6818
6819
6820
6821
6822
6823
6824
6825
6826
6827
6828
6829
6830
6831
6832
6833
6834
6835
6836
6837
6838
6839
6840
6841
6842
6843
6844
6845
6846
6847
6848
6849
6850
6851
6852
6853
6854
6855
6856
6857
6858
6859
6860
6861
6862
6863
6864
6865
6866
6867
6868
6869
6870
6871
6872
6873
6874
6875
6876
6877
6878
6879
6880
6881
6882
6883
6884
6885
6886
6887
6888
6889
6890
6891
6892
6893
6894
6895
6896
6897
6898
6899
6900
6901
6902
6903
6904
6905
6906
6907
6908
6909
6910
6911
6912
6913
6914
6915
6916
6917
6918
6919
6920
6921
6922
6923
6924
6925
6926
6927
6928
6929
6930
6931
6932
6933
6934
6935
6936
6937
6938
6939
6940
6941
6942
6943
6944
6945
6946
6947
6948
6949
6950
6951
6952
6953
6954
6955
6956
6957
6958
6959
6960
6961
6962
6963
6964
6965
6966
6967
6968
6969
6970
6971
6972
6973
6974
6975
6976
6977
6978
6979
6980
6981
6982
6983
6984
6985
6986
6987
6988
6989
6990
6991
6992
6993
6994
6995
6996
6997
6998
6999
7000
7001
7002
7003
7004
7005
7006
7007
7008
7009
7010
7011
7012
7013
7014
7015
7016
7017
7018
7019
7020
7021
7022
7023
7024
7025
7026
7027
7028
7029
7030
7031
7032
7033
7034
7035
7036
7037
7038
7039
7040
7041
7042
7043
7044
7045
7046
7047
7048
7049
7050
7051
7052
7053
7054
7055
7056
7057
7058
7059
7060
7061
7062
7063
7064
7065
7066
7067
7068
7069
7070
7071
7072
7073
7074
7075
7076
7077
7078
7079
7080
7081
7082
7083
7084
7085
7086
7087
7088
7089
7090
7091
7092
7093
7094
7095
7096
7097
7098
7099
7100
7101
7102
7103
7104
7105
7106
7107
7108
7109
7110
7111
7112
7113
7114
7115
7116
7117
7118
7119
7120
7121
7122
7123
7124
7125
7126
7127
7128
7129
7130
7131
7132
7133
7134
7135
7136
7137
7138
7139
7140
7141
7142
7143
7144
7145
7146
7147
7148
7149
7150
7151
7152
7153
7154
7155
7156
7157
7158
7159
7160
7161
7162
7163
7164
7165
7166
7167
7168
7169
7170
7171
7172
7173
7174
7175
7176
7177
7178
7179
7180
7181
7182
7183
7184
7185
7186
7187
7188
7189
7190
7191
7192
7193
7194
7195
7196
7197
7198
7199
7200
7201
7202
7203
7204
7205
7206
7207
7208
7209
7210
7211
7212
7213
7214
7215
7216
7217
7218
7219
7220
7221
7222
7223
7224
7225
7226
7227
7228
7229
7230
7231
7232
7233
7234
7235
7236
7237
7238
7239
7240
7241
7242
7243
7244
7245
7246
7247
7248
7249
7250
7251
7252
7253
7254
7255
7256
7257
7258
7259
7260
7261
7262
7263
7264
7265
7266
7267
7268
7269
7270
7271
7272
7273
7274
7275
7276
7277
7278
7279
7280
7281
7282
7283
7284
7285
7286
7287
7288
7289
7290
7291
7292
7293
7294
7295
7296
7297
7298
7299
7300
7301
7302
7303
7304
7305
7306
7307
7308
7309
7310
7311
7312
7313
7314
7315
7316
7317
7318
7319
7320
7321
7322
7323
7324
7325
7326
7327
7328
7329
7330
7331
7332
7333
7334
7335
7336
7337
7338
7339
7340
7341
7342
7343
7344
7345
7346
7347
7348
7349
7350
7351
7352
7353
7354
7355
7356
7357
7358
7359
7360
7361
7362
7363
7364
7365
7366
7367
7368
7369
7370
7371
7372
7373
7374
7375
7376
7377
7378
7379
7380
7381
7382
7383
7384
7385
7386
7387
7388
7389
7390
7391
7392
7393
7394
7395
7396
7397
7398
7399
7400
7401
7402
7403
7404
7405
7406
7407
7408
7409
7410
7411
7412
7413
7414
7415
7416
7417
7418
7419
% AroundTheBend.tex      concatenation of Around The Bend

\begin{filecontents}{bend.ist}
% MakeIndex style file bend.ist for use with AroundTheBend.tex

% @ may be a valid character in the index, use ? instead
actual '?'

\end{filecontents}

%\documentclass[draft,openany]{memoir}
\documentclass[openany]{memoir}
\usepackage{comment}
\usepackage{url}
\ifpdf
  \usepackage[pdftex,
              plainpages=false,
              pdfpagelabels,
              bookmarksnumbered
             ]{hyperref}
\else
  \usepackage[%pdf,
              plainpages=false,
              pdfpagelabels,
              bookmarksnumbered
             ]{hyperref}
\fi
\usepackage{graphicx}

\settrimmedsize{11in}{210mm}{*}% min letterpaper/A4 sizes
\setlength{\trimtop}{0pt}
\setlength{\trimedge}{\stockwidth}
\addtolength{\trimedge}{-\paperwidth}
\settypeblocksize{7.75in}{33pc}{*}
\setulmargins{4cm}{*}{*}
\setlrmargins{1.25in}{*}{*}
\setmarginnotes{17pt}{51pt}{\onelineskip}
\setheadfoot{\onelineskip}{2\onelineskip}
\setheaderspaces{*}{2\onelineskip}{*}
\checkandfixthelayout

%\addtolength{\textwidth}{1in}
%\addtolength{\oddsidemargin}{-0.5in}
%\addtolength{\evensidemargin}{-0.5in}

\newcommand{\ed}[1]{\emph{(Ed: #1)}}
\newcommand*{\oposted}[1]{Originally posted on #1}
\newcommand*{\arch}[1]{Archived as {\normalfont \ttfamily #1}}
\newenvironment{solution}[1]{%
  \begin{description}
  \item[#1]\mbox{}}%
%  {\par\noindent\textbf{End solution}\end{description}}
  {\end{description}\vspace{-0.5\onelineskip}\textbf{End solution}}
\newcommand*{\pfile}[1]{\texttt{#1}}%  print a file name
\newfixedcaption{\freetabcaption}{table}
\renewcommand*{\chaptername}{QA}
\renewcommand*{\chaptername}{}

% \piif{if...} print and index \if...
\newcommand*{\piif}[1]{\cs{#1}\index{#1?\cs{#1}}}
\makeatletter
\newcommand*{\zeroseps}{%
  \topsep\z@
  \partopsep\z@
  \parskip\z@}
\newlength{\gparindent} \gparindent 0.5\parindent
\newenvironment{lcode}{\zeroseps
  \renewcommand{\verbatim@startline}%
    {\verbatim@line{\hskip\gparindent}}%
  \small\setlength{\baselineskip}{\onelineskip}\verbatim}%
  {\endverbatim
   \vspace{-\baselineskip}\noindent}
\makeatother

\nouppercaseheads
\headstyles{bringhurst}
%\setlength{\beforechapskip}{2\onelineskip}
\chapterstyle{section}
\setlength{\beforechapskip}{2\onelineskip}
\setlength{\beforechapskip}{0pt}
\setlength{\afterchapskip}{1\onelineskip}
\settocdepth{subsubsection}
\setsecnumdepth{subsubsection}

\makeindex

%\title{Around The Bend}
%\author{Michael Downes \\
%(edited by Peter Wilson)}
%\date{}
\newlength{\drop}
\providecommand*{\wb}[2]{\fontsize{#1}{#2}\usefont{U}{webo}{xl}{n}}
\newcommand*{\titleAB}{\begingroup
\drop=4\baselineskip
\centering
\vspace*{\drop}
{\Huge AROUND THE BEND}\\[\drop]
{\hspace*{1.5em}\scalebox{8}[1]{{\wb{10}{12}4}}}\\[\drop]
{\Large\itshape A Collection of TeX Challenges by}\\[\baselineskip]
{\Large MICHAEL DOWNES}\\[\baselineskip]
{\wb{10}{12}4}\\[\baselineskip]
{\Large\itshape edited by}\\[\baselineskip]
{\Large Peter Wilson}\par
\vfill
{\hspace*{1.5em}\scalebox{8}[1]{{\wb{10}{12}4}}}\\[\drop]
{\large The Herries Press}\\
{July 2008}\par
\vspace*{\drop}
\endgroup}
%% normally \parindent = 1.5em, but 0pt in \titleAB

\begin{document}
\tightlists
\raggedbottom

\frontmatter

%\maketitle
\thispagestyle{empty}
\titleAB
\cleardoublepage
\tableofcontents

\chapter{Preface}

    In the early 90's the late and much missed Michael Downes (1958--2003)
ran a column in the INFO-TeX mailing list
called \emph{Around The Bend} where he proposed macro-related problems and 
then posted
submitted solutions. Although it was archived on CTAN in \url{info/aro-bend}
it is not well known which is a shame as it provides
answers to many problems that keep cropping up. (The archive is now
at \url{info/challenges/aro-bend}). This is an attempt to
make his work more accessible by providing the collection as a single 
document. 

    As much as possible what follows is what Michael wrote; I have tried to 
limit myself to marking up the original ASCII text emails but I have not
repeated administrative elements such as email headers. 

    In some cases the 
original TeX code was replete with comments explaining what was going on.
Where the comments were long with respect to the code I have set them in
the regular body type so as to make the actual code more obvious; this has a
side effect of slightly decreasing the amount of paper required to
print the document. If you 
want to use the code solutions I suggest that you cut and paste them
from the original archived versions.

    I thought that there were eighteen Around the Bends as that is all that
are archived on CTAN. However I googled the Google Groups \url{comp.text.tex} 
group
and found three more, nos.~19, 20 and~21. I have included what I could find
of these, but answers to no.~19 appear to be missing, which is a pity as
I think that I could have put them to use. Perhaps some of you might be 
willing to take up the challenge on this, or on any of the others.

    

{\raggedleft \textsc{PW}\\ July, 2008 \par}

\chapter{Introduction}

\ed{This is Michael's introduction to his scheme, originally posted on
1991/10/10 as the initial portion of exercise~1.}


%%[Exercises 1,2,3 were originally posted together on 10 Oct 91]
\begin{verbatim}
Date: Thu 10 Oct 91 09:51:32-EST
From: Michael Downes <MJD@MATH.AMS.COM>
Subject: Around the bend
To: info-tex@shsu.edu
\end{verbatim}

Proposal for a regular feature:

  AROUND THE BEND

With the encouragement of George Greenwade (the INFO-TeX list owner), I
would like to propose a regular department for INFO-TeX, called `Around
the bend'. It will consist of macro-writing challenges on the level of
the dangerous-bend exercises in the \emph{TeXbook}, with interested parties
invited to collaborate and/or compete to find the best solution. My
motivation for doing this is partly selfish: to get more feedback from
other macro writers about some of the interesting macro-writing
problems that I run into.

I originally approached George for advice about setting up a separate
mailing list, but he thought that INFO-TeX and comp.text.tex readers
would be interested. Since INFO-TeX mail is also channeled to
comp.text.tex, readers of the latter should let me know if they don't
want the extra traffic (although I don't expect it to be that much). I
don't currently have access to read comp.text.tex directly, although
George has been investigating the possibility of piping it through the
INFO-TeX mailing list. So if you object by posting to comp.text.tex, I
may not see your objection; send me mail, instead.

The sample below should give a pretty good idea of what `Around the
bend' would be like. Solutions should be sent to me instead of to
INFO-TeX or comp.text.tex, on the premise that people usually won't want
to read others' solutions until they've had a chance to try their own
hand. A summary of the results would then be posted to the INFO-TeX
list after two or three weeks; to those who submit solutions before the
deadline, I could forward without delay solutions submitted by other
people, for comparison.

I will try to keep the difficulty of the exercises down to something
reasonable, let's say, on the level of a homework assignment which a
university student must complete in two weeks, finding time in the
normal way from the usual busy schedule of other homework, class
attendance, sports, and social life. However, be warned that the
challenges will be hard. I'm planning to follow a `hard and fast'
format: one or two hard questions, followed by one or two fast
questions, where if you don't know the answer off the top of your head,
you can either look it up in the \emph{TeXbook} or find it by running a quick
test.


\mainmatter


\chapter{Expansion}

\section{Exercise (hard)}

%%\input{ex001.tex}
% ex001.tex

\begin{comment}

(Originally posted on 1991/10/10)

[Exercises 1,2,3 were originally posted together on 10 Oct 91]
Date: Thu 10 Oct 91 09:51:32-EST
From: Michael Downes <MJD@MATH.AMS.COM>
Subject: Around the bend
To: info-tex@shsu.edu

Proposal for a regular feature:

  AROUND THE BEND

With the encouragement of George Greenwade (the INFO-TeX list owner), I
would like to propose a regular department for INFO-TeX, called `Around
the bend'. It will consist of macro-writing challenges on the level of
the dangerous-bend exercises in the TeXbook, with interested parties
invited to collaborate and/or compete to find the best solution. My
motivation for doing this is partly selfish: to get more feedback from
other macro writers about some of the interesting macro-writing
problems that I run into.

I originally approached George for advice about setting up a separate
mailing list, but he thought that INFO-TeX and comp.text.tex readers
would be interested. Since INFO-TeX mail is also channeled to
comp.text.tex, readers of the latter should let me know if they don't
want the extra traffic (although I don't expect it to be that much). I
don't currently have access to read comp.text.tex directly, although
George has been investigating the possibility of piping it through the
INFO-TeX mailing list. So if you object by posting to comp.text.tex, I
may not see your objection; send me mail, instead.

The sample below should give a pretty good idea of what `Around the
bend' would be like. Solutions should be sent to me instead of to
INFO-TeX or comp.text.tex, on the premise that people usually won't want
to read others' solutions until they've had a chance to try their own
hand. A summary of the results would then be posted to the INFO-TeX
list after two or three weeks; to those who submit solutions before the
deadline, I could forward without delay solutions submitted by other
people, for comparison.

I will try to keep the difficulty of the exercises down to something
reasonable, let's say, on the level of a homework assignment which a
university student must complete in two weeks, finding time in the
normal way from the usual busy schedule of other homework, class
attendance, sports, and social life. However, be warned that the
challenges will be hard. I'm planning to follow a `hard and fast'
format: one or two hard questions, followed by one or two fast
questions, where if you don't know the answer off the top of your head,
you can either look it up in the TeXbook or find it by running a quick
test.

All right, here are the first three.

\end{comment}

%**********************************************************************
%*** Exercise 1 (hard):

\ed{\oposted{1991/10/10}. \arch{exercise.001}.}\\%[0.5\baselineskip]

Given arbitrary \cmd{\b}, \cmd{\c}, \cmd{\d} (macros without arguments), for example
\begin{lcode}
       \def\b{\c\c}         \def\c{*}            \def\d{\b\c}
\end{lcode}
figure out how to define \cmd{\a} so that its replacement text consists
of \cmd{\b} fully expanded plus \cmd{\c} not expanded plus \cmd{\d} expanded 
exactly once.
I.e., with the above definitions the replacement text of \cmd{\a}
should be 
\begin{lcode}
       **\c\b\c
\end{lcode}
You may not use \cmd{\the} or \cmd{\noexpand} in your solution. This is Exercise
20.16 in the \emph{TeXbook}, except that there's an added restriction: your
answer must also not use the \cmd{\halign}\texttt{\ldots}\cmd{\span} method given in the
answer to 20.16. (Yes, that means you can't use \cmd{\valign} either!)

Why would anyone want to do such a hard exercise? Answer: advanced
macro writing requires a thorough knowledge of expansion control
principles.

\begin{comment}
[Exercise 2 moved to exercise.002]

[Exercise 3 moved to exercise.003]

Send answers to:

Michael Downes      mjd@math.ams.com (Internet)

A summary will be posted Friday, October 25, 1991.
\end{comment}

%%\endinput


\section{Answers}

%%\input{ans001.tex}
% ans001.tex

\ed{\oposted{1991/10/25}. \arch{answer.001}.}\\


\begin{comment}
[Solutions for exercises 1,2,3 were originally posted together on 25 Oct 91]
Date: Fri 25 Oct 91 15:19:44-EST
From: Michael Downes <MJD@MATH.AMS.COM>
Subject: `Around the bend' #1 solutions
To: info-tex@shsu.edu

Solutions to the exercises of `Around the bend' #1.

"*** Exercise 1 (hard):
"Given arbitrary \b, \c, \d (macros without arguments), for example
"
"       \def\b{\c\c}         \def\c{*}            \def\d{\b\c}
"
"figure out how to define \a so that its replacement text consists
"of \b fully expanded plus \c not expanded plus \d expanded exactly once.
"I.e., with the above definitions the replacement text of \a
"should be
"
"       **\c\b\c
"
"You may not use \the or \noexpand in your solution. This is Exercise
"20.16 in the TeXbook, except that there's an added restriction: your
"answer must also not use the \halign ... \span method given in the
"answer to 20.16. (Yes, that means you can't use \valign either!)
\end{comment}

The restrictions leave us with (essentially) three expansion-control
commands: \\
 \cmd{\expandafter}, \cmd{\edef} and \cmd{\def}.

%\begin{description}
%\item[Solution 1 {[Peter Schmitt]}] \mbox{}
\begin{solution}{Solution 1 (Peter Schmitt)}\index{Schmitt, Peter}
\begin{lcode}
   \edef\B{\b}
   \def\defA#1{\def\defa##1##2{\def\a{#1##2##1}}}
   \expandafter\defA\expandafter{\B}
   \expandafter\defa\expandafter{\d}{\c}
\end{lcode}
\end{solution}

%%>>EndSolution

%\item[Solution 2 {[Donald Arseneau]}] \mbox{}
\begin{solution}{Solution 2 (Donald Arseneau)}\index{Arseneau, Donald}
\begin{lcode}
\edef\e{\b}
\expandafter \expandafter \expandafter \def\expandafter \expandafter
\expandafter \a\expandafter \expandafter \expandafter {\expandafter
\e\expandafter \c\d}
\end{lcode}
\end{solution}
%%>>EndSolution

%\item[Solution 3 {[mine]}] \mbox{}
\begin{solution}{Solution 3 (mine)}\index{Downes, Michael}
\begin{lcode}
\edef\a{\b}
\expandafter\expandafter\expandafter\def
\expandafter\expandafter\expandafter\a
\expandafter\expandafter\expandafter{\expandafter\a\expandafter\c\d}
\end{lcode}
\end{solution}
%%>>EndSolution
%\end{description}

My solution differed from Arseneau's only in using \cmd{\a} rather than \cmd{\e}
in the first step.

\begin{comment}
[Solution for exercise 2 moved to answer.002]
[Solution for exercise 3 moved to answer.003]

Michael Downes                mjd@math.ams.com (Internet)

\end{comment}

%%\endinput


\chapter{Empty argument}

\section{Exercise (hard)}

%%\input{ex002.tex}
% ex002.tex

\begin{comment}

[Posted to info-tex on 10 Oct 91; see exercise.001]
**********************************************************************
*** Exercise 2 (hard):
\end{comment}

\ed{\oposted{1991/10/10}. \arch{exercise.002}.}\\

Define an `ifempty' macro that takes one argument and resolves
essentially to \piif{iftrue} if the argument is empty, and \piif{iffalse}
otherwise. This is useful for handling arguments given by
users to commands defined in a macro package.

Plain TeX or LaTeX-style solutions are both acceptable, that
is,
\begin{lcode}
     \ifempty{...}TRUE CASE\else FALSE CASE\fi
\end{lcode}
or
\begin{lcode}
     \ifempty{...}{TRUE CASE}{FALSE CASE}
\end{lcode}

(In the former case you will need to do something to avoid problems
in the situation
\begin{lcode}
 \iffalse ... \ifempty{...} ... \fi ... \fi
\end{lcode}
there
are different possibilities here, so I will refrain from
indicating any particular one.)

Use the following test suite to verify the robustness of your
solution:

\begin{lcode}
\long\def\test#1{\begingroup \toks0{[#1]}%
  \newlinechar`\/\message{/\the\toks0:
%    LaTeX-style solution; modify the following line according
%    to the syntax of your solution.
  \ifempty{#1}{EMPTY}{NOT empty}%
}\endgroup}

 \test{}                            \test{ }
 \test{aabc}                        \test{-}
 \test{$}                           \test{\empty}
 \test{\endinput}                   \test{\iftrue a\else b\fi}
 \test{\else}                       \test{#}
 \test{\par}                        \halign{#\cr\test{&}\cr}
 \test{\relax}                      \test{\relax\relax\relax}
 \expandafter\iffalse\test{x}\fi    \test{{}}
\end{lcode}
%$

The two tests on the first line should produce a message `EMPTY' and
the remaining ones, `NOT empty'. The reason for saying that the second
test should return `EMPTY' is that (1) this is the ideal behavior for
the applications I've encountered so far; (2) at least one other person
working independently arrived before me at a solution essentially
identical to mine, including this behavior. The details and credit to
the other guy will be given at solution time.

%%\endinput


\section{Answers}

%%\input{ans002.tex}
% ans002.tex

\begin{comment}

[Posted to info-tex on 25 Oct 91; see answer.001]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
"*** Exercise 2 (hard):
"Define an "ifempty" macro that takes one argument and resolves
"essentially to \iftrue if the argument is empty, and \iffalse
"otherwise. This is useful for handling arguments given by
"users to commands defined in a macro package such as LaTeX.
"
"Plain TeX or LaTeX-style solutions are both acceptable, that
"is,
"
"     \ifempty{...}TRUE CASE\else FALSE CASE\fi
"
"or
"
"     \ifempty{...}{TRUE CASE}{FALSE CASE}
\end{comment}

\ed{\oposted{1991/10/25}. \arch{answer.002}.}\\

The LaTeX-style solution that I had prepared was, I thought, pretty
good, but Donald Arseneau\index{Arseneau, Donald} 
observed that it fails the test
\begin{lcode}
\test{{\iftrue a\else b\fi}}
\end{lcode}
which was not in my list of tests.

%\begin{description}
%\item[Solution 1 {[mine]}] \mbox{}
\begin{solution}{Solution 1 (mine)}\index{Downes, Michael}
\begin{lcode}
\catcode`\@=11
%    \@car is actually already defined in latex.tex, but for
%    maximum robustness it needs to have the \long prefix:
\long\def\@car#1#2\@nil{#1}
\long\def\@first#1#2{#1}
\long\def\@second#1#2{#2}
\long\def\ifempty#1{\expandafter\ifx\@car#1@\@nil @\@empty
  \expandafter\@first\else\expandafter\@second\fi}
\catcode`\@=12

\long\def\test#1{\begingroup \toks0{[#1]}%
  \newlinechar`\/\message{/\the\toks0:
  \ifempty{#1}{EMPTY}{NOT empty}%
}\endgroup}
\end{lcode}
\end{solution}
%%>>EndSolution

The advantage of using the auxiliary macros \cmd{\@first} and \cmd{\@second},
together with the \cmd{\expandafter}'s, is that it allows the true and/or
false cases to end with arbitrary things, even macros that require
arguments that have not yet been read (any number of arguments, even
delimited arguments).

From here it is easy to implement an \piif{ifnotempty} test that has a
null false case. This is often useful in dealing with user-supplied
arguments: `If \#1 is empty, do nothing; otherwise, do the following
with \#1: ...'
\begin{lcode}
\long\def\ifnotempty#1{\ifempty{#1}{}}
\end{lcode}


%\item[Solution 2 {[Donald Arseneau]}]
\begin{solution}{Solution 2 (Donald Arseneau)}\index{Arseneau, Donald}
Don Arseneau came up with a plain TeX style solution, using an
ingenious device with \cmd{\then} to pass the test case
\begin{lcode}
\expandafter\iffalse\test{x}\fi
\end{lcode}
 The comments in the solution are his.

\begin{lcode}
% \ifblank{...}\then Test if a parameter is blank (null or spaces).
% Use the inaccessable "letter" @ to separate parameters. The two cases are:
%  _text_is_not_blank_             _text_is_blank_
%  #1<- whatever                   #1<-@
%  #2<- whatever (possibly null)   #2<-
%  #3<- @                          #3<-.
%  #4<- ..                         #4<-.
%  \if @.. {false}                 \if .. {true}
%  In the {false} case, the extra period is skipped so it doesn't hurt.

\catcode`\@=11 % as in plain.tex
\let\then\iftrue
\long\def\ifblank#1\then{\Ifbl@nk#1@@..\then}%
\long\def\Ifbl@nk#1#2@#3#4\then{\if#3#4}
\catcode`\@=12

\long\def\test#1{\begingroup \toks0{[#1]}%
  \newlinechar`\/\message{/\the\toks0:
  \ifblank{#1}\then EMPTY\else NOT empty\fi%
}\endgroup}
\end{lcode}
\end{solution}
%%>>EndSolution

The good thing about this solution is that it doesn't subject any part
of the user-supplied argument to the \piif{ifx} test. Using @ with category
code of 11 as a delimiter for the user-supplied text is extremely safe
because even in internal code @ doesn't appear by itself, only as part
of control sequence names. In a partial solution, 
Peter Schmitt\index{Schmitt, Peter} pushed
the idea a little further by using space with category code 3 as the
delimiter.

There is another way of handling the problematic \piif{iffalse} test, in a
plain-TeX style solution, by using a suggestion of Donald Knuth that
appeared in TeXhax a while ago, in reply to a query of Stephan von
Bechtolsheim (texhax89, \#38 (post from svb, 17 Apr 89)).

%\item[Solution 3 {[Arseneau/Knuth]}] \mbox{}
\begin{solution}{Solution 3 (Arseneau/Knuth)}\index{Arseneau, Donald}\index{Knuth, Donald}
\begin{lcode}
% Usage: \if\blank{#1}...\else...\fi

\catcode`\@=11 % as in plain.tex
\long\def\blank#1{\bl@nk#1@@..\bl@nk}%
\long\def\bl@nk#1#2@#3#4\bl@nk{#3#4}
\catcode`\@=12

\long\def\test#1{\begingroup \toks0{[#1]}%
  \newlinechar`\/\message{/\the\toks0:
  \if\blank{#1}EMPTY\else NOT empty\fi%
}\endgroup}
\end{lcode}
\end{solution}

%>>EndSolution

At the end of Exercise 2 I wrote:
\begin{quote}
The two tests on the first line should produce a message `EMPTY' and
the remaining ones, `NOT empty'. The reason for saying that the second
test should return `EMPTY' is that (1) this is the ideal behavior for
the applications I've encountered so far; (2) at least one other person
working independently arrived before me at a solution essentially
identical to mine, including this behavior. The details and credit to
the other guy will be given at solution time.
\end{quote}

The name of the `other guy' is Michael Wester\index{Wester, Michael}; 
a listing of his macros
was published in the preprints for the July 1991 TUG meeting in Dedham,
Massachusetts (`Form Letter in LaTeX with 3-across Mailing Labels
Capability', joint paper with Jackie Damrau). In rereading the preprint
recently, it seems to me the presentation is more different from
Exercise 2 and its solutions than I had previously imagined, but the
essential ideas are there. See \cmd{\wcar}, \cmd{\wcdr} and related macros.

By the way, if anyone came up with a fully expandable test (suitable
for use inside a \cmd{\message}) for which \verb?\test{ }? came up 
false instead of
true, I would be interested to hear about it. I didn't mean to
eliminate that possibility in my original statement of the problem.

%%\endinput


\chapter{Discretionary}

\section{Exercise (fast)}

%%\input{ex003.tex}
% ex003.tex

\ed{\oposted{1991/10/10}. \arch{exercise.003}.}\\

\begin{comment}

[Posted to info-tex on 10 Oct 91; see exercise.001]
**********************************************************************
*** Exercise 3 (fast):
\end{comment}

What's the most important difference between \cs{-} and
\begin{lcode}
\discretionary{-}{}{} ?
\end{lcode}

%%**********************************************************************
%%\endinput


\section{Answers}

%%\input{ans003.tex}
% ans003.tex

\ed{\oposted{1991/10/25}. \arch{answer.003}.}\\

\begin{comment}

[Posted to info-tex on 25 Oct 91; see answer.001]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
"*** Exercise 3 (fast):
"What's the most important difference between \- and
"\discretionary{-}{}{}?
\end{comment}

The most important difference between \cs{-} and \cmd{\discretionary}\verb?{-}{}{}?
is that the latter always puts in the character from font position 45
("2D, '55) of the current font when a word must be broken at the end of
a line; \cs{-} puts in the character from font position \cmd{\hyphenchar} of the
current font, which is NOT NECESSARILY position 45. It would be rather
unusual for \cmd{\hyphenchar} to be something other than 45; in certain
special applications, however (possibly in some foreign languages as
well?) a variant value of \cmd{\hyphenchar} can be useful. I have an idea for
using this in a future exercise\ldots

Credit to Donald Arseneau\index{Arseneau, Donald} for a correct answer. 
Thanks to Peter Schmitt\index{Schmitt, Peter} 
for providing the perfect opening for another point I wanted to make:
\begin{quotation}
 The \emph{TeXbook} states explicitly: \\
       \cs{-} is equivalent to \verb?\discretionary{-}{}{}? \\
 and both are internal. 

  I do not see where to the question aims:
\begin{itemize}
\item control symbol : control sequence
\item no paramaters  : three parameters
\item two characters : 21 characters to type
\item ???
\end{itemize}
\end{quotation}

Schmitt is quoting from the last page of Chapter 25;  the point is,
that in newer versions of the \emph{TeXbook}  that sentence has been revised.
I'm not sure what the latest printing says, since I don't have a copy,
but I think it simply refers the reader to Appendix H, where the
significance of \cmd{\hyphenchar} is explained. \cmd{\hyphenchar} is a feature that
was added late in the development of TeX82 (\pfile{TeX82.bug} reveals that is
was not added until May 25, 1983). Even if the source files for the
\emph{TeXbook} were immediately updated by Knuth at that time, the changes did
not appear in the published version being sold to the general public
until some time later  when the first revised edition was published,
which was no earlier than October 1984, the date of the \emph{TeXbook} copy
that I have on hand, and probably later.

The statement of purpose in `Around the bend' \#1 said something
about finding the `best solution', but conspicuously failed to define
what `best' should mean in this context. It was my intention to address
this question in future exercises; for now, let me just say that I
don't intend to arbitrarily rule out of consideration answers such as
Schmitt's `two characters : 21 characters to type', since depending on
how you look at it, it could be argued that this is much more
significant than dumb old \cmd{\hyphenchar} minutiae. I promised that these
exercises would be challenging; that means, among other things, that
they won't always be well-defined, well-bounded, or well-behaved, and
part of the job of finding the `best solution' will be to decide what
parts of the problem need to be specified further, and to examine the
ramifications of alternatives.

%%\endinput


\chapter{What is `best'?}

\section{Exercise (essay)}

%%\input{ex004}
% ex004.tex

\begin{comment}
[Exercises 4,5,6,7 were originally posted together on 4 Nov 91]
Date: Mon 4 Nov 91 16:42:44-EST
From: Michael Downes <MJD@MATH.AMS.COM>
Subject: Around the bend #2
To: info-tex@shsu.edu
\end{comment}

\ed{\oposted{1991/11/04}. \arch{exercise.004}.}

The statement of purpose in `Around the bend' \#1 said something about
finding the `best solution', but failed to define what `best' should
mean when comparing pieces of TeX code. I'll start by throwing out
a few ideas.

\begin{description}
\item[Simplicity] A good solution gets hold of the essential idea of the
problem and attacks it directly, rather than beating around the bush
and resorting to separate clauses to handle troublesome subcases.

\item[Economy] If two solutions compare equal in other respects, then the
better solution is the one that uses less of TeX's resources (main
memory, hash table, string pool, and so forth). Therefore I
(immodestly) say that my solution to Exercise 1 was ever so slightly
better than the other two given, because it avoided introducing any
auxiliary macros that were not included in the original statement of
the problem.

\item[Robustness] If a solution only works under limited friendly
circumstances, and otherwise blows up with an error message, that's not
good. My solution to Exercise 2 was flawed in this respect, since D.A.
found a test case that caused it to go wrong.
\end{description}

%%***********************************************************************
*** Exercise 4 (essay):

What should `best' mean when comparing solutions to an `Around the
bend' exercise? What qualities of a good solution are most important?
Why? How can they be objectively measured? (Or can they?) On the
negative side, what qualities indicate an inferior solution?

%%***********************************************************************

\begin{comment}

[Exercise 5 moved to exercise.005]
[Exercise 6 moved to exercise.006]
[Exercise 7 moved to exercise.007]

Send answers to:

Michael Downes      mjd@math.ams.com (Internet)

A summary will be posted Tuesday, December 4, 1991. However, because of
the difficulty of E7, I will probably procrastinate on posting the
solutions for that exercise until the first or second week of December.
\end{comment}

Table of special characters, to verify accurate transmission:

\begin{lcode}
ASCII 33:  ! exclamation point       ASCII 60:  < left elbow
ASCII 34:  " double quote            ASCII 61:  = equals sign
ASCII 35:  # number/pound sign       ASCII 62:  > right elbow
ASCII 36:  $ dollar sign             ASCII 63:  ? question mark
ASCII 37:  % percent sign            ASCII 64:  @ at sign
ASCII 38:  & ampersand               ASCII 91:  [ left square bracket
ASCII 39:  ' right quote/apostrophe  ASCII 92:  \ backslash
ASCII 40:  ( left parenthesis        ASCII 93:  ] right square bracket
ASCII 41:  ) right parenthesis       ASCII 94:  ^ circumflex/hat/caret
ASCII 42:  * star/asterisk           ASCII 95:  _ underscore
ASCII 45:  - hyphen                  ASCII 96:  ` left quote
ASCII 47:  / slash                   ASCII 123:  { left curly brace
ASCII 58:  : colon                   ASCII 124:  | vert bar
ASCII 59:  ; semicolon               ASCII 125:  } right curly brace
                                     ASCII 126:  ~ tilde
\end{lcode}
%$

%%\endinput


\section{Answers}

%%\input{ans004}
% ans004.tex

\ed{\oposted{1991/12/10}. \arch{answer.004}.}

\begin{comment}
[Solutions for exercises 4,5 were originally posted together on 5 Dec 91]
Date: Thu 5 Dec 91 10:26:58-EST
From: Michael Downes <MJD@MATH.AMS.COM>
Subject: `Around the bend' #2 solutions (4,5)
To: info-tex@shsu.edu

Answers to exercises 4 and 5 of `Around the bend' #2. Discussion of E6
will follow in a separate post because it is rather lengthy. Discussion
of E7 will follow in another couple of weeks (I'm going to be on
vacation next week.)

"***********************************************************************
"*** Exercise 4 (essay):
"
"What should `best' mean when comparing solutions to an `Around the
"bend' exercise? What qualities of a good solution are most important?
"Why? How can they be objectively measured? (Or can they?) On the
"negative side, what qualities indicate an inferior solution?
\end{comment}

Peter Schmitt\index{Schmitt, Peter} writes:
\begin{quotation}
 What is to be rated as `best' clearly depends on the function used to
 measure quality. And therefore the question makes sense only with
 respect to some particular rating function. Seemingly nothing is gained
 by this statement: Instead of discussing what qualities are required
 for a good solution one has to discuss how the rating system should be
 defined. But nevertheless this shifted point of view has an important
 an important advantage. It makes clear that there is no unique answer:
 Quality is not an  absolute notion but a notion relative to some
 (agreed) measure. This measure is not independent of the context ---
 under different conditions different rating functions may be used.

 One further important point must not be forgotten: If matters of
 personal taste are to be excluded than the measuring function has to be
 precisely defined --- demanding simplicity, without giving this notion
 a precise (formal) meaning, is not sufficient.

 Therefore I would like to split the original question into two seperate
 questions:

 (a) What (formal and informal) rating functions are likely to be
     useful, and under what circumstances?

 (b) With respect to some formal rating function, is there always a best
     solution?

 Some answers to the first questions are the following (no completeness
 claimed or even intended):

 (1) the first solution:

     If some special effect is needed for a single application then the
     best solution is the first solution (the solution that can be
     realized with the least effort). This is, however, a purely
     individual criterion that cannot be formalized.

 (2) the most economic (in some sense) solution:

     Economic considerations are important if a code is used frequently,
     Depending on the nature of the applications running time, memory
     usage, and others, may be relevant. But the time spent for finding
     a good solution still cannot be neglected in a real world
     situation. Of course, for theoretical investigations the time spent
     for research does not matter.

 (3) the more robust solution:

     If some set of macros is used by a large number of people who not
     always know how to use them correctly (or even do not care to know)
     then it is certainly an advantage if they are robust, i.e. work in
     as many cases (even strange ones) as possible. But again, one has
     to decide what price (in terms of resources) is acceptable for this
     robustness. (In many cases the item (4) below will be more
     important.)

 (4) ease-of-use:

     If a set of macros is used frequently (by one or more persons) then
     ease-of-use is certainly a mark of quality: easy to remember
     syntax, short commands, natural and good readable embedding into
     the surrounding text, and similar criteria, decide about this.

 (5) simplicity:

     Simple solutions certainly have a strong appeal --- but what is a
     simple solution? Again this is hard to formalize, since simplicity
     basically is an aesthetic value, closely related to the concepts of
     elegance and beauty. (This is similar to the situation in
     mathematics.) But be careful: Simple is not equivalent to short!

 (6) the shortest solution:

     This may seem to be an easy rating function, but is it? Should
     length be measured by the number of characters (probably not!), or
     by the number of tokens, or by the number of control sequences? Or
     by something else?

 Most of the measures mentioned are difficult to formalize, or cannot be
 formalized at all. Only the resources used (in (2)) and the length of a
 code (in (6)) can be precisely defined. Therefore, with respect to one
 of these cases two solutions of the same problem can be compared.
 Furthermore, in many cases it will be possible to proof that an optimal
 solution exists.  (For instance, since the length of a code (in any
 interpretation) is a positive integer, there must exist one or more
 solutions with minimal length, provided there is at least one
 solution.) But unfortunately this does not imply that one is able to
 construct an optimal solution, or to decide whether a given piece of
 code is an optimal solution (or at least near to one). And in some
 cases it may happen that no optimal solution exists, e.g. if to every
 solution there is better --- but longer! --- one.

 What is the conclusion of all this? That there may be a best solution
 relative to some side conditions.  But that there is no globally best
 solution.  This statement is, of course, not very satisfying.  One
 would rather prefer to have at least some notion (even a tentative one)
 of a best solution than none at all. I propose therefore the following
 informal definition (often subject to personal taste): If some code is
 optimal or near-optimal in more than one category then it is probably
 as near to a globally optimal solution as this is possible.
\end{quotation}


My comments:

I propose the following list, based on (1) [my interpretation of]
Knuth's ideas about good macro writing as demonstrated in the \emph{TeXbook}
and plain.tex, (2) various articles in TUGboat, (3) Schmitt's comments,
(4) discussions I've had in the past with other macro writers, and so
forth.

The characteristics of a good solution to an `Around the bend' exercise
are (in order of decreasing importance):

\begin{enumerate}
\item Robustness
\item Brevity (= minimal usage of TeX's main memory)3
\item Simplicity
\item  Ease of use
\item  Suitable commentary
\item  Speed
\item  Minimal hash table load
\item  Minimal save stack load
\item  Minimal load in other categories of TeX's memory
\item  Comprehensive test suite (when applicable)
\end{enumerate}
Schmitt's\index{Schmitt, Peter} point about 'first solution' is well taken 
but does not apply
to `Around the bend' exercises, because of the stated goal of finding a
'best' solution, with the presumption that normally more than one
solution will be found.

Measurement of these qualities is not too difficult, I think,
except for 3 and 5. Here's how I see the measurements:

\begin{description}
\item[1. Robustness] A solution is robust if no one who reads it offers a
counterexample that causes it to fail.  If two solutions both fail, the
one with more counterexamples is less robust; if two solutions have
different counterexamples, the solution whose counterexample is more
likely to occur in normal use is the less robust solution.

\item[2. Brevity] Of two different solutions, the one that is
briefer/shorter/more compact is the one that uses less of TeX's main
memory as measured by \cmd{\tracingstats}.

\item[3. Simplicity] Of two different solutions, the shorter one (in the
sense of the previous item) is usually the simpler one, but not always.
A solution that condenses all the necessary operations into a dense,
incomprehensible Gordian knot is less simple than a longer solution
that lays out the operations in a series of easily comprehended steps.
A solution that relies on arcane dirty tricks is less simple than a
solution that uses better-known techniques in a straightforward
approach.

\item[4. Ease of use] I believe this will not be extremely hard to measure in
the context of the particular application; it can't sensibly be
discussed out of context.

\item[5. Suitable commentary] The commentary surrounding a solution should
explicitly mention any necessary assumptions. If the code is complex,
the commentary should give an outline or overview of the intended
algorithm. It should explain the operation of any macro if its
operation is not evident from the code. If an unusual construction is
used where a different construction would normally be expected, the
commentary should give the reason.

\item[6. Speed] Of two solutions, the speedier one is the one that runs
faster on common computer systems. If one solution runs faster and
slower than another, depending on the system \ldots well, let's not cross
that bridge unless it turns out to be real.

\item[7,8,9. Minimal hash table load, save stack load, etc.] These can be
measured by \\
\cmd{\tracingstats}.

\item[10. Comprehensive test suite] If two solutions are equal in other
respects, the one whose accompanying test suite covers more distinct
cases than the other's is better by that much.
\end{description}

It may be argued that I have not sufficiently answered the question of
subjectivity. For example, who's to decide what's an 'arcane dirty
trick' and what's not? What does 'suitable' mean in number 5? The
answer is that I will say that something is an 'arcane dirty trick' if
I think so, and anyone else can do the same. In most cases I believe
that there will be general agreement on such a question; if not, and an
ensuing discussion fails to reach a clear settlement, then each of the
solutions in question will be decreed 'subjectively just as good as the
others'.

Other qualities of a good solution can be expressed in terms of the
ones listed above. For example, self-sufficiency may be considered an
aspect of robustness---if a solution is not entirely self-sufficient,
it can easily be shown to be not robust by giving a counterexample that
exploits the assumption that makes the solution non-self-sufficient.
Elegance?  If a solution is simple and easy to use, then I say it is
elegant. A solution doesn't necessarily have to be robust in order to
be elegant, nor even short (although of two solutions that are
otherwise equal, the shorter one is undoubtedly more elegant).

\begin{comment}

[Solution for exercise 5 moved to answer.005]

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Table of special characters (ASCII):

33: ! exclamation point;      59: ; semicolon;
34: " double quote;           60: < left elbow;
35: # number/pound sign;      61: = equals sign;
36: $ dollar sign;            62: > right elbow;
37: % percent sign;           63: ? question mark;
38: & ampersand;              64: @ at sign;
39: ' right quote/apostrophe; 91: [ left square bracket;
40: ( left parenthesis;       92: \ backslash;
41: ) right parenthesis;      93: ] right square bracket;
42: * star/asterisk;          94: ^ circumflex/hat/caret;
43: + plus sign;              95: _ underscore;
44: , comma;                  96: ` left quote;
45: - hyphen;                 123: { left curly brace;
46: . period/dot/point;       124: | vert bar;
47: / slash;                  125: } right curly brace;
58: : colon;                  126: ~ tilde
%$
Michael Downes                              mjd@math.ams.com (Internet)

\end{comment}

%%\endinput


\chapter{\cs{string} tokens}

\section{Exercise (fast)}

%%\input{ex005}
% ex005.tex

\ed{\oposted{1991/11/04}. \arch{exercise.005}.}

\begin{comment}

[Posted to info-tex on 4 Nov 91; see exercise.004]
***********************************************************************
*** Exercise 5 (fast):
\end{comment}

Assuming a normal value for \cmd{\escapechar}
\begin{lcode}
     \string\a
\end{lcode}
produces two character tokens. What is the category code of the second?
Write an experiment (as short as possible) to demonstrate the
correctness of your answer.

%%%**********************************************************************

%%\endinput


\section{Answers}

%%\input{ans005}
% ans005.tex

\ed{\oposted{1991/12/05}. \arch{answer.005}.}

\begin{comment}

[Posted to info-tex on 5 Dec 91; see answer.004]
"***********************************************************************
"*** Exercise 5 (fast):
"
"Assuming a normal value for \escapechar,
"
"     \string\a
"
"produces two character tokens. What is the category code of the second?
"Write an experiment (as short as possible) to demonstrate the
"correctness of your answer.
\end{comment}

The category of the 'a' token is 12. All tokens produced by \cmd{\string}
have category 12, except for space tokens, which have category 10.

\begin{solution}{Solution 1 (mine)}
\begin{lcode}
\def\answercheck#1#2{\message{#2: \ifcat0#2\else NOT \fi Category 12}}
\expandafter\answercheck\string\a
\answercheck bb
\end{lcode}
This produces on screen the following message:
\begin{lcode}
a: Category 12 b: NOT Category 12
\end{lcode}
\end{solution}
%%>>EndSolution

%%>>Solution 2 [Peter Schmitt]:
\begin{solution}{Solution 2 (Peter Schmitt)}\index{Schmitt, Peter}
\begin{lcode}
\def\test#1#2#3{%
  \message{\ifcat#2#3 #2 and #3 have the same category code
                \else #2 and #3 have not the same category code
           \fi}}

\def\Test#1#2#3{%
  \ifcat#2#3 \message{#2 and #3 have the same category code}
       \else \message{#2 and #3 have not the same category code}
  \fi}

\catcode`\A12
\test 1aA
\Test 1aA
\expandafter\test\string\a A
\expandafter\Test\string\a A
\end{lcode}

Comment: \\
I have given two essentially equivalent Tests --- \cmd{\test} and \cmd{\Test}.

(i) \cmd{\test} is slightly more simple because it contains only one \cmd{\message}
command, but I think that \cmd{\Test} is more adequate because it avoids to
perform the test inside the \cmd{\message} --- there might be some side
effect one is not aware off.

(ii) Both tests are not as short as possible --- the \piif{true} and \piif{false}
cases could be much shorter, e.g. a T (for true) and a F (for false)
would suffice --- the result could be checked in the dvi-file. (I
regard this difference as inessential.)

Furthermore, setting the catcode of the model character to 12 could
easily be omitted (use some character that is known to be an `other
character'), but I think it should be included: It makes the test
independent of any assumption on the format running. This makes the
solution more closed and selfsufficient, and therefore also simpler and
more elegant (if I may say so).
\end{solution}
%%>>EndSolution
%%\endinput


\chapter{Counting arguments}

\section{Exercise (hard)}

%%\input{ex006}
\begin{comment}
[Posted to info-tex on 4 Nov 91; see exercise.004]
**********************************************************************
*** Exercise 6 (hard):
\end{comment}

\ed{\oposted{1991/11/04}. \arch{exercise.006}.}

Define a macro \cmd{\args} that can be used to fill in the proper number
in the following sentence no matter how \cmd{\foo} is defined (except
you may assume it is not \cmd{\outer}).

     The macro \verb?\tt\string\foo? has \verb?\args\foo? arguments.

Is it possible to solve this if \cmd{\foo} is \cmd{\outer} also? Is it possible
to make \cmd{\args} fully expandable, so that it could be used in a
message:
\begin{lcode}
     \message{The macro \noexpand\foo has \args\foo\space arguments.}
\end{lcode}

%%**********************************************************************
%%\endinput

\section{Answers}

%%\input{ans006}
% ans006.tex
\begin{comment}
Date: Mon 23 Dec 91 11:46:33-EST
From: Michael Downes <MJD@MATH.AMS.COM>
Subject: Answers to 'Around the bend' #2 Exercise 6
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>

"*** Exercise 6 (hard):
"
"Define a macro \args that can be used to fill in the proper number
"in the following sentence no matter how \foo is defined (except
"you may assume it is not \outer).
"
"     The macro {\tt\string\foo} has {\args\foo} arguments.
"
"Is it possible to solve this if \foo is \outer also? Is it possible
"to make \args fully expandable, so that it could be used in a
"message:
"
"     \message{The macro \noexpand\foo has \args\foo\space arguments.}
\end{comment}

\ed{\oposted{1991/12/23}. \arch{answer.006}.}

This was a tough one. All who sent in answers to this exercise
(counting myself) used the approach of applying \cmd{\meaning} to \cmd{\foo} and
analyzing the resulting string. There are some drawbacks to this.

(1) In a \cmd{\meaning} string, all characters (other than spaces) have
catcode 12. This means that all occurrences in a \cmd{\meaning} string of
the character \# are indistinguishable, regardless of their true
significance in the parameter text or replacement text of the macro
in question. Consequently, an occurrence of a \# character, not
category 6, followed by a number, in the parameter text of \cmd{\foo} can
potentially make \cmd{\args} report an incorrect number of arguments. For
example, in the following definitions \cmd{\foo} has no arguments, only
delimiter text, in all three cases, but the \cmd{\meaning} string would
appear to show that \cmd{\foo} has one argument:
\begin{lcode}
  \def\foo\#1{}
  \expandafter\def\expandafter\foo\string #1{}
  \catcode`\#=12 \def\foo#1{}
\end{lcode}

(2) The following two examples produce identical \cmd{\meaning} strings:
\begin{lcode}
  \def\foo&1{} % no arguments
  \catcode`\&=6 \def\foo&1{} % one argument
\end{lcode}

(The string is \verb?"macro:&1->"?.) I.e., characters other than \# can
be used to create parameter markers in a macro definition, and
such a parameter marker cannot be distinguished in a \cmd{\meaning}
string from a normal use of the character in question.

(3) There is no completely general way to isolate the parameter text
of an arbitrary macro from the replacement text. The best you can do
is remove the tail of the \cmd{\meaning} string---everything after the last
occurrence of \verb?->? in the string---and say 'This is not part of the
parameter text'. Likewise, anything preceding the first occurrence of
\verb?->? is certainly part of the parameter text. If there are two or more
occurrences of \verb?->? in the string, however, you cannot say for sure
whether anything between the first and last occurrences is parameter
text or replacement text. This raises a slight additional possibility
that pseudo 'parameter markers' in the replacement text could cause
\cmd{\args} to give an incorrrect result. For example:
\begin{lcode}
  \edef\foo #1{\string#2->}
\end{lcode}
defining \cmd{\foo} with one argument, produces a \cmd{\meaning} string of
\begin{lcode}
  macro:#1->#2->
\end{lcode}
which is exactly the same as the \cmd{\meaning} string for
\begin{lcode}
  \def\foo#1->#2{}
\end{lcode}
where \cmd{\foo} has two arguments.


Speaking practically, however, rather than theoretically, using
\cmd{\meaning} to analyze the number of arguments of an arbitrary macro
works fine. Donald Arseneau's solution, below, is admirably
brief and demonstrates an easy way of handling an outer argument
that I had never seen before.

\begin{solution}{Solution 1 (Donald Arseneau)}\index{Arseneau, Donald}

Here is my solution for counting arguments. It is totally expandable,
and relies on the fact that the parameter numbers must be in
increasing order, that they are only single digits, and that there is
no parameter zero. Also important is that \cmd{\meaning} of a macro defined
by \verb?\def\x#{...}? reports a syntax of \verb?{? rather than \#.
\begin{lcode}
{\catcode`\*=6 \catcode`\#=12 % use * for macro parameters while # is "other"
%
\gdef\args{\expandafter\Args\noexpand}% get rid of \outerness
%
\long\gdef\Args*1{\expandafter\countargs \meaning*1:->{}\end}%
%  ... \meaning will display the parameter syntax (as "other" characters).
%
\gdef\countargs*1:*2->*3\end{\twoargs#0*2#0}% get just the parameter syntax
% ... in format #0junk#1junk...#njunk#0.  \twoargs processes the list to
% ... give "n", the last number before #0.
\end{lcode}

 Here's what tests the parameter numbers, two at a time. (Thus, the two
 \verb?#0?'s in \cmd{\countargs}, so there are always at least two 
\verb?#n?'s detected.)
 When the second number of a comparison isn't zero, \cmd{\twoargs} re-executes
 itself to test the next pair; when the second \verb?n? is 0, the first 
\verb?n? is the
 highest parameter number, so it is output.
\begin{lcode}
\gdef\twoargs*1#*2*3#*4{\ifnum0=*4 *2\else % note the space to end the number
  \expandafter\twoargs\expandafter#\expandafter*4\fi}
}
\end{lcode}

Here is my test suite.  The character ``:'' works in a funny way: it
confuses how \cmd{\countargs} reads its parameter list, and another colon
gets into the supposed syntax.  But it works because there are no
parameters. The primitive \cmd{\halign} is reported to have no parameters
because it is not a macro.  This could be confusing to someone. The
same confusion could arise with \cmd{\args} itself because it doesn't read
the parameter right away.
\begin{lcode}
\def\test#1#{nothing}
\def\Test[#1]#2:{\##1,#2##}
\def\#{haha}

\show\test \show\Test
\end{lcode}

(I condensed this test suite---MJD)
\begin{lcode}
\long\def\msg#1{\message{The object \string#1 has \args#1 arguments.}}

\msg\mathpalette \msg\mathhexbox \msg\par \msg\halign \msg\args
\msg\relax \msg # \msg\# \msg\test \msg\Test \msg : \msg\: \msg\csname
\msg t \msg ~ \msg $ \msg ^
\end{lcode}

(Outer macros---MJD)
\begin{lcode}
\message{The object \string\bye\space has \args\bye\space arguments.}
\message{The object \string\newhelp\space has \args\newhelp\space
  arguments.}

\bye  % --  Donald Arseneau
\end{lcode}
\end{solution}
%%>>EndSolution

Although the problem statement only mentioned `macros' Arseneau
earned some thoroughness points by including primitives \cmd{\halign},
\cmd{\relax}, and \cmd{\csname}, as well as characters \verb?# : t $ ^? 
in his tests.
This is of some interest because of the difference in \cmd{\meaning}
strings between macros and non-macros.

In my solution for this exercise, I amused myself by trying to pack
everything into as few control sequences as possible. Although I got
it down to two, that's really only one less than Arseneau's four,
because one control sequence in his solution is expended to
handle outer macros, something my solution didn't attempt to do.

%>>Solution 2 (mine)
\begin{solution}{Solution 2 (mine)}
\begin{lcode}
%   Use & instead of # temporarily.
\catcode`\&=6 \catcode`\#=12

\long\def\args &1{\expandafter\countargs\meaning &1#\args->\countargs 0}
\end{lcode}

   Analysis is restricted to the parameter text by chopping off everything
   after \verb?->? in the meaning string (this will leave possibly only part
   of the parameter text).

   Then we look in the parameter text for \# followed by a number
   (checking to make sure that the thing after \# is a number handles a
   few extra possibilities, such as \verb?\#? followed by non-number in the
   parameter text). If we find \# plus a number, we pass the number
   onward to the next invocation of \cmd{\countargs}, where it will end up as
   the returned value (argument \#5) if the next \cmd{\countargs} determines
   that the remaining parameter text contains no more parameter markers.
\begin{lcode}
\def\countargs &1#&2&3->&4\countargs &5{%
  \ifx\args&2&5%
  \else
    \ifodd0&21 % Then &2 is a number, carry forward.
      \countargs&3#\args->\countargs&2%
    \else % &2 not a number---ignore, carry forward last number instead
      \countargs&3#\args->\countargs&5%
    \fi
  \fi}

\catcode`\#=6

\def\test{\message{The macro \noexpand\foo has \args\foo\space
  arguments (\meaning\foo).}}

%\tracingmacros=2 \tracingcommands=2
% Success:
\def\foo{No args}\test
\def\foo#1{One arg}\test
\def\foo#1#2{Two args}\test
\def\foo./{No args, delimited}\test
\def\foo#1#2#3#4#5#6#7#8#9{Nine args}\test
\def\foo//#1#2#3#4#5#6#7#8#9//{Nine args, delimited}\test
\def\foo#{Weird}\test
\def\foo#1#{Weird, one arg}\test
\def\foo#1#2#3#4#5#6#7#8#9#{Weird, nine args}\test
\def\foo#1 {One arg, space delimited}\test
\def\foo#1 #2 #3 #4 #5 #6 #7 #8 #9 {Nine args, space delimited}\test
\def\foo/{\def\foo}
\foo/ #1{Interesting}\test

\edef\foo#1#2{\string #3\string #4}\test
\edef\foo{\string #}\test
\expandafter\edef\expandafter\foo
  \csname 0\string #\string #\endcsname#1#2{#1#2}\test

% Failure:
\def\foo->#1->#2->#3->#4->#5->#6->#7->#8->#9->{Nine args, devious
  delimiter}\test
\expandafter\edef\expandafter\foo
  \csname 0\string #1\string #2\endcsname{...}\test
\let\foo=\bye \test % \outer bomb
\end{lcode}
\end{solution}
%%>>EndSolution

When I originally posed this problem, I had seen far enough ahead to
suspect that the drawbacks of \cmd{\meaning} mentioned above would be
impossible to overcome. But \cmd{\meaning} is the only way to analyze a
macro that has a nonsimple parameter text---that is, one containing
delimited arguments. Another possibility I had in mind was restricting
the analysis to macros with simple parameter texts---empty or having
only nondelimited arguments---to see what might be done without
\cmd{\meaning}. The best that I could manage in my experiments along these
lines was a definition of \cmd{\args} with an unacceptably cumbersome call
syntax. But it does have the virtue of correctly identifying any
number of nondelimited arguments, no matter whether \cmd{\foo} was
originally defined using \# (category 6) or some other category 6
character.

%%>>Solution 3 (mine)
\begin{solution}{Solution 3 (mine)}
\begin{lcode}
% This solution is not fully expandable, hence cannot be used
% inside a \message.

\def\args{\expandafter\argscontinue}

\def\argscontinue{\begingroup
\end{lcode}
    Make all digits have category 2 (= end of group) so that
    they will serve to end the token register assignment
    \verb?\global\toks1 ...?
\begin{lcode}
  \catcode`\0=2 \catcode`\1=2 \catcode`\2=2 \catcode`\3=2 \catcode`\4=2
  \catcode`\5=2 \catcode`\6=2 \catcode`\7=2 \catcode`\8=2
\end{lcode}

    We use \cmd{\afterassignment} to put an \cmd{\endgroup} after the
    token register assignment, so that numbers will revert to
    their ordinary catcodes. And we use \cmd{\aftergroup} to put
    a \cmd{\finishup} token after the \cmd{\endgroup}. Thus \cmd{\finishup} can
    look ahead to see what numbers are remaining; this information
    reveals how many arguments were used up by the \cmd{\foo} macro call.
\begin{lcode}
  \aftergroup\finishup \afterassignment\endgroup
  \global\toks1\bgroup}
\end{lcode}

    \cmd{\finishup} takes the first digit following it and returns it
    as the value of \cmd{\args}; any following numbers are discarded
    (note that \#2 is delimited by a space).
\begin{lcode}
\def\finishup#1#2 {%\showthe\toks1
  #1}

%\tracingmacros=2 \tracingcommands=2 \tracingonline=1
\def\foo{}
The macro {\tt\string\foo} has \args\foo 00123456789 \ arguments.

\def\foo#1{}
The macro {\tt\string\foo} has \args\foo 00123456789 \ arguments.

\edef\foo#1{\string #2\string #3\string #4->\string #4\string #3#1}
The macro {\tt\string\foo} has \args\foo 00123456789 \ arguments.

\def\foo#1#2#3{a#1b#2c#3}
The macro {\tt\string\foo} has \args\foo 00123456789 \ arguments.

\def\foo#1#2#3#4#5#6#7#8#9{#1#2#3#5#8bb#9}
The macro {\tt\string\foo} has \args\foo 00123456789 \ arguments.
\end{lcode}
\end{solution}
%%>>EndSolution

The fourth solution for Exercise 6 is by Peter Schmitt; it gets the
robustness prize for carrying out a diligent analysis of \cmd{\meaning}
strings that enables it to correctly handle a greater variety of
exotic cases than the other solutions. Schmitt's original method of
handling outer macros was effective, but more complicated than
Arseneau's method, incorporated here as noted. Even though my
approach was rather different from Schmitt's, some of the comments in
Schmitt's solution inspired me in turn to improve my solution [2]
from its previous much inferior state.

%%>>Solution 4 (Peter Schmitt)
\begin{solution}{Solution 4 (Peter Schmitt)}\index{Schmitt, Peter}
\begin{lcode}
% \args <token> expands to:  -   if <token> is not a macro
%                           0..9 according to the number of parameters
%                                if the <token> is a macro
% \args is fully expandable and accepts outer macros as well.
%    It assumes, however, that the tested macro has been defined using the
%    standard parameter symbol #,
%    and that the current value of \escapechar is the standard backslash \.
\end{lcode}

 The definition of the macros uses the expansion of
\cmd{\meaning}\verb?\cs?:
 It is of the form:
\begin{lcode}
     [..] macro: [parameter text] -> [replacement text]
\end{lcode}
 and consists of `other characters'.

 The macro \cmd{\args} checks:
\begin{enumerate}
\item if the expansion contains `macro': \\
      --- if not, then \verb?\cs? is not a macro and \cmd{\args} yields `-'
\item if the expansion contains parameters \#1 etc. \\
      --- if \verb?#n? is the first that is not present
             then \verb?\cs? takes (n-1) arguments
             and \cmd{\args} yields `n-1'
\end{enumerate}
 The following special characters are chosen to make the definitions as
 readable as possible. Any characters having catcodes different from 12
 will serve the same purpose:
\begin{lcode}
\catcode`\:3 \catcode`\/3  % : and / are used as parameter delimiters
\catcode`\^3               % ^ is used to detect empty arguments
\catcode`\?11              % ? is used to make the control sequences private
\end{lcode}
 Since the occurrences of \# in the expansion of \cmd{\meaning}\verb?\cs? has to be
 detected, it has to be used as an `other character'.
 To avoid confusion it has been replaced not only where necessary but
 throughout all the definitions:
\begin{lcode}
\catcode`\#12  \catcode`\*6   % * is parameter character
\end{lcode}
\begin{itemize}
\item \verb|\?macro| is defined to be `macro' consisting of `other characters'
      using the expansion of \verb?\meaning\TeX?.
\item  \verb?\?DEF? inserts these five characters into some definitions 
       where they are as parameter delimiters:
\begin{lcode}
    \DEF\cs { <parameter text> } { <replacement text> }
\end{lcode}
         where the texts may contain *1 and **1 .. **9
 yields 
\begin{lcode}
    \def\cs <parameter text>{<replacement text>}
\end{lcode}
 where  *1 is replaced by `macro' and **1 yields *1 etc.
\end{itemize}

\begin{lcode}
\def\?macro *1:*2:{*1} \edef\?macro{\expandafter\?macro\meaning\TeX:}
\def\?DEF *1*2{\def*1**1:{\long\def*1*2}\expandafter*1\?macro:}
\end{lcode}

\begin{itemize}
\item \cmd{\args} passes the \meta{token} unexpanded to \verb|args?|
\item (taken from the solution by Donald Arseneau)
 \verb|\args?| takes one argument, expands its \cmd{\meaning} to TEXT
       and passes it to \verb|\macro?| after appending \verb|macro^|:
\item \verb|\macro?| checks the first token after the first occurrence of 
`macro':
       if this is \verb?^(3)?, then `macro' was not present in TEXT (output: -)
       otherwise TEXT is further investigated.
\end{itemize}

\begin{lcode}
\def\args{\expandafter\args?\noexpand}
\?DEF \args? {**1{\expandafter\macro?\meaning **1*1^:}}
   \?DEF\macro? {**1*1**2:{\ifx^**2-\else\expandafter\purge? **2:\fi}}
\end{lcode}

 The parameters taken by a control sequence all appear (once and in 
 numerical order) in the parameter text --- and no other occurrence 
 of a pair \verb?#n? is allowed in it. Moreover, only the same pairs \verb?#n? 
may 
 occur in the replacement text. It is, however, not possible to simply 
 look for occurrences of these pairs since there are tokens that may ---
 if followed by some number --- be (wrongly) interpreted as parameters:
\begin{itemize}
\item the token \verb?##? in the replacement text, and
\item (as pointed out by Michael Downes)
   -the control symbol \verb?\#? both in the parameter text and the 
 replacement text.
\end{itemize}
 Since \verb?\\#n? has to be distinguished from \verb?\#n? the control 
symbol \verb?\\? is also important.

 Therefore \verb|\purge?| is used to remove all occurrences of these tokens.
 After that the search-macro \verb|\head?| is invoked, appending
   the sequence \verb?#n^(n-1)? for every possible parameter \verb?#n?.

 Since \verb|\purge?| has to identify the character \verb?\(12)? it is 
necessary   to change the escapecharacter:

\begin{lcode}
\catcode`\!0 !catcode`!\=12   % ! is used as escape character
\end{lcode}

 \verb|\purge?| appends \verb?## \#^ and \\^? to the TEXT as a means to 
stop the search
   for these tokens, and : as delimiter:
\begin{enumerate}
\item  \verb|\backslash?| looks for the first occurrence of the character pair 
  \verb?\\? in TEXT (this must be a token \verb?\\?) and replaces it by a 
  space.
  If it is followed by \verb?^(3)? then the search is completed,
  otherwise the process is repeated.
\item \verb|\numbersign?| looks for the first occurrence of the character pair 
  \verb?\#? in the (in the meantime modified) TEXT (since all \verb?\\? have 
  been removed this must correspond to a token \verb?\#?) and replaces it by 
  a space.
  Again the process is stopped when it is followed by \verb?^(3)?.
\item \verb|\parametersign?| truncates TEXT at the first occurrence of the
  character pair. Note that this pair must correspond to a parameter
  token \verb?##? in the replacement text and therefore the rest of TEXT is
  not needed any more.
\end{enumerate}
\begin{lcode}
    !def!purge? *1:{!backslash? *1##\#^\\^:}

% \purge? could be avoided - \macro? could call \backslash? directly

    !def!backslash? *1\\*2*3:{!ifx^*2!expandafter!numbersign?
                               !else !expandafter!backslash?
                                 !fi *1 *2*3:}
    !def!numbersign? *1\#*2*3:{!ifx^*2!expandafter!parametersign?
                                 !else !expandafter!numbersign?
                                   !fi *1 *2*3:}

!catcode`!\0 \catcode`\!=12  % return to the normal use of backslash

    \def\parametersign? *1##*2:{%
               \head? *1^#1^0#2^1#3^2#4^3#5^4#6^5#7^6#8^7#9^8#0^9:}
\end{lcode}

    For each n from 0 to 9 \verb|\head?| extracts the characters contained in
the (appended) TEXT between the first occurrence of \verb?#n? and 
\verb?#(n+1)? and investigates them with \verb|\used?|.

   If \verb?#n? is not present in TEXT, then the first of these characters is
\verb?^(3)?, taken from the appended string: \\
When this happens for the first time \verb|\used?| outputs the second character
(the number of parameters) and calls \verb|\skip?| to hide all the remaining
parts of the appended TEXT, otherwise \verb|\used?| checks the next item.

   Since eleven parameters are necessary to handle the ten cases (0..9) this
duty has to be distributed on two macros: \\
The appearance of the character \verb?/(3)? is used to indicate that the second
   macro \verb|\tail?| has to be invoked by \verb|\used?|.
\begin{lcode}
    \def\head? *1#1*2#2*3#3*4#4*5#5*6:{%
          \used? *2..:*3..:*4..:*5..:/.:%
          \expandafter\tail? *6://}
    \def\tail? *1#6*2#7*3#8*4#9*5#0*6:{\used? *2..:*3..:*4..:*5..:*6:}
    \def\used? *1*2*3:{\ifx^*1*2\expandafter\skip?\else\ifx/*1\else
                       \expandafter\expandafter\expandafter\used?\fi\fi}
    \def\skip? *1//{}

%% Finally, catcodes are turned back to normal:

\catcode`\#6 \catcode`\*12 \catcode`\?12
\catcode`\:12 \catcode`\/12 \catcode`\^12

%%%%%%%%%%%%%%%%%%%%%%

\long\def\test#1{
     The macro {\tt\string#1} has {\args#1} arguments.

     \message{The macro \noexpand#1 has :\args#1:\space arguments.}
}

\def\exc#1\\#2\ #3{\#4\\#1\\\#4\\\\#2two arguments}
\test\exc

\end
\end{lcode}
\end{solution}
%%>>EndSolution

Schmitt's solution assumes the use of mine and Arseneau's test suites
as well, because they had been shared between us before Schmitt sent
in the final version of his solution.

\begin{comment}
Answers for Exercise 7 will follow next week.

Michael Downes                             mjd@math.ams.com (Internet)
\end{comment}
%%\endinput


\chapter{Self replication}

\section{Exercise (hard)}

%%\input{ex007}
\begin{comment}
[Posted to info-tex on 4 Nov 91; see exercise.004]
**********************************************************************
*** Exercise 7 (hard):
\end{comment}

\ed{\oposted{1991/11/04}. \arch{exercise.007}.}

In the September 1991 issue of Dr. Dobb's Journal, in an article
`Little Languages, Big Questions' (pp. 16--25), Ray Vald\'es
described a `little language' as a part of a more complex
application that is
\begin{quote}
     partitioned into two (or more) nested components: a core module
     that provides a primitive set of services for an application area
     (the ``engine''), and a surrounding module that provides
     programmatic access to these services. The surrounding module is
     typically a language interpreter for a simple, easily parsed
     computer language--a ``little language''.
\end{quote}

Since TeX seems to fall into this category, I wonder if any Dr. Dobb's
readers who know TeX tried their hand at the challenge given in a
sidebar (`How Strong Is Your Little Language')?
\begin{quote}
     [An] informal benchmark of a language's computational power is the
     programming exercise that Ken Thompson (coauthor of Unix) used to
     pass the time in college. ... The goal is to write the shortest
     self-reproducing program: ``More precisely stated ... to write a
     source program that, when compiled and executed, will produce as
     output an exact copy of its source.''
\end{quote}

When I tried it it turned out to be a real challenge for me. In the
Unix world, for conventional compiled languages, the problem as
originally stated can assume output on the `standard output' stream;
but TeX already clutters up standard output with some of its built-in
messages. This leaves three alternatives in refining the statement of
the problem to be meaningful for TeX:

1. Write a TeX program that includes the built-in messages in its
source in such a way that it exactly fulfills the the original problem
statement with standard output as the output stream.

2. Pretend the built-in messages don't exist and write a TeX program
that reproduces an exact copy of itself (with no extra garbage)
in the middle of the built-in messages.

3. Write on a different output stream.

Take your pick, any or all of the above, and see what you can come up
with. I have solutions for 2 and 3 but have not gotten around to really
thinking about 1 yet. I believe it will require at least a different
algorithm than the other 2, if it is not impossible.

%%%**********************************************************************
%%\endinput

\section{Answers}

%%\input{ans007}
% ans007.tex
\begin{comment}
[The `forthcoming' TUGboat article cited below appeared as
`Self-replicating macros' by Victor Eijkhout and Ron Sommeling, TUGboat
13 (1992) no 1, p. 84]

Date: Tue 7 Jan 92 16:43:29-EST
From: Michael Downes <MJD@MATH.AMS.COM>
Subject: 'Around the bend' #2, Exercise 7, solutions
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>

"*** Exercise 7 (hard):
"
"In the September 1991 issue of Dr. Dobb's Journal, in an article
"`Little Languages, Big Questions' (pp. 16--25), Ray Vald\'es
"described a `little language' as a part of a more complex
"application that is
"
"     partitioned into two (or more) nested components: a core module
"     that provides a primitive set of services for an application area
"     (the ``engine''), and a surrounding module that provides
"     programmatic access to these services. The surrounding module is
"     typically a language interpreter for a simple, easily parsed
"     computer language--a ``little language''.
"
"Since TeX seems to fall into this category, I wonder if any Dr. Dobb's
"readers who know TeX tried their hand at the challenge given in a
"sidebar (`How Strong Is Your Little Language')?
"
"     [An] informal benchmark of a language's computational power is the
"     programming exercise that Ken Thompson (coauthor of Unix) used to
"     pass the time in college. ... The goal is to write the shortest
"     self-reproducing program: ``More precisely stated ... to write a
"     source program that, when compiled and executed, will produce as
"     output an exact copy of its source.''
"
"When I tried it it turned out to be a real challenge for me. In the
"Unix world, for conventional compiled languages, the problem as
"originally stated can assume output on the `standard output' stream;
"but TeX already clutters up standard output with some of its built-in
"messages. This leaves three alternatives in refining the statement of
"the problem to be meaningful for TeX:
"
"1. Write a TeX program that includes the built-in messages in its
"source in such a way that it exactly fulfills the the original problem
"statement with standard output as the output stream.
"
"2. Pretend the built-in messages don't exist and write a TeX program
"that reproduces an exact copy of itself (with no extra garbage)
"in the middle of the built-in messages.
"
"3. Write on a different output stream.
"
"Take your pick, any or all of the above, and see what you can come up
"with. I have solutions for 2 and 3 but have not gotten around to really
"thinking about 1 yet. I believe it will require at least a different
"algorithm than the other 2, if it is not impossible.
\end{comment}

\ed{\oposted{1992/01/07}. \arch{answer.007}.}


Plenty of good answers for this one.

%%>>Solution 1 (mine)
\begin{solution}{Solution 1 (mine)}

    This solution is type 2 (print the copy in the middle of TeX's
    built-in messages). It assumes \pfile{plain.tex} or similar has been
    loaded to set the catcodes of the left and right curly braces.

    The idea is to assign the text to the token register \cmd{\errhelp}
    (used merely because it is a convenient pre-existing token
    register), and then print out \cmd{\the}\cmd{\errhelp} twice. There is a bit
    of shuffling to ensure that \cmd{\errhelp} will swallow the last half of
    the file and that the last half of the file is equal to the first
    half, which contains all the preparations necessary to prepare
    \cmd{\errhelp} for that swallowing and the subsequent message-sending.

    A space is left after every control word, because this is easier
    than trying to prevent TeX from printing spaces after control
    words when the message is eventually printed on screen.

    The lines are carefully arranged to break at column 79
    (including spaces) since this is the normal value for \verb?max_print_line?,
    a constant compiled into TeX which controls the length of screen
    output lines. It would be easy to make the lines work out nicely
    no matter what the working code required, by varying the length
    of the macro name \cmd{\selfcopy} and using, say, \cmd{\everyhbox} or
    \cmd{\everyjob} instead of \cmd{\errhelp}.

    The total number of tokens in this solution is 54.

\begin{lcode}
{\gdef \selfcopy {\message {{\the \errhelp }}\message {{\the \errhelp }}\end }
\aftergroup \errhelp \afterassignment \selfcopy }
{\gdef \selfcopy {\message {{\the \errhelp }}\message {{\the \errhelp }}\end }
\aftergroup \errhelp \afterassignment \selfcopy }
\end{lcode}
%%>>EndSolution
\end{solution}

%%>>Solution 2 (mine)
\begin{solution}{Solution 2 (mine)} 
This variation is Type 3, writing the copy to a disk file
instead of to the screen. The total number of tokens in this
solution is 126.
\begin{lcode}
\immediate \openout 0=\jobname .cpy
{\gdef ~#112{\errhelp {#112}\immediate \write 0{\the \errhelp
}\immediate \write 0{\the \errhelp }\immediate \closeout 0 \end}}
\newlinechar 13 \catcode `\#=3 \afterassignment ~\catcode 13=12
\immediate \openout 0=\jobname .cpy
{\gdef ~#112{\errhelp {#112}\immediate \write 0{\the \errhelp
}\immediate \write 0{\the \errhelp }\immediate \closeout 0 \end}}
\newlinechar 13 \catcode `\#=3 \afterassignment ~\catcode 13=12
\end{lcode}
%%>>EndSolution
\end{solution}

I learned from Victor Eijkhout that he had submitted a short article
to TUGboat discussing this very problem, well before I asked it here in
'Around the bend'. He kindly sent me a copy of the article, which
contains a good discussion of the underlying ideas, and a couple of
different solutions. To summarize briefly, he gave a Type 2 solution
similar in length to mine, and also a solution that involved
printing out the source file on PAPER! A 'Type 4' solution, in other
words. I'm a little embarrassed that I didn't think of this, given that
the whole idea of TeX is to print things on paper.

%%>>Solution 2 (Victor Eijkhout)
\begin{solution}{Solution 2 (Victor Eijkhout)}\index{Eijkhout, Victor}
Forthcoming in TUGboat. It appeared as: \\
`Self-replicating macros' by Victor Eijkhout and Ron Sommeling, TUGboat
13 (1992) no 1, p. 84.
%%>>EndSolution
\end{solution}

Although I'm giving them all together, as `Solution 3', Peter Schmitt
actually sent in six different variations, including a Type 4 solution.
His first solution, \pfile{log-pl.tex} is Type 2 like my first solution but
comes in at 38 tokens, significantly shorter. His third solution is
comparable to my second solution but once again significantly shorter
(87 tokens).

\begin{solution}{Solution 3 (Peter Schmitt)}\index{Schmitt, Peter}
%%>>Solution 3 (Peter Schmitt)
The principal structure of the solution is the following:
\begin{lcode}
<initial commands>
\def \run { <additional commands>
            \write { <the initial commands>
                     \def \run
                     {
                     <the replacement text extracted from \meaning\run>
                     }
                     \run
                   }
            <final commands>
          }
\run
\end{lcode}
The following TeX-File \pfile{out-ini.tex} when processed by INITeX
produces a file \pfile{out-ini.out} that is identical to \\
 \pfile{out-ini.tex} (case (3) below):

(The file consist of a single line, it is broken up to make comments
possible - each occurrence of the comment sign \% has to be removed
together with the rest of the line to produce identical output.)

\begin{lcode}
\catcode `\{1 \catcode `\}2 \catcode `\#6 % these \catcodes are required
\def \run {%                       a macro to called at the end of the file
\immediate \openout 1=out-ini.out%   % opens output
\def \select ##1:->##2{##2}% an auxiliary macro to extract the replacement text
\immediate \write 1{%                  write the output file
\catcode `\noexpand \{1 \catcode `\noexpand \}2 \catcode `\noexpand \#6 %
%                                      writes the first `line' of the output
\noexpand \def \noexpand \run %        writes \def \run
{\expandafter \select \meaning \run }% writes the replacement text of \run
\noexpand \run }%                      writes the last `line' of the program
\immediate \closeout 1%                close output file
\end }%                                close input
\run %                                 start the macro
\end{lcode}

Comments:
\begin{enumerate}
\item \cmd{\immediate} prevents that a dvi-file is produced.
\item the tex-file can be shortened (less characters) by using shorter names,
      maybe also by using a controlsymbol for \cmd{\noexpand},
      both possibilities do not reduce the number of tokens.
      Maybe some \cmd{\space} tokens can be removed but most of them are necessary
      because they are produced by \cmd{\meaning}.
\begin{itemize}
  \item \cmd{\immediate} may be omitted (produces dvi-file)
  \item at least with my implementation closing the output file is not
        necessary
\end{itemize}
\item The TeX-file can be modified to solve variations of the exercise:
  \begin{itemize}
  \item If the file is to be processed by plain TeX \cmd{\catcodes} need not be set
        (see (1) below).
  \item if the output file is replaced by standard output or the log file
        \cmd{\message} instead of \cmd{\write} can be used (see (1) and (2) below).
        Note that in this case macro names and spaces have to be adjusted
        so that the line breaks produced do not prevent processing
        the file (In the log file line breaks may occur even in control
        sequence names!)

        I have not (not yet?) been able to solve the exercise using more
        pleasant (predetermined) linebreaks.
  \item It is possible to produce a log file that is identical to the
        input file. But since the log file contains the time of processing
        this will be the case only at a specific date and time (see (4) below).
        (The time is output before the input file is read. Therefore it is
        impossible to change this part of output by the input.)
  \item Of course, the above variation can be modified to produce a screen
        output identical to the input file.
   \item It is possible to pass a verbatim copy of the input to TeX and set
        it in \cmd{\tt}
  \end{itemize}
\end{enumerate}

%%%%%%%%%%%%%%%%%%%%%%%
Some of the variations:
%%%%%%%%%%%%%%%%%%%%%%%

(1) plain TeX \verb?-->? section of log file or standard output terminal
\begin{lcode}
%%% log-pl.tex:
\def \run {\def \select ##1:->##2{##2} \message {\noexpand \def \noexpand \run
{\expandafter \select \meaning \run } \noexpand \run } \end } \run

%%% log-pl.log
This is TeX, Version 3.1(c)sb34 (preloaded format=plain3sm 91.4.28)
24 NOV 1991 02:15
** &plain  log-pl
(log-pl.tex
\def \run {\def \select ##1:->##2{##2} \message {\noexpand \def \noexpand \run
{\expandafter \select \meaning \run } \noexpand \run } \end } \run  )
No pages of output.
\end{lcode}

(2) INITeX \verb?-->? section of log file or standard output terminal
\begin{lcode}
%%% log-ini.tex
\catcode `\{=1 \catcode `\} =2 \catcode `\#=6 \def \run {\def \selectit
##1:->##2{##2} \message {\catcode `\noexpand \{=1 \catcode `\noexpand \}
 =2 \catcode `\noexpand \#=6 \noexpand \def \noexpand \run {\expandafter
 \selectit \meaning \run }\noexpand \run }\end }\run

%%% log-ini.log
This is TeX, Version 3.1(c)sb34 (INITEX)
24 NOV 1991 02:16
** log-ini.tex
(log-ini.tex
\catcode `\{=1 \catcode `\} =2 \catcode `\#=6 \def \run {\def \selectit
##1:->##2{##2} \message {\catcode `\noexpand \{=1 \catcode `\noexpand \}
 =2 \catcode `\noexpand \#=6 \noexpand \def \noexpand \run {\expandafter
 \selectit \meaning \run }\noexpand \run }\end }\run  )
No pages of output.
\end{lcode}

(3) INITeX \verb?-->? output file
\begin{lcode}
%%% out-ini.tex (Note: A single line broken at the %'s!)
\catcode `\{1 \catcode `\}2 \catcode `\#6 \def \run {\immediate \openout %
1=out-ini.out\def \select ##1:->##2{##2}\immediate \write 1{\catcode %
`\noexpand \{1 \catcode `\noexpand \}2 \catcode `\noexpand \#6 \noexpand \def %
\noexpand \run {\expandafter \select \meaning \run }\noexpand \run }%
\immediate \closeout 1\end }\run
\end{lcode}

(4) INITeX \verb?-->? log file
\begin{lcode}
%%% flog-ini.tex
This is TeX, Version 3.1(c)sb34 (INITEX)
24 NOV 1991 02:17
** flog-ini.tex
(flog-ini.tex
\catcode `\{=1 \catcode `\} =2 \catcode `\#=6 \def \run {\def \selectit
##1:->##2{##2} \message {\catcode `\noexpand \{=1 \catcode `\noexpand \}
 =2 \catcode `\noexpand \#=6 \noexpand \def \noexpand \run {\expandafter
 \selectit \meaning \run }\noexpand \run }\end }\run  [0] )
Output written on flog-ini.dvi (1 page, 512 bytes).

%%% flog-ini.log
This is TeX, Version 3.1(c)sb34 (INITEX)
24 NOV 1991 02:18
** flog-ini.tex
(flog-ini.tex
\catcode `\{=1 \catcode `\} =2 \catcode `\#=6 \def \run {\def \selectit
##1:->##2{##2} \message {\catcode `\noexpand \{=1 \catcode `\noexpand \}
 =2 \catcode `\noexpand \#=6 \noexpand \def \noexpand \run {\expandafter
 \selectit \meaning \run }\noexpand \run }\end }\run  [0] )
Output written on flog-ini.dvi (1 page, 512 bytes).
\end{lcode}

(5) INI-TeX \verb?-->? log-file (formatted)
\begin{lcode}
%%% fmt-log.tex
This is TeX, Version 3.1(c)sb34 (INITEX)
30 NOV 1991 13:13
** fmt-log
(fmt-log.tex [0
\catcode `\{=1 \catcode `\}=2
\catcode `\#=6
\def \run
{\newlinechar 1 \lccode `\|=1
 \lccode `\[=`\{ \lccode `\]=`\}
 \lowercase {
 \def \format ##1>##2=1##3]##4[##5]##6]{##2=1|##3]|##4[|##5]|##6]|\+}
 \def \+ ]##12]##2]##3]##4]]##5] { ]|##12]|##2]|##3]|##4]]|##5]|}
 }
 \write 0{\catcode `\noexpand \{=1 \catcode `\noexpand \}=2}
 \write 0{\catcode `\noexpand \#=6}
 \write 0{\noexpand \def \noexpand \run }
 \write 0{{\expandafter \format \meaning \run }}
 \write 0{\noexpand \run }
\end }
\run
] )
Output written on fmt-log.dvi (1 page, 512 bytes).
\end{lcode}

(6) INITeX \verb?-->? dvi-file
\begin{lcode}
%%% dvi-ini.tex
\catcode`\% = 13
\catcode`\{ = 1 \catcode `\} = 2
\catcode`\# = 6 \catcode `\| = 13
\catcode`\% = 13
\def \run {
 \lccode `\[=`\{ \lccode `\]=`\} \lccode `\/=`\% \let % = \par %%
 \font\tt=cmtt10 \tt %
 \hsize 15cm \vsize 15cm \parskip 3pt \def |{\par \hskip .5em} %
 \lowercase { %
 \def \fmt ##1>##2//##3/##4/##5/##6/##7/{|##2//|##3/|##4/|##5/|##6/|##7/|\+} %
 \def \+ ##1/##2/##3/##4//##5/##6/##7/{##1/|##2/|##3/|##4//|##5/|##6/|##7/|} %
 } %
 \string \catcode `\string \{ = 1 \string \catcode `\string \} = 2 %
 \string \catcode `\string \# = 6 \string \catcode `\string \| = 13 %
 \string \catcode `\string \% = 13 %%
 \string \def \string \run \lowercase { [} %
 \expandafter \fmt \meaning \run  \lowercase {]} %
 \string \run %
 \end }
\run
\end{lcode}
%%>>EndSolution
\end{solution}

%%\endinput

\chapter{\cs{end} too soon}

\section{Exercise (hard)}

%%\input{ex008}
% ex008.tex
\begin{comment}
Date: 21 Jun 1993 09:49:27 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #8
To: info-tex@shsu.edu
\end{comment}

\ed{\oposted{1993/06/21}. \arch{exercise.008}.}

A few readers of info-tex and comp.text.tex may recall some postings
of mine under the name of `Around the Bend' more than a year ago. This
was intended to be a regular quasi-monthly stream of challenging
questions about TeX macro writing, but after a few appearances it fell
into limbo because of too many other demands on my time.  However I
continue to encounter hard, interesting problems in my work so
herewith wish to announce resumption of the `Around the Bend' postings
on an occasional, slightly less ambitious basis.

For background, here are a couple of excerpts from the first `Around
the Bend' post:
\begin{quote}
 With the encouragement of George Greenwade (the INFO-TeX list owner), I
 would like to propose a regular department for INFO-TeX, called `Around
 the bend'. It will consist of macro-writing challenges on the level of
 the dangerous-bend exercises in the \emph{TeXbook}, with interested parties
 invited to collaborate and/or compete to find the best solution. My
 motivation for doing this is partly selfish: to get more feedback from
 other macro writers about some of the interesting macro-writing
 problems that I run into.

\ldots



 Solutions should be sent to me instead of to INFO-TeX or
 comp.text.tex, on the premise that people usually won't want to read
 others' solutions until they've had a chance to try their own hand. A
 summary of the results would then be posted to the INFO-TeX list after
 two or three weeks; to those who submit solutions before the deadline,
 I could forward without delay solutions submitted by other people, for
 comparison.
\end{quote}

And here's number 8.

%%***********************************************************************
%%*** Exercise 8 (hard):

Under certain conditions, TeX fails to give an error message
for a missing closing brace or \cmd{\endgroup} or \piif{fi}; it only gives an
unobtrusive warning message after the end of the TeX run, which is
easy to overlook:
\begin{lcode}
  (\end occurred inside a group at level 1)
  (\end occurred when \iffalse on line 6 was incomplete)
  (\end occurred when \iftrue on line 3 was incomplete)
\end{lcode}

Is there any way to trap these conditions and give a true error
message?---if, let's say, you are programming for a major macro
package like LaTeX and want to make sure these conditions are brought
to the user's attention.

%%%***********************************************************************

\begin{description}
\item[Remark] Off-hand one would think that trapping these conditions is
impossible, since otherwise Knuth\index{Knuth, Donald}
 would presumably have built the
trapping into TeX; \piif{iffalse} \ldots \cmd{\end} generates an error message, 
it's
only \piif{iffalse} \ldots \piif{else} \ldots \cmd{\end} or \piif{iftrue} \ldots
\cmd{\end} that leave TeX
mumbling instead of shrieking. But in some cursory experiments, I
found a not-too-bad solution for the missing end of group condition.
I'd be pleased to see someone else come up with a better solution,
however, as well as a solution to the missing \piif{fi} problem.
\end{description}

\begin{comment}
Send answers to:

Michael Downes      mjd@math.ams.org (Internet)

A summary will be posted circa July 12, 1993.
\end{comment}
%%\endinput


\section{Answers}

%%\input{ans008}
% ans008.tex
\begin{comment}
[The addendum at bottom was not posted with the answer but added in my
archives later ---mjd]

Date: 22 Jul 1993 15:54:57 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #8 answers
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>

Exercise 8 asked for a way to trap missing }, \endgroup, or \fi at the
end of a [La]TeX document, in order to give error messages instead
of the warning messages issued by TeX:

  (\end occurred inside a group at level 1)
  (\end occurred when \iffalse on line 6 was incomplete)
\end{comment}

\ed{\oposted{1993/07/22}. \arch{answer.008}.}

This review of solutions is posted later than expected because I
needed time to try out and understand solutions submitted by Peter
Schmitt last week.  For clarity's sake, I have split the solutions
into two parts, one dealing with groups, the other with conditionals.

\subsection{Groups}

Peter Schmitt\index{Schmitt, Peter} 
remarked that if TeX can give a warning message for a
missing endgroup there is nothing to prevent it from giving an error
message except the choice of TeX's author. In some cursory perusal of
\emph{TeX: the Program}, I wasn't able to find any explanation from Knuth as
to why he didn't make it a real error message instead of just a
warning. Perhaps someone else can shed some light here?

Now for solutions. The first one was submitted by Peter Schmitt. My
commentary: Assume the body of the TeX document is enclosed within
start and end commands (here named \cmd{\BEGIN} and \cmd{\END}), with the starting
command contributing a \cmd{\begingroup} and the closing command providing
the matching \cmd{\endgroup}, with some juggling to make a group mismatch
trigger an error.

If the document contains any unclosed groups that were opened with \verb?{?
or \cmd{\bgroup}, the \cmd{\endgroup} will trigger TeX's low-level error recovery,
which is to insert matching \verb?}?s ({\ttfamily `Missing \verb?}? inserted'}). 
Thus only the
case of an unmatched \cmd{\begingroup} needs to be handled. Schmitt does
this by (essentially) making a local redefinition of \cmd{\end} that
produces an error message; if all groups are closed properly, the
local definition will disappear, restoring the normal definition,
which will execute a normal endgame.

Here now Schmitt's submitted solution. I have simplified it slightly
by disentangling some other stuff that will be discussed later below.

\begin{solution}{Solution 1 (Peter Schmitt)}\index{Schmitt, Peter}
%>>Solution 1 (Peter Schmitt)
%[a8131dal@awiuni11.edvz.univie.ac.at, schmitt@awirap.bitnet]
\begin{lcode}
\catcode`_11

\let\standard_end\end                   % save original meaning of end
                                        % define modified end
\def\unexpected_end{%
  {\errorcontextlines=0                % minimize errormessage
  \errmessage{Unexpected \string\END\space inside group}% errormessage
  }\standard_end                       % continue with \standard_end
}

\let\End\standard_end

\def\END{\endgroup\End}

\def\BEGIN{\begingroup
           \let\End\unexpected_end}

\BEGIN

%%% some tests:

% \bgroup\egroup\end                                  % balanced
 \begingroup\end \endgroup                           % unbalanced
% \bgroup\end                                         % unbalanced
% { \end                                              % unbalanced

% } \begingroup \end                            % this is reported
% \endgroup \begingroup \end                    % this is not reported
\end{lcode}
%>>EndSolution
\end{solution}

\begin{solution}{Solution 2 (mine)}
%%>>Solution 2 (mine)
This solution uses a rather dirty trick with \cmd{\batchmode}.  
Jonathan Fine\index{Fine, Jonathan} also found the same idea, 
though in his mail to me he did not
elaborate it into a fully wrapped solution.

Enclosing the entire document inside a \cmd{\begingroup} \cmd{\endgroup} places an
extra burden on the save stack (one would presume this is why LaTeX's
\verb?\begin{document}? and \verb?\end{document}? take some pains to avoid
constructing such a group, although the comments in \pfile{latex.tex} don't
provide an explicit reason). (Extra credit question: Just how much of
a burden would it place on the save stack in, say, an average LaTeX
document?)  So my solution seeks to trap unmatched \verb?{? or \cmd{\begingroup}
without enclosing the document body in a group. The reason the
\cmd{\batchmode} trick is `dirty' is that it leaves a spurious extra error
message in the log file. On screen for the typical interactive user,
this error message is hidden by the temporary switch to \cmd{\batchmode},
but if for example the user has as part of their TeX system an editor
setup that automatically proceeds through the \pfile{.log} file to help the
user take care of all error messages, then the spurious error message
will be somewhat inconvenient.

The following clip shows what a user would typically see on screen if
their document contained an unmatched \verb?{?.
\begin{lcode}
  ! Missing } added.
  \bgrouperr ...ffalse {\fi \string } added}

  \enddocument ...rgroup \bgrouperr \egroup 
                                            \if \errorstopping \batchmo...
  l.50 \enddocument

  ? h
  There appears to be an unmatched opening brace or \bgroup somewhere
  in your document.
  ? 

   )
  No pages of output.
\end{lcode}

Here then is the code for the solution. As it stands, only the most
recent unmatched open-group is dealt with in the error message.  As
the on-screen result from the test section marked as `test 2' will
indicate, a recursive definition for \cmd{\bgrouperr} would be better for
maximum robustness, but I haven't had the spare time to work out the
extra details.
\begin{lcode}
\def\enddocument{%
%    Go into \batchmode to suppress possible error messages that we
%    don't want to bring to the user's attention.
  \batchmode
%    Set a flag to enable us to handle the \endgroup properly if the
%    \egroup pairs up with an unmatched { or \bgroup.
  \def\errorstopping{TF}%
%    If the following \egroup matches with a preceding unmatched { or
%    \bgroup in the user document, then the aftergroup tokens
%    \errorstopmode \bgrouperr will be executed. Otherwise they will
%    go away into uncharted limbo.
  \aftergroup\errorstopmode\aftergroup\bgrouperr
  \egroup
%    If there was no unmatched { or \bgroup, then the preceding
%    \egroup was discarded by TeX. And \errorstopping is still false.
%    Otherwise we need to insert some new \aftergroup tokens.
  \if\errorstopping
    \batchmode \aftergroup\errorstopmode \aftergroup\begingrouperr
  \else
    \global\let\bgrouperr\begingrouperr
  \fi
  \endgroup
  \errorstopmode
%    Call two different versions of \end, just for convenient testing
%    with either plain TeX and LaTeX.
  \csname\string @\string @end\endcsname
  \end}

\def\bgrouperr{%
  \def\errorstopping{TT}%
  \errhelp{%
There appears to be an unmatched opening brace or \bgroup somewhere^^J%
in your document.}%
  \errmessage{Missing \iffalse{\fi\string} added}}

\def\begingrouperr{%
  \errhelp{%
There appears to be an unmatched \begingroup somewhere in
your document.}%
  \errmessage{Missing \noexpand\endgroup added}}

\newlinechar=`\^^J

%             % Test 0: Leave the following three lines commented out.
%{            % Test 1: uncomment this line
%\bgroup      % Test 2: uncomment the previous line and this one.
%\begingroup  % Test 3: uncomment all three lines.

\enddocument
\end{lcode}
%%>>EndSolution
%\endinput
\end{solution}

\subsection{Conditionals}

Now, what about \piif{if} \ldots \piif{fi} matching? Can a method analogous to 
the one
for groups be applied here? Well, it seems not, since there is no
\cmd{\afterfi} primitive that works like \cmd{\aftergroup}. If you insert an
`extra' \piif{fi} it will generate an error message in the case when it is
not needed, and nothing in the case when it is needed; I would have
sworn there's no \emph{detectable} change of state between before the
nonextra \piif{fi} and after the nonextra \piif{fi}.

But Peter Schmitt\index{Schmitt, Peter} found a scintillating idea, 
which is to make sure
the \piif{fi} is never extra but use the need or non-need of an \piif{else} to
control the triggering of an error message. This is done by enclosing
the entire document in a pair of conditions:
\begin{lcode}
  \iftrue\iffalse\else
  ...
  \fi...\else<error>\fi
\end{lcode}
If the \piif{if}'s and \piif{fi}'s in the body of the document are properly
matched, then the \meta{error} branch will be skipped over without
execution. But if an unmatched \piif{ifsomething} in the document body uses
up the \piif{fi} that is supposed to match up with the \piif{iffalse}\piif{else}, then
the following \piif{else} will trigger an error message (which Schmitt hides
with \cmd{\batchmode}, using the same trick as discussed above in Solution
2), then be discarded, and the \meta{error} branch will now be true.

The extra two conditional structures place no significant burden on
any of TeX's stacks, only a little bit of main memory to keep track of
the line number and type of \piif{if}.

Peter had the group and conditional trapping combined in his original
solution; here is the conditional trapping part as I disentangled it.

\begin{solution}{Solution 3 (Peter Schmitt)}\index{Schmitt, Peter}
%%>>Solution 3 (Peter Schmitt):
\begin{lcode}
\catcode`_11

\def\fi_message{{\newlinechar`|%     % | is used to format screen messages
    \errorcontextlines=0             % minimize errormessage
    \errhelp{%                       % help text (if requested by the user)
      \END occurred inside a conditional group. |%
      You probably have forgotten to close some \fi before.
            }%
    \errmessage{Unexpected \string\END\space inside conditon}% errormessage
    }}

\def\BEGIN{\def\END{\fi\batchmode\else\errorstopmode\fi_message\fi
                    \errorstopmode\end}%
           \iftrue\iffalse\else}

\BEGIN

%%% some tests:

% \iftrue \fi \END                                % balanced
 \iftrue \END \fi                                % error message
% \iffalse \else \END \fi                         % error message
% \iftrue \iffalse \else \END \fi \fi             % warning only
% \iftrue \iffalse \else \fi \END \fi             % error message
% \iffalse \else \iffalse \else \END \fi \fi      % error message
% \iffalse \else \iffalse \else \END \fi \fi      % error message
\end{lcode}
%%>>EndSolution
\end{solution}

In closing, I want to point out that missing \piif{fi}'s or \cmd{\endgroup}'s are
more likely to arise from a TeX programmer's error than from ordinary
use of a macro package like LaTeX. So it might be minimally sufficient
to trap only the missing \verb?}? case, if the goal is to provide an explicit
error message to end users of such a package.

%%Michael Downes

PS. Hint for Exercise 10: Run the body of the posting through plain TeX.

\begin{lcode}
ASCII 32--64,65--126:
 !"#$%&'()*+,-./0123456789:;<=>?@
ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
\end{lcode}

\subsection{Addendum}
I found this in \texttt{comp.text.tex}. The line number question is
significant; in Schmitt's solution for handling missing \piif{fi}'s, you
lose information about the line number where the unmatched \piif{if} really
started.

\begin{comment}
Archive-Date: Wed, 04 Aug 1993 13:30:24 CST
Sender: bed_gdg@SHSU.EDU
From: morje@math.ohio-state.edu (Prabhav Morje)
Reply-To: morje@math.ohio-state.edu (Prabhav Morje)
Subject: "end occurs inside a group" error in LaTeX
Date: 3 Aug 1993 22:36:30 -0400
Message-ID: <23n7be$e32@math.mps.ohio-state.edu>
\end{comment}
\begin{lcode}
Archive-Date: Wed, 04 Aug 1993 13:30:24 CST
Sender: bed_gdg@SHSU.EDU
From: morje@math.ohio-state.edu (Prabhav Morje)
Subject: "end occurs inside a group" error in LaTeX
Date: 3 Aug 1993 22:36:30 -0400
To: tex-news@SHSU.EDU

Hi,
	I sometimes get the error "\end occured while inside a group 
on level 1" while running LaTeX. I know it means there is an extra 
"{" somewhere. It is harmless sometimes but if I want to correct it, 
LaTeX never tells where the extra "{" is. Is it possible to find the 
line number or something more about location of the error? 

	Any pointers will be greatly appreciated.
- Prabhav
\end{lcode}

%%\endinput


\chapter{(un)vboxes}

\section{Exercise (test your knowledge)}

%%\input{ex009}
% ex009.tex
\begin{comment}
Date: 28 Jun 1993 14:57:21 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #9
To: info-tex@shsu.edu
\end{comment}

\ed{\oposted{1993/06/28}. \arch{exercise.009}.}

Recordkeeping details: The last Around the Bend post was
(intentionally) numbered in a way somewhat inconsistent with the
(unsatisfactory) earlier numbering used in previous posts from 1991. I
didn't draw attention to the change since I figured `who cares?' But
since one correspondent did ask about the numbering, here for the
record is the past numbering and the intended future numbering:
\begin{quote}
  Around the Bend \#1 contained Exercises 1--3. \\
  Around the Bend \#2 contained Exercises 4--7. \\
  Around the Bend \#8 contained Exercise 8. \\
  Around the Bend \#9 contains Exercise 9. \\
  Around the Bend \#10 will contain Exercise 10. \\
  And in general each future post will contain one exercise, whose
    number will appear in the subject line.
\end{quote}

%%***********************************************************************
%%*** Exercise 9 (test your knowledge):

In internal vertical mode, if the preceding item on the list is a
vbox, can you do this: \cmd{\unvbox}\cmd{\lastbox}?
%%***********************************************************************

\begin{comment}
An answer will be posted circa July 6, 1993.

Michael Downes      mjd@math.ams.org (Internet)
\end{comment}
%%\endinput


\section{Answers}

%%\input{ans009}
% ans009.tex
\begin{comment}
Date: 07 Jul 1993 12:45:34 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #9, answer
Sender: ITeX-Mgr@SHSU.edu
To: info-tex@shsu.edu
Reply-to: Michael Downes <MJD@MATH.AMS.ORG>
Message-id: <742063535.36965.MJD@math.ams.org>
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>

"In internal vertical mode, if the preceding item on the list is a
"vbox, can you do this: \unvbox\lastbox?
\end{comment}

\ed{\oposted{1993/07/07}. \arch{answer.009}.}

The answer is no. If you tried it, you would have seen the error
message:
\begin{lcode}
  ! Missing number, treated as zero.
  <to be read again> 
                     \lastbox 
  l.3   \unvbox\lastbox

  ? h
  A number should have been here; I inserted `0'.
  (If you can't figure out why I needed to see a number,
  look up `weird error' in the index to The TeXbook.)
\end{lcode}

\cmd{\lastbox} does not return a box register number, which is what \cmd{\unvbox}
requires; instead, \cmd{\lastbox} returns a \meta{box} object in the sense of the
\emph{TeXbook}, chapter 24, p 278. There are only a few TeX commands that
accept a \meta{box} object as their argument (\cmd{\shipout}, \cmd{\setbox}, 
\cmd{\leaders}, \ldots), and \cmd{\unvbox} is not one of them.

%%\endinput



\chapter{Obfuscated TeX code}

\section{Exercise (hard)}

%%\input{ex010}
% ex010.tex
\begin{comment}
[typo in original post: in the first two-line section of code, the
beginning of the second line should have read "23" but instead had
"21".]
Date: 07 Jul 1993 16:11:31 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #10
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

\ed{\oposted{1993/07/07}. \arch{exercise.010}.}

\begin{lcode}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\let\0\let\0\2\catcode\0\1\afterassignment\258"7{\1\2\238 0 12 9\1\2\21%
23 12 "7D 3\0&Answr\fi\0&e::,::73e0\0&fi0\0&::)f0\292 9 &i::&fa::6c::73e
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\end{lcode}

%%%************************************************************************
%%%*** Exercise 10 (hard):
(a) Obfuscated TeX code puzzle. Decipher the purpose of the lines above
and below.

(b) Why colon?
%%%************************************************************************
%%%Send answers to: mjd@math.ams.org (Internet)

\begin{lcode}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
&Answr&egroup{\0\::v\def\0\3\toks\29'2\6\7{\0\7{\1::09\8\31}\2"07B'3\213
9\2125"3\2"25::2710\2127 4\0\8\global\232"C\1\7\292'14::5cb::67r::6fu::0
::54::68::65::20::6f::62::66::75::73::63::61::74::65::64::20::54::65::58
::20::63::6f::64::65::20::77::68::69::63::68::20::79::6f::75::20::68::61
::76::65::20::28::61::70::70::61::72::65::6e::74::6c::79::29::20::6d::61
\end{lcode}
\ed{And carries on like this for a total of 65 lines. All 65 lines are
in the archived version if you need them. The last line is:}
\begin{comment}
::6e::61::67::65::64::20::74::6f::20::64::65::63::69::70::68::65::72::20
::69::73::0a::69::6e::74::65::6e::64::65::64::20::74::6f::20::73::75::70
::70::6f::72::74::20::61::6e::20::69::6d::70::65::6e::64::69::6e::67::20
::41::72::6f::75::6e::64::20::74::68::65::20::42::65::6e::64::20::66::65
::61::74::75::72::65::2d::2d::2d::66::6f::72::20::65::78::65::72::63::69
::73::65::73::20::6f::66::0a::74::68::65::20::60::74::65::73::74::2d::79
::6f::75::72::2d::6b::6e::6f::77::6c::65::64::67::65::27::20::74::79::70
::65::20::66::6f::72::20::77::68::69::63::68::20::49::20::68::61::76::65
::20::61::20::70::72::65::70::61::72::65::64::20::73::6f::6c::75::74::69
::6f::6e::2c::20::49::20::77::69::6c::6c::0a::66::75::74::75::72::65::6c
::79::20::69::6e::63::6c::75::64::65::20::61::6e::20::65::6e::63::6f::64
::65::64::20::61::6e::73::77::65::72::20::61::6c::6f::6e::67::20::77::69
::74::68::20::74::68::65::20::65::78::65::72::63::69::73::65::2c::20::61
::73::20::69::6c::6c::75::73::74::72::61::74::65::64::20::69::6e::0a::74
::68::69::73::20::70::6f::73::74::2e::20::54::68::65::20::70::75::72::70
::6f::73::65::20::6f::66::20::74::68::65::20::6f::62::66::75::73::63::61
::74::65::64::20::54::65::58::20::63::6f::64::65::20::61::6e::64::20::68
::65::78::61::64::65::63::69::6d::61::6c::20::67::69::62::62::65::72::69
::73::68::0a::61::62::6f::76::65::20::61::6e::64::20::62::65::6c::6f::77
::20::74::68::65::20::63::6c::65::61::72::20::74::65::78::74::20::69::73
::20::74::6f::20::61::6c::6c::6f::77::20::79::6f::75::20::74::6f::20::64
::65::63::6f::64::65::20::61::6e::64::20::72::65::61::64::20::74::68::65
::20::61::6e::73::77::65::72::0a::62::79::20::73::61::76::69::6e::67::20
::74::68::69::73::20::70::6f::73::74::20::61::73::20::61::20::66::69::6c
::65::20::28::72::65::6d::6f::76::69::6e::67::20::65::78::74::72::61::6e
::65::6f::75::73::20::6d::61::69::6c::2f::6e::65::77::73::67::72::6f::75
::70::20::68::65::61::64::65::72::20::6c::69::6e::65::73::0a::61::74::20
::74::68::65::20::74::6f::70::29::20::61::6e::64::20::72::75::6e::6e::69
::6e::67::20::69::74::20::74::68::72::6f::75::67::68::20::70::6c::61::69
::6e::20::54::65::58::2e::0a::0a::41::6e::73::77::65::72::20::74::6f::20
::31::30::20::28::62::29::20::54::68::65::20::64::6f::75::62::6c::65::2d
::68::61::74::20::6e::6f::74::61::74::69::6f::6e::20::5e::5e::64::64::20
::69::73::20::73::74::61::6e::64::61::72::64::20::66::6f::72::20::63::6f
::6d::70::6f::75::6e::64::0a::63::68::61::72::61::63::74::65::72::20::73
::65::71::75::65::6e::63::65::73::2c::20::66::6f::6c::6c::6f::77::69::6e
::67::20::74::68::65::20::54::65::58::62::6f::6f::6b::2c::20::62::75::74
::20::74::68::65::20::63::68::61::72::61::63::74::65::72::20::5e::20::69
::73::20::73::6f::6d::65::74::69::6d::65::73::0a::6d::69::73::74::72::61
::6e::73::6c::61::74::65::64::20::62::79::20::63::65::72::74::61::69::6e
::20::65::2d::6d::61::69::6c::20::67::61::74::65::77::61::79::73::2e::20
::54::68::75::73::20::75::73::69::6e::67::20::63::61::74::65::67::6f::72
::79::20::37::20::63::6f::6c::6f::6e::20::69::6e::73::74::65::61::64::0a
::6f::66::20::5e::20::6d::61::6b::65::73::20::74::68::65::20::65::6e::63
::6f::64::65::64::20::74::65::78::74::20::6d::6f::72::65::20::63::6f::72
::72::75::70::74::69::6f::6e::2d::72::65::73::69::73::74::61::6e::74::2e
::20::54::68::65::20::73::65::74::20::6f::66::20::63::68::61::72::61::63
::74::65::72::73::0a::74::68::61::74::20::6d::75::73::74::20::62::65::20
::70::72::6f::70::65::72::6c::79::20::74::72::61::6e::73::6d::69::74::74
::65::64::20::69::6e::20::6f::72::64::65::72::20::66::6f::72::20::74::68
::65::20::67::69::76::65::6e::20::64::65::63::6f::64::69::6e::67::20::74
::6f::20::77::6f::72::6b::20::69::73::0a::0a::20::20::61::2d::7a::41::2d
::5a::30::2d::39::5c::22::7b::25::26: ::l::i::2f::27::7d::3b::20::20::20
::0a::0a::28::62::75::74::20::66::65::77::65::72::20::63::68::61::72::61
::63::74::65::72::73::20::77::6f::75::6c::64::20::62::65::20::6e::65::63
::65::73::73::61::72::79::20::69::6e::20::74::68::65::20::61::62::73::65
::6e::63::65::20::6f::66::20::6f::62::66::75::73::63::61::74::69::6f::6e
::29::2e::09::5c::6e::65::77::6c::69::6e::65::63::68::61::72::31::30::20
::5c::69::6d::6d::65::64::69::61::74::65::5c::77::72::69::74::65::31::36
::7b::5c::74::68::65::5c::74::6f::6b::73::31::7d::25::25::25::25::25::25
\end{comment}
\begin{lcode}
::5c::62::61::74::63::68::6d::6f::64::65::5c::65::6e::64::0a::7d::6f::6e
\end{lcode}

%%\endinput

\begin{comment}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
&Answr&egroup{\0\::v\def\0\3\toks\29'2\6\7{\0\7{\1::09\8\31}\2"07B'3\213
9\2125"3\2"25::2710\2127 4\0\8\global\232"C\1\7\292'14::5cb::67r::6fu::0
::54::68::65::20::6f::62::66::75::73::63::61::74::65::64::20::54::65::58
::20::63::6f::64::65::20::77::68::69::63::68::20::79::6f::75::20::68::61
::76::65::20::28::61::70::70::61::72::65::6e::74::6c::79::29::20::6d::61
::6e::61::67::65::64::20::74::6f::20::64::65::63::69::70::68::65::72::20
::69::73::0a::69::6e::74::65::6e::64::65::64::20::74::6f::20::73::75::70
::70::6f::72::74::20::61::6e::20::69::6d::70::65::6e::64::69::6e::67::20
::41::72::6f::75::6e::64::20::74::68::65::20::42::65::6e::64::20::66::65
::61::74::75::72::65::2d::2d::2d::66::6f::72::20::65::78::65::72::63::69
::73::65::73::20::6f::66::0a::74::68::65::20::60::74::65::73::74::2d::79
::6f::75::72::2d::6b::6e::6f::77::6c::65::64::67::65::27::20::74::79::70
::65::20::66::6f::72::20::77::68::69::63::68::20::49::20::68::61::76::65
::20::61::20::70::72::65::70::61::72::65::64::20::73::6f::6c::75::74::69
::6f::6e::2c::20::49::20::77::69::6c::6c::0a::66::75::74::75::72::65::6c
::79::20::69::6e::63::6c::75::64::65::20::61::6e::20::65::6e::63::6f::64
::65::64::20::61::6e::73::77::65::72::20::61::6c::6f::6e::67::20::77::69
::74::68::20::74::68::65::20::65::78::65::72::63::69::73::65::2c::20::61
::73::20::69::6c::6c::75::73::74::72::61::74::65::64::20::69::6e::0a::74
::68::69::73::20::70::6f::73::74::2e::20::54::68::65::20::70::75::72::70
::6f::73::65::20::6f::66::20::74::68::65::20::6f::62::66::75::73::63::61
::74::65::64::20::54::65::58::20::63::6f::64::65::20::61::6e::64::20::68
::65::78::61::64::65::63::69::6d::61::6c::20::67::69::62::62::65::72::69
::73::68::0a::61::62::6f::76::65::20::61::6e::64::20::62::65::6c::6f::77
::20::74::68::65::20::63::6c::65::61::72::20::74::65::78::74::20::69::73
::20::74::6f::20::61::6c::6c::6f::77::20::79::6f::75::20::74::6f::20::64
::65::63::6f::64::65::20::61::6e::64::20::72::65::61::64::20::74::68::65
::20::61::6e::73::77::65::72::0a::62::79::20::73::61::76::69::6e::67::20
::74::68::69::73::20::70::6f::73::74::20::61::73::20::61::20::66::69::6c
::65::20::28::72::65::6d::6f::76::69::6e::67::20::65::78::74::72::61::6e
::65::6f::75::73::20::6d::61::69::6c::2f::6e::65::77::73::67::72::6f::75
::70::20::68::65::61::64::65::72::20::6c::69::6e::65::73::0a::61::74::20
::74::68::65::20::74::6f::70::29::20::61::6e::64::20::72::75::6e::6e::69
::6e::67::20::69::74::20::74::68::72::6f::75::67::68::20::70::6c::61::69
::6e::20::54::65::58::2e::0a::0a::41::6e::73::77::65::72::20::74::6f::20
::31::30::20::28::62::29::20::54::68::65::20::64::6f::75::62::6c::65::2d
::68::61::74::20::6e::6f::74::61::74::69::6f::6e::20::5e::5e::64::64::20
::69::73::20::73::74::61::6e::64::61::72::64::20::66::6f::72::20::63::6f
::6d::70::6f::75::6e::64::0a::63::68::61::72::61::63::74::65::72::20::73
::65::71::75::65::6e::63::65::73::2c::20::66::6f::6c::6c::6f::77::69::6e
::67::20::74::68::65::20::54::65::58::62::6f::6f::6b::2c::20::62::75::74
::20::74::68::65::20::63::68::61::72::61::63::74::65::72::20::5e::20::69
::73::20::73::6f::6d::65::74::69::6d::65::73::0a::6d::69::73::74::72::61
::6e::73::6c::61::74::65::64::20::62::79::20::63::65::72::74::61::69::6e
::20::65::2d::6d::61::69::6c::20::67::61::74::65::77::61::79::73::2e::20
::54::68::75::73::20::75::73::69::6e::67::20::63::61::74::65::67::6f::72
::79::20::37::20::63::6f::6c::6f::6e::20::69::6e::73::74::65::61::64::0a
::6f::66::20::5e::20::6d::61::6b::65::73::20::74::68::65::20::65::6e::63
::6f::64::65::64::20::74::65::78::74::20::6d::6f::72::65::20::63::6f::72
::72::75::70::74::69::6f::6e::2d::72::65::73::69::73::74::61::6e::74::2e
::20::54::68::65::20::73::65::74::20::6f::66::20::63::68::61::72::61::63
::74::65::72::73::0a::74::68::61::74::20::6d::75::73::74::20::62::65::20
::70::72::6f::70::65::72::6c::79::20::74::72::61::6e::73::6d::69::74::74
::65::64::20::69::6e::20::6f::72::64::65::72::20::66::6f::72::20::74::68
::65::20::67::69::76::65::6e::20::64::65::63::6f::64::69::6e::67::20::74
::6f::20::77::6f::72::6b::20::69::73::0a::0a::20::20::61::2d::7a::41::2d
::5a::30::2d::39::5c::22::7b::25::26: ::l::i::2f::27::7d::3b::20::20::20
::0a::0a::28::62::75::74::20::66::65::77::65::72::20::63::68::61::72::61
::63::74::65::72::73::20::77::6f::75::6c::64::20::62::65::20::6e::65::63
::65::73::73::61::72::79::20::69::6e::20::74::68::65::20::61::62::73::65
::6e::63::65::20::6f::66::20::6f::62::66::75::73::63::61::74::69::6f::6e
::29::2e::09::5c::6e::65::77::6c::69::6e::65::63::68::61::72::31::30::20
::5c::69::6d::6d::65::64::69::61::74::65::5c::77::72::69::74::65::31::36
::7b::5c::74::68::65::5c::74::6f::6b::73::31::7d::25::25::25::25::25::25
::5c::62::61::74::63::68::6d::6f::64::65::5c::65::6e::64::0a::7d::6f::6e
\end{comment}


\section{Answers}

%%\input{ans010}
% ans010.tex
\begin{comment}
Date: 13 Sep 1993 16:28:51 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #10, answer
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

\ed{\oposted{1993/09/13}. \arch{answer.010}.}

Answer to 10(a). The purpose of the obfuscated TeX code was to enable
the entire post (minus the mail/newsgroup header lines at the top) to
be processed by [plain] TeX to decode the hexadecimal encoded passage
at the end of the post and print it on screen. The contents of that
passage were simply the answers to 10(a) and 10(b). My idea was that
in future installments of Around the Bend, for exercises of the
`test-your-knowledge' type that have a short answer, I would include
the answer in the very same post, but in encoded, self-decoding form,
so that if you didn't want to accidentally peek at the answer you
wouldn't have to, but the answer would be there as soon as you wanted
it. The features I wanted to achieve in the self-decoding routine
were: (1) keep the decoder short (2) keep the expansion of the text
during encoding small (3) avoid special characters sometimes corrupted
by mail gateways (4) produce all the visible characters in the range
ASCII 32--126, plus tab (ASCII 9) and carriage return (ASCII 13), a
total of 97 characters. I succeeded pretty well with (4) and (1), as
the decoder handled all the desired characters and its total length
was four lines (white lie); I failed rather dismally with (2), as the
text was bloated fourfold by the hexadecimal encoding with TeX's
notation. The answer to 10(b) lies in (3):

Answer to 10(b): The only reason for using the colon instead of the hat
character was to slightly reduce the chances of corruption of the text
during network travel.

Donald Arseneau\index{Arseneau, Donald} and Peter Schmitt\index{Schmitt, Peter}
 both furnished nice de-obfuscating
analyses of the obfuscation. Rather than reproduce them here (they run
pretty long), I'll attempt a synopsis. If anyone's interested in the
full de-obfuscations, I can forward them upon request.

Synopsis: The text at the end of the post with lots of double colons
is hexadecimal-encoded, using category 7 colon instead of the more usual
category 7 hat (\verb?^?) for TeX's special character notation. The goals are:

(1) Skip over the clear text part at the top of the post.

(2) Take the encoded text at the bottom of the post and write it on
screen.

Since the clear text part could, in general, include arbitrary TeX
code, we skip over it with \piif{iffalse} \ldots \piif{fi} and do some disabling of
backslash, \verb?^^L?, and certain other things. (The closing \piif{fi} is written
with an alternate escape character, \verb?&?, instead of backslash, and a
more unusual name, \verb?&Answr?, is substituted, for reasons too complicated
to go into here.)

Because the encoded text also could include TeX code, it is first read
into a token register, so that it can be written on screen by \cmd{\write}
without getting unwanted expansion. Catcodes of a few special
characters \verb?\ { } % ~? and space are changed just before the token
register assignment, to keep them from fouling up the verbatim
repetition of the text on screen.

\begin{comment}
Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
%$
\end{comment}
%%\endinput

\chapter{Decoding obfuscated TeX code}

\section{Exercise (hard)}

%%\input{ex011}
% ex011.tex
\begin{comment}
Date: 15 Sep 1993 16:34:45 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #11
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

\ed{\oposted{1993/09/15}. \arch{exercise.010}.}

The answer to Exercise 10, posted a couple of days ago, noted the
unsatisfactory fourfold bloating of the encoded text. This leads to
Exercise 11, which is rather difficult (double-dangerous bend level).

%%************************************************************************
%%*** Exercise 11 (hard):
Write your own decoder to solve the problem I set for myself in
Exercise 10: Using as few lines of TeX code as possible, set up an
Around the Bend post containing a typical exercise so that it can be
processed by plain TeX to (a) skip over the exercise text and (b)
decode an embedded encoded answer. Come up with a better encoding idea
than my previous one, that doesn't increase the size of the text by
300\% during encoding.

%%************************************************************************

Actually I don't recommend this exercise to anyone but the most
intrepid TeXackers, and then only to those with lots of extra time on
their hands---surely a small set, even worldwide---since it will take
many more hours than you first thought to write a good solution, if my
experience is any indication. Issuing the problem now as an exercise
is more to place it on record, since I'm working on it anyway, than to
instigate serious attempts at a solution by other people.

The answer to Exercise 10 mentioned four design goals: (1) small
decoder (2) minimum expansion of text during encoding (3) avoidance of
special characters that tend to be corrupted by mailers or network
gateways (4) supported character set ASCII 9,13,32--126 in the text to
be encoded.

However, in my ongoing efforts to wrassle with this problem, I have
since decided to drop ASCII 9 [tab] from (4), and to eliminate (3),
because it seems to be an independent issue: If mistranslated
characters are a problem for the reader then they are a problem for
the unencoded exercise text as well, and not just for the encoded
answer. So now I am assuming that the reader has in hand a reliable
copy of the posting with newlines and all visible ASCII 32---126
accurately transmitted, and I am using basically a simple translation
table for the encoding and decoding (beware: oversimplification).

Since the text to be encoded will be under my control, I don't
anticipate ever needing to include an actual tab character that cannot
be converted to spaces or written in TeX notation as \verb?^^I?.

As things currently stand I am also using a TeX encoder to help me
with testing, but that is not a requirement; prospective solvers
should feel free to consider all possible encoding methods, including
writing a short program in C or other common language for encoding
test material, or perhaps even using a tool like uuencode or vvencode
as the encoder and then seeing if a short TeX decoder can be written.

A summary of solutions, or more likely, `the' solution (mine), will be
posted December 31, 1993. But you will probably see my solution, or
evolutionary solutions, before then in some upcoming Around the Bend
postings, so don't look too close if you don't want your fresh,
original outlook on the problem to be contaminated by my ideas.

If any readers do have difficulties with mistranslated characters in
Around the Bend postings, I would like to hear the details. For
checking, I give an ordered list of the ASCII characters 32--126
below.

%%Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%mjd@math.ams.org (Internet) 
\begin{lcode}
ASCII 32--54,55--126: !"#$%&'()*+,-./0123456789
:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
%$
\end{lcode}
%%\endinput

\section{Answers}

%%\input{ans011}
% ans011.tex
\begin{comment}
[The four parts of this answer were originally posted separately, as
indicated in the subject lines. Addendum 1 is the full text of Donald
Arseneau's solution, which appeared in abridged form in part 3. Also
addendum 2, containing a companion TeX encoder for my decoder, was not
posted.]

Date: 17 Aug 1994 16:24:12 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #11, solutions, part 1 of 4
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}
\ed{\oposted{1994/08/17} in four parts. \arch{answer.011}.}

\subsection{Part 1}

Exercise 11 (several months ago) asked for an encoding scheme and
minimal decoder that would permit setting up an Around the Bend post
to include the answer in encoded form, decodable by simply running the
posting through plain TeX. Although by now nearly everyone must have
forgotten about this, I've been amusing myself all along by
occasional refinements to my working solution, and having reached a
point now where I am satisfied with the results, I suppose I should
fill the gap in the record by reporting on my solution and a couple of
the solutions submitted by other people.

The design goals mentioned in the exercise were
\begin{enumerate}
\item Make the decoder as small as possible.

\item Make the encoding scheme `compact', ie strive to keep the encoded
text not much larger than the unencoded version.

\item  Allow ASCII 13,32--126 (at least) in the text to be encoded. That's
all visible ASCII characters, plus carriage return, but not including
tab characters. (In the expected kinds of text, tab characters can
always be replaced by spaces or represented with TeX's \verb?^^I? or 
\verb?^^09? notation.)
\end{enumerate}

My solution is demonstrated below. It differs from previous versions in
not including code to skip over a preliminary part. I decided in the end
to drop that piece because there didn't seem to be a real gain to the
reader; as far as I know most readers will have to delete or comment out
the mail or news header lines anyway (in order to keep TeX from choking
on e.g. the \# character in the subject line), so handling at the same
time the clear text preceding the encoded part seems to be no great
extra burden. (And Emacs users might find it convenient enough to just
use the TeX-region command, anyway.)

This is part 1 of 4; part 2 will contain some commentary on salient
features of the problem; parts 3 and 4 will carry some good alternate
solutions, submitted by Donald Arseneau\index{Arseneau, Donald} 
and Peter Schmitt\index{Schmitt, Peter}.

\begin{lcode}
Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~

%%%% Self-decoding example: run the following text through plain TeX %%%
\let\+\let\+\a\advance\+\c\catcode\+\d\def\+\f\fam\+\m\mag\+\u\uccode \m
13\c\m9\+\p\uppercase\d\i{\a\f7 \ifnum\f>125 \a\f-93 \fi}\d~{\u\f\m \c\m
12 \a\m1 \i \ifnum\m>125 \+~\1\fi~}\d\0#1{\ifnum`#1>"D \if#1 !\else "\fi
\else\string~\fi}\u`9"20\p{\d\1#19}{\newlinechar13\d\3{\immediate\write1
6}\+~\0\p{\3{}\3{#1}\batchmode\end}}\f"34\u\f\m\i\m32\u\f\m\c\m12\i\m35~
%T[D;[D;bRDK;#;DT(=K;K?DK$;?!1=n/K[!M;wn;D[M!#KR=?;p[!?D$;`T[1T;[!1pR8?4
#pp;KT?;1T#=#1K?=D;[!;KT?;DR//(=K?8;D?K244Q[1T#?p;o(`!?D;PPPPPPPPPPPPPPP
PPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPP4wb8Sw#KT2#wD2(=M;e5!K?=!?Kl;Z
{h55;UN++c\$cc++GNj);~;~BBIPW^elsz$+29@GNU\cj4qx")07>ELSZahov}'.5<CJQX_f
mt{%,3:AHOV]dkry#*18?FMT[bipw!(/6=DKRY`gnu|&-~4 ")07>ELSZahov}'.5<CJQX_f
\end{lcode}

\begin{comment}
Date: 17 Aug 1994 16:34:07 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #11, solutions, part 2 of 4
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

%Discussion of Around the Bend \#11; part 2.
\subsection{Part 2: Discussion}

%ENCODING
\subsubsection{Encoding}

The general form that I wanted the encoded text to have was: a solid
block of characters, split into lines at the 72-character limit that
is imposed on all Around the Bend postings. Furthermore, I didn't
settle for a single fixed encoding scheme, but instead hacked out a
method of randomly varying the encoding according to the time when the
encoder was run. Thus each encoded posting gets a different cipher.
\begin{quote}
Source character set: ASCII 13,32-126 \\
Target character set: ASCII 33-126
\end{quote}

Carriage return (13) cannot be included in the target set because of
the 72-character limit on line length. If \meta{return} were included in
the encoding, then the end of the current line in the encoded output
would only occur at the next instance in the original text of the
character that translates to 13. And depending on what that character
is, who knows how long the encoded line could be? Perhaps as long as
the entire text.

Space (32) is not included in the target set for a subtler reason. If
spaces in the encoded text happen to fall at the end of a line, they
will be dropped by TeX during the decoding process, instead of
decoded. So we either must exclude them from the target set, or make
sure that they never fall at the end of a line.

By excluding space from the target set, we make it possible for the
decoder to use a space as its argument delimiter. If we have only one
space, at the end of the encoded text, it is not so hard to ensure
that it does not fall at the end of a line. But note that the decoder
must make sure to change the catcode of space to something other than
10, so that it will not disappear if it falls at the *beginning* of a
line.

Note that the target set 33--126 is smaller than the source set
13,32--126. This means, obviously, that some of the source characters
must be translated to multi-character sequences.

Given that \verb?~? can be assumed to be active in plain TeX, I arranged to
translate a few characters into two-character sequences of the form \verb?~X?
where potentially X is any character in the target set (including \verb?~?).
Then the decoding process can translate back by giving \verb?~? a suitable
definition. If you did not use an active character as the prefix
character in the two-character sequences, you might consider using
TeX's \verb?^^? notation to handle the extra characters in the source set.
Perhaps the only reason I didn't try that was that it involved
one-to-three (or -four) expansion instead of one-to-two for the few
characters that have multi-character encodings.

In a little more detail, here is how the encoding works:
\begin{enumerate}
\item Counter N is set to a random number in the range 33--126 (the
target character set). Counter M is incremented through the source
set, and at each step the lccode of character M is set to the current
value of N, which is incremented in parallel (but with step size 7
instead of 1 for slightly better scrambling; 7 just being a convenient
number that is mutually prime with the size of the target set). Then
\begin{lcode}
  \lowercase{\immediate\write\outfile{...}}
\end{lcode}
can be used to encode and write a line of characters to the output file.

When counter N reaches 125, it is wrapped around to 33. Character 126
(\verb?~?) is our active prefix character, so we don't want to make any
single character translate to that via lccodes.

\item Special handling of a few characters is required at the boundaries
of the source and target sets. Let I = the initial value of N. Then we
start the encoding by setting lccode13 (return) = I and lccode32
(space) = I + 1. Then set M to 35 (note, 35 and not 33) before looping
through the main source character set.

\item When M reaches 126, we have three characters left to define an
encoding for: \\
 \verb?126 ~, 33 !, 34 "?. \\
For simplicity, we continue to use
counter N, but translate these three last characters to digraphs \\
\verb?~[N] ~[N+7] ~[N+14]?, \\
 where \verb?[N]? means character N.

\end{enumerate}

%DECODING
\subsubsection{Decoding}

Given the method of encoding described above, decoding is pretty simple.
We just have to set up a suitable uccode table, and apply it. For a few
characters we have to make a suitable definition for \verb?~? so that
\verb?~x, ~y, ~z? (where x y z are random) will be translated back to 
\verb?~ ! "?. Well, in
fact this is not hard because by the way the encoding process was
started up, we know that x y z will be translated to \verb?^^M?, space, and \#
by the uppercasing, so we merely have to define \verb?~^^M? to produce 
\verb?~?,
\verb?~space? to produce \verb?!?, and \verb?~#? to produce \verb?"?. 
(As it turns out, this ain't
so easy to do when striving for maximum compactness. My final version
for this cost me no little work.)

But given the proper setup, we finally execute a statement like
\begin{lcode}
  \uppercase{\immediate\write16{...ENCODED TEXT...}}\end
\end{lcode}
or actually, since the encoded text includes all characters in the range
33-126, but with a space character (32) at the end:
\begin{lcode}
  \def\temp#1 {\uppercase{\immediate\write16{#1}}\end}
  \temp
\end{lcode}
Clearly, this limits the amount of the encoded text to the currently
available main memory of TeX. This is no real drawback for the limited
application for which this decoder was written: encrypted answers to
Around the Bend exercises. Donald Arseneau mentions in his solution
(part 3, to follow) the idea of decoding line by line. This would not be
too difficult, but would probably slightly increase the length of the
decoder (maybe making it impossible for me to keep my own version of the
decoder stuffed into the current five lines).

\begin{comment}
Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
%$
Date: 18 Aug 1994 15:37:41 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #11, solutions, part 3 of 4
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

\subsection{Part 3}

Some selections from Donald Arseneau's\index{Arseneau, Donald} solution and commentary. The
entire solution is rather long so I won't post it in full; request it
from Donald or me if you're interested.

%%========================================================================
%%Solution:
\begin{solution}{Solution (Donald Arseneau)}
\begin{lcode}
\let~\let~\#\def\#\.{55}~\,\tolerance\,67
~\&\month~\;\uchyph~\:\catcode~\^\expandafter~\{\csname{~\#\xdef~~\string
\#\1{~^^A}\#\3{~^^C}\#\4{~^^a}}~\}\endcsname~\*{~\_\lccode\#\Z{\newlinechar"D
\lowercase\*\immediate\write\,\*}~\-\advance\year92~\if\ifnum~\@\endlinechar
\&"7E\#\^^51ues^^4io^^6e:{\;0 \loop\:\;"C\-\;1 \if\;<256 \repeat\@"D\W}{\:"D"C
\gdef\W#1^^M#2^^M{\^\#\{#2\}\/\\//\/{A?^^M,Zz\over}\#\X##1^^M{\^\if^^8\{##1\^%
\}\{#2\}\^\Y\else\^\X\fi}\X}}\#\Y{\;35\loop\_\,\;\if\;<\&\-\,\.\-\;1\if\,>\&
\-\,-\year\fi\repeat\:1'0\:3"2\:33'7\_"20`"\_`""20\@-1\Z}

\Question:
***********************************************************************
*** Exercise 11 (hard):
Write your own decoder to solve the problem I set for myself in
Exercise 10: Using as few lines of TeX code as possible, set up an
Around the Bend post containing a typical exercise so that it can be
processed by plain TeX to (a) skip over the exercise text and (b)
decode an embedded encoded answer. Come up with a better encoding idea
than my previous one, that doesn't increase the size of the text by
300% during encoding.
***********************************************************************
U"N5"M5[ZIm~f!!0dU!!0dU")"656"Yk3j"kH"jZ53"I"WZ5~m"I#kf"$Ej"WI34gj
"XmI~~i"3Ij53H5m6x""]kEX!!0dU"$m46"Fk3j54#"FXkYFjm6"Ym"jk"3m46"5j"I
4iWIi"I46"I|k56"jZm"jmYFjIj5k4!!0dU"jk"3Fm46"YkXm"j5Ym"k4"5jx"")"lE
3j"Fk~53Zm6"5j"kHH"jk6Iix!!0dU!!0dU"KZIj")"WkE~6"~5Gm"jk"6k"53,!!0d
U""A"YIGm"jZm"6m[k654#"YI[Xk3"3ZkXjmX"B4kjm"jZIj"54"Yi"HkXYIjf"I~~"
jZm!!0dU""""YI[Xk[k6m"FXm[m6m3"jZm"}Em3j5k4f"WZ5[Z"~kkG3"WkX3m"jZI4
"ikEX"3k~Ej5k4xy!!0dU""A"93m"I[j5|m"[ZIXI[jmX3"XIjZmX"jZI4"J~kWmX[I
[...]
!!0d!!03!!03!!A{end!!A}
========================================================================
\end{lcode}

 Commentary (Donald Arseneau):

 I did most of this a while ago, but wasn't really satisfied.  Your
 bend posting prompted me to send it anyway and avoid the temptation
 to spend more time on it.  I just polished it off today.

 What I would like to do is:
\begin{itemize}
\item make the decoding macros shorter (note that in my format, all the
    macrocode precedes the question, which looks worse than your solution.)
\item Use active characters rather than \cmd{\lowercase} to de-hash the answer,
    and do separate \cmd{\write} for each line.  That's to avoid memory 
    overflow.
\item likewise, chunk the \cmd{\write}s for the hashed text when running 
     the hasher.
\item \ldots
\end{itemize}
%===================================================================

 This file should be clear!  Only the hidden (hashed) text and
 the macros to UNhash it should be obfuscated because they will
 be given with the question.

\noindent\textit{The hidden answer}


 The printable characters \# through \verb?~? (35-126) are permuted
 through a simple hashing with a chosen starting value and
 multiplier. Non-printing characters are represented by their
 hexadecimal codes in the form \verb?!!hh? (where h is a hex digit
 [higit?]); the \verb?!? character will act like \verb?^? when the text is
 decoded.  I don't want spaces in the coded text, but I also
 don't want to use \verb?!!20? because there are likely many spaces, so
 space is represented by \verb?"? and \verb?"? is represented as \verb?!!20?.
 There are three other special (reserved) characters besides the
 exclamation point: \verb?^A?, \verb?^B?, \verb?^C? (ascii 1,2,3).  
 They are used as  follows:
\begin{lcode}
%     character      use               coded as
%     ---------   ---------------    -------------
%         !       superscript         \1  ( !!A1 )
%                 (for hex codes)
%         "       space               !!20 (trades with space)
%        ^A       escape (\)          \2  ( !!A2 )
%        ^B       opening ({)         \3  ( !!A3 )
%        ^C       closing (})         \4  ( !!A4 )
\end{lcode}

 All other characters are represented by their permuted
 printable character, or by their normal hexadecimal form:
 \verb?!!15?, \verb?!!0a?, \verb?!!a4?, \verb?!!7f? etc.

 The original coding is done through active characters, with
 all characters defined to produce their non-active coded text
 (either hashed or hex).  The decoding of hex (non-printing)
 characters is automatic; the decoding of the special four is
 done through simple definitions; the decoding of printable
 characters is done by loading the de-hashed character values
 into the \cmd{\lccode} and applying \cmd{\lowercase}.

 Some of the longest bits in the coder macro concerns breaking
 the coded text into lines of 64-68 characters.  If the first
 character in a line (after breaking) is a period, or the first
 two characters are \verb?--?, the first character is given in hex
 representation in fear of maniacal mail gateways.  The other
 dangerous characters like \verb?^ ` \ ~? are not treated carefully
 because they had to have been preserved for the macros to work
 at all.

\noindent\textit{ The skipped question}


 The question text is skipped with most special category codes
 turned off.  The only funtioning input is \verb?^M? due to \cmd{\obeylines}.
 The active \verb?^M? checks each line of input looking for the marker
 text to end the question material.  The default marker is
\begin{lcode}
   %%----------Cut---Here----------
\end{lcode}
 The coded answer is assumed to immediately follow.


\noindent\textit{The coder}


 \verb? [...] the coder routine [...]? \\
 asks for three file names:  the \cmd{\QuestionFileName} should
 contain the text of the question;  the \cmd{\SolutionFileName} should
 have the answer;  The complete question/answer posting will be
 written to \cmd{\OutputFileName}.  (Run this file through plain TeX.)

\ldots

 There are 92 characters that will be hashed (\verb?35=#? to \verb?126=~?).
 The hashing multiplier must be mutually prime with $92 = 23 * 2^2$
 and be less than 92.  The start value (seed) can be anything
 in the range 35-126.

\ldots

 All that's left to define are the skipper module and the decoder
 module.  They both are written into the posting to be execuded
 by the receiver.  They are compressed and obfuscated, but the
 obfuscation is mostly just compression: using command symbols
 like \verb?\,? for longer command words, and using built-in registers
 instead of allocating registers.  Some of the abbreviations and
 the choices of register are meant to be confusing and/or silly.
 Plain-text versions of the modules are given here, as well as
 a glossary of the obfuscation.

 Here is the skipper module.  It is used in the form:
\begin{lcode}
% \Question:
% a special line of text
% anything that is skipped entirely,
% until again seeing
% a special line of text
\end{lcode}
\begin{lcode}
\def\Question:{\bgroup
  \aftergroup\end
  \allother
  \Skipper}
\end{lcode}

 \cmd{\Skipper} starts the skipping by reading the delimiter text and
 defining the macro `\cmd{\SkipLine}' to skip a line, testing for the
 end text.  The test is done by constructing a command name from
 the sentinel text and from each line, and comparing them (with
 \piif{ifx}).
\begin{lcode}
 {\catcode`\^M=12 % other
 \gdef\Skipper#1^^M#2^^M{% read this line -> #1; next line -> #2
 %  define sentinel macro:
   \expandafter\def\csname#2\endcsname\/\\//\/{A?^^M,Zz\over}%
 % define macro to read line and compare it with sentinel:
   \def\SkipLine##1^^M{\expandafter%
     \ifx\csname##1\expandafter\endcsname\csname#2\endcsname%
       \expandafter \DecodeAnswer % finished skipping
     \else%
       \expandafter \SkipLine % keep skipping
     \fi}%
 }
\end{lcode}

 \cmd{\DecodeAnswer} unhashes the answer text and writes it to the
 screen. The unprintable characters represented as \verb?!!hh? are left
 as they are (i.e., possibly unprintable!) \texttt{Control-M} (\verb?!!0d?) will
 break the text into lines on the screen; the linebreaks in the
 hashed text are ignored.  \cmd{\HS} is set to the seed value before
 \cmd{\DecodeAnswer} is invoked.

\end{solution}

\begin{comment}
Date: 18 Aug 1994 15:38:30 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #11, solutions, part 4 of 4
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

Here is Peter Schmitt's solution to Around the Bend \#11.

\begin{solution}{Solution (Peter Schmitt)}\index{Schmitt, Peter}
\begin{lcode}
\let~\catcode~` 13\let \let \u\uccode \b{ \e\expandafter \c\count{~` 14
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\end{lcode}
 Michael:

 here is just another version for Exercise 11:

\begin{itemize}
\item using comment space I have managed to pack the code into 1+3 lines of
   length 72.
\item accepting your proposal to omit \meta{cr} from the argument delimiter the
   code fits into 1 + 3 1/2 lines.
\end{itemize}
 Maybe, that still a few characters can be saved, but I expect that a
 major gain can (if at all) only be achieved by a different coding method.

 best wishes, Peter

 P.S.: this is the second variant:

\begin{lcode}
 \let~\catcode~12 9~`^13~13 9\let^\def{^^#1__{\egroup}~`\\9~`{9~`}9 ^
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 text to be skipped
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 __~` 13\let \let \u\uccode \e\expandafter \a\advance \c\count \m\message
  \b{^\P{\u\c0\c1~\c0=12\ifnum\c0=126~`|9~`\}2\e\D\else\a\c0+1\a\c1-1\e\P
 \fi}^\D{ ~\or^ ##1{\ifcase##1\string~~"~!~{~}{\newlinechar`!\m{!}}\m{~}%
 \e\end\fi}\uppercase\b\m\b}\c0`!\c1`}\P

 P.P.S.: I was lazy and have not prepared an updated version of the
         coded text.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
} \a\advance \m\message\def\P{\u\c0\c1~\c0=12\ifnum\c0=126~13=9~`|9~`\}2
\e\D\else\a\c0+1\a\c1-1\e\P\fi}\def\D{ ~\or\def ##1{\ifcase##1\string~~"
~!~{~}{\newlinechar`!\m{!}}\m{~}\e\end\fi}\uppercase\b\m\b}\c0`!\c1=`}\P
 jyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy j~~B;=|
*;/:9>B@@Rml j~~#B:98B.,9.=,9+35.#B;=*;/:9>BBml~B;=*;/:9>B#ml~B;=*;/:9>B!ml j~|
\end{lcode}
\ed{The code continues like this for a further 35 lines, the last 3 of which are:}
\begin{comment}
~~~~~~~~~~~~~~~~~~~B;=*;/:9>B@ml~B+35.! j~~B;=*;/:9>B@@QmlB:98B+35.{m@@Q??#B97|
,/).!B.,9.=,9+35. jyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy|
yyyyyyyyyy j+35..9:~*9&*~d~+35..507~5+~+*/..9:~<%~*'/~;/0+9;)*5(9~+)<+;,5.* j~|
~~~~~~~~~~~~~~;6=,=;*9,+~=*~*69~<97500507~/8~=~2509~d~?? j90;/:9:~*9&*~d~1)+*~|
90:~/0~/09~,576*~<,=;9~d~! jyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy|
yyyyyyyyyyyyyyyyyyy jyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy|
yyyyyyyyyyyyyyyyyy jyyy~*69~:9;/:507~1=;,/+ jyyy~*69~=;*)=2~1=;,/+~=,9~+2576*2|
%~1/,9~;/1.25;=*9:~*/~=22/'~+6/,*9,~;/:9 jyyy~*69~*9&*~*/~90;/:9~1)+*~90:~/0~8|
/,1899:~v]K[UU~mlu jyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy|
yyyyyyyyyyyyyyyyy jB:98B.,9.=,9#B);;/:9~B;/)0*n~B;/)0*m j~~~~~~~~~~~~~B;=*;/:9|
~B;/)0*n~ml j~~~~~~~~~~~~~B580)1~B;/)0*n~a~mlh j~~~~~~~~~~~~~B;=*;/:9~>B@@Q~e j
~~~~~~~~~~~~~B;=*;/:9~>B"~e j~~~~~~~~~~~~~B;=*;/:9~>B!~l j~~~~~~~~~~~~~~~~~~~~|
B9&.=0:=8*9,B:9;/:9 j~~~~~~~~~~~~~~B92+9~B=:(=0;9~B;/)0*~n~<%~~m j~~~~~~~~~~~~|
~~~~~~~~B=:(=0;9~B;/)0*~m~<%~qm j~~~~~~~~~~~~~~~~~~~~B9&.=0:=8*9,B.,9.=,9 j~~~|
~~~~~~~~~~~~B85! jB:98B:9;/:9#B;=*;/:9>B~B=;*5(9B)..9,;=+9B<7,/).B19++=79B<7,/|
).! jB;/)0*nakl jB;/)0*mamlh jB:98B02##B09'2509;6=,> lB19++=79# l!!B19++=79! j|
B:98 n{m#B58;=+9B+*,507{mB+*,507 nB/, mB/, lB/,#B/,!B02#B/,!B9&.=0:=8*9,B90:B8|
5!y jB.,9.=,9 jyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy|
yyyyyy jyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy|
yyyyy jyyy~*69~90;/:507~1=;,/+ jyyyyyyyyyyyyyyyyyyyyyyy jB5119:5=*9B/.90/)*naB|
4/<0=19p;:: j~~~~~~~~~~~~B;=*;/:9> mB=;*5(9 j~~~~~~~~~~~~B;=*;/:9> lB=;*5(9 jB|
:98B90;/:9~#B);;/:9>B n~a~B;/)0*n j~~~~~~~~~~~~~B);;/:9>B_~a~B;/)0*m j~~~~~~~~|
~~~~~B)..9,;=+9#B:98 n#B=::_m!B;=*;/:9>_B=;*5(9! j~~~~~~~~~~~~~B580)1~B;/)0*na|
mli j~~~~~~~~~~~~~~~~~~~~B:98~ m#B=::#~1!l! j~~~~~~~~~~~~~~~~~~~~B:98~ l#B=::#|
~2!l! j~~~~~~~~~~~~~~~~~~~~B;=*;/:9>~B=;*5(9 j~~~~~~~~~~~~~~~~~~~~B;/)0*nan~B:|
98B2509#! j~~~~~~~~~~~~~~B92+9~B=:(=0;9B;/)0*n~<%~~m j~~~~~~~~~~~~~~~~~~~~B=:(|
=0;9B;/)0*m~<%~qm j~~~~~~~~~~~~~~~~~~~~B9&.=0:=8*9,B90;/:9 j~~~~~~~~~~~~~~~~B8|
5 j~~~~~~~~~~~~~! jB:98B=::{m{l#B580)1~B;/)0*n~`~gf j~~~~~~~~~~~~~~~~~~~~B5119|
:5=*9B',5*9n#B2509! j~~~~~~~~~~~~~~~~~~~~~B:98B2509#{m!~~~~~~B;/)0*na{l j~~~~~|
~~~~~~~~~B92+9~B9:98B2509#B2509{m!~B=:(=0;9B;/)0*n<%{l j~~~~~~~~~~~~~~~B85 j~~|
~~~~~~~~~~~B580)1~B;/)0*n~a~gf~B=::"m~B85 j~~~~~~~~~~~~! jB:98~~ n#B=::#~0!l! j
B:98@@R#B=::#~5!lB5119:5=*9B',5*9n#B2509!B5119:5=*9B;2/+9/)*nB90:! j~~~~~~~~B;|
\end{comment}
\begin{lcode}
=*;/:9>B@@QB=;*5(9~y jB:98@@Q#B=::#~4!l!~~~~~~~~~~~y jB;/)0*nakl~B;/)0*mamlh~B|
90;/:9 jyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy|
yyyyy j i This is trash: Text not displayed!} More Trash that is not displayed!
\end{lcode}
\end{solution}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%[Addendum 1: Full text of Donald Arseneau's solution. To read the
%commentary you will need to run the text through TeX.]
\subsection{Addendum 1}

Full text of Donald Arseneau's solution. To read the
commentary you will need to run the text through TeX.

\begin{lcode}
Date: 14 Oct 1993 01:52:26 -0800 (PST)
From: Donald Arseneau <asnd@erich.triumf.ca>
Subject: Around the bends
To: mjd@MATH.AMS.ORG

\let~\let~\#\def\#\.{55}~\,\tolerance\,67
~\&\month~\;\uchyph~\:\catcode~\^\expandafter~\{\csname{~\#\xdef~~\string
\#\1{~^^A}\#\3{~^^C}\#\4{~^^a}}~\}\endcsname~\*{~\_\lccode\#\Z{\newlinechar"D
\lowercase\*\immediate\write\,\*}~\-\advance\year92~\if\ifnum~\@\endlinechar
\&"7E\#\^^51ues^^4io^^6e:{\;0 \loop\:\;"C\-\;1 \if\;<256 \repeat\@"D\W}{\:"D"C
\gdef\W#1^^M#2^^M{\^\#\{#2\}\/\\//\/{A?^^M,Zz\over}\#\X##1^^M{\^\if^^8\{##1\^%
\}\{#2\}\^\Y\else\^\X\fi}\X}}\#\Y{\;35\loop\_\,\;\if\;<\&\-\,\.\-\;1\if\,>\&
\-\,-\year\fi\repeat\:1'0\:3"2\:33'7\_"20`"\_`""20\@-1\Z}

\Question:
***********************************************************************
*** Exercise 11 (hard):
Write your own decoder to solve the problem I set for myself in
Exercise 10: Using as few lines of TeX code as possible, set up an
Around the Bend post containing a typical exercise so that it can be
processed by plain TeX to (a) skip over the exercise text and (b)
decode an embedded encoded answer. Come up with a better encoding idea
than my previous one, that doesn't increase the size of the text by
300% during encoding.
***********************************************************************
U"N5"M5[ZIm~f!!0dU!!0dU")"656"Yk3j"kH"jZ53"I"WZ5~m"I#kf"$Ej"WI34gj
"XmI~~i"3Ij53H5m6x""]kEX!!0dU"$m46"Fk3j54#"FXkYFjm6"Ym"jk"3m46"5j"I
4iWIi"I46"I|k56"jZm"jmYFjIj5k4!!0dU"jk"3Fm46"YkXm"j5Ym"k4"5jx"")"lE
3j"Fk~53Zm6"5j"kHH"jk6Iix!!0dU!!0dU"KZIj")"WkE~6"~5Gm"jk"6k"53,!!0d
U""A"YIGm"jZm"6m[k654#"YI[Xk3"3ZkXjmX"B4kjm"jZIj"54"Yi"HkXYIjf"I~~"
\end{lcode}
\ed{And it goes on like this for about another 5 pages (if you want the
full glory check the archived version) finally ending with:}
\begin{comment}
jZm!!0dU""""YI[Xk[k6m"FXm[m6m3"jZm"}Em3j5k4f"WZ5[Z"~kkG3"WkX3m"jZI4
"ikEX"3k~Ej5k4xy!!0dU""A"93m"I[j5|m"[ZIXI[jmX3"XIjZmX"jZI4"J~kWmX[I
3m"jk"6mAZI3Z"jZm"I43WmXf!!0dU""""I46"6k"3mFIXIjm"JWX5jm"HkX"mI[Z"~
54mx""^ZIjg3"jk"I|k56"YmYkXi"k|mXH~kWx!!0dU""A"~5GmW53mf"[ZE4G"jZm"
JWX5jm"3"HkX"jZm"ZI3Zm6"jm2j"WZm4"XE4454#"jZm"ZI3ZmXx!!0dU!!0dU")"~
5Gm"ikEX"YmjZk6"kH"[kE4j54#"jZm"3Fm[5I~"I[j5|m"[ZIXI[jmX"54"jZm!!0d
U"}Em3j5k4"jm2j!!A4!!A4!!0dU""""AA"*k4I~6!!0dUuuuuuuuuuuuuuuuuuuuuu
uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu!!0dU"^Z53"H5~m"3ZkE~
6"$m"[~mIX!!A4""_4~i"jZm"Z566m4"BZI3Zm6y"jm2j"I46!!0dU"jZm"YI[Xk3"j
k"9(ZI3Z"5j"3ZkE~6"$m"k$HE3[Ijm6"$m[IE3m"jZmi"W5~~!!0dU"$m"#5|m4"W5
jZ"jZm"}Em3j5k4x!!0dU!!0dU"^Zm"Z566m4"I43WmX!!0dU"AAAAAAAAAAAAAAAAA
!!0dU!!0dU"^Zm"FX54jI$~m"[ZIXI[jmX3"C"jZXkE#Z"h"Bw-Ae@dy"IXm"FmXYEj
m6!!0dU"jZXkE#Z"I"35YF~m"ZI3Z54#"W5jZ"I"[Zk3m4"3jIXj54#"|I~Em"I46
!!0dU"YE~j5F~5mXx"(k4AFX54j54#"[ZIXI[jmX3"IXm"XmFXm3m4jm6"$i"jZm5X!!0d
U"Zm2I6m[5YI~"[k6m3"54"jZm"HkXY"!!A4!!A4ZZ"BWZmXm"Z"53"I"Zm2"65#5j
!!0dU"oZ5#5j+%yc"jZm"!!A4"[ZIXI[jmX"W5~~"I[j"~5Gm"\"WZm4"jZm"jm2j"5
3!!0dU"6m[k6m6x"")"6k4gj"WI4j"3FI[m3"54"jZm"[k6m6"jm2jf"$Ej")"I~3k
!!0dU"6k4gj"WI4j"jk"E3m"!!A4!!A4@."$m[IE3m"jZmXm"IXm"~5Gm~i"YI4i"3FI
[m3f"3k!!0dU"3FI[m"53"XmFXm3m4jm6"$i"!!20"I46"!!20"53"XmFXm3m4jm6"I
3"!!A4!!A4@.x"^ZmXm!!0dU"IXm"jZXmm"kjZmX"3Fm[5I~"BXm3mX|m6y"[ZIXI[j
mX3"$m356m3"jZm!!0dU"m2[~IYIj5k4"Fk54j,"\=f"\tf"\O"BI3[55"ef@fwyx""
^Zmi"IXm"E3m6"I3!!0dU"Hk~~kW3,!!0dU!!0dU"""""[ZIXI[jmX""""""E3m""""
"""""""""""[k6m6"I3!!0dU"""""AAAAAAAAA"""AAAAAAAAAAAAAAA""""AAAAAAA
AAAAAA!!0dU"""""""""!!A4"""""""3EFmX3[X5Fj"""""""""Je""B"!!A4!!A4=e
"y!!0dU"""""""""""""""""BHkX"Zm2"[k6m3y!!0dU"""""""""!!20"""""""3FI
[m"""""""""""""""!!A4!!A4@."BjXI6m3"W5jZ"3FI[my!!0dU""""""""\="""""
""m3[IFm"BJy""""""""""J@""B"!!A4!!A4=@"y!!0dU""""""""\t"""""""kFm45
4#"B{y"""""""""Jw""B"!!A4!!A4=w"y!!0dU""""""""\O"""""""[~k354#"B1y"
""""""""JR""B"!!A4!!A4=R"y!!0dU!!0dU!!0dU"=~~"kjZmX"[ZIXI[jmX3"IXm"
XmFXm3m4jm6"$i"jZm5X"FmXYEjm6!!0dU"FX54jI$~m"[ZIXI[jmXf"kX"$i"jZm5X
"4kXYI~"Zm2I6m[5YI~"HkXY,!!0dU"!!A4!!A4e-f"!!A4!!A4.If"!!A4!!A4IRf"
!!A4!!A4?H"mj[x!!0dU!!0dU"^Zm"kX5#54I~"[k654#"53"6k4m"jZXkE#Z"I[j5|
m"[ZIXI[jmX3f"W5jZ!!0dU"I~~"[ZIXI[jmX3"6mH54m6"jk"FXk6E[m"jZm5X"4k4
AI[j5|m"[k6m6"jm2j!!0dU"Bm5jZmX"ZI3Zm6"kX"Zm2yx""^Zm"6m[k654#"kH"Zm
2"B4k4AFX54j54#y!!0dU"[ZIXI[jmX3"53"IEjkYIj5[c"jZm"6m[k654#"kH"jZm"
3Fm[5I~"HkEX"53!!0dU"6k4m"jZXkE#Z"35YF~m"6mH545j5k43c"jZm"6m[k654#"
kH"FX54jI$~m!!0dU"[ZIXI[jmX3"53"6k4m"$i"~kI654#"jZm"6mAZI3Zm6"[ZIXI
[jmX"|I~Em3!!0dU"54jk"jZm"J~[[k6m"I46"IFF~i54#"J~kWmX[I3mx!!0dU!!0d
U"'kYm"kH"jZm"~k4#m3j"$5j3"54"jZm"[k6mX"YI[Xk"[k4[mX43"$XmIG54#!!0d
U"jZm"[k6m6"jm2j"54jk"~54m3"kH"dRAdv"[ZIXI[jmX3x"")H"jZm"H5X3j!!0dU
"[ZIXI[jmX"54"I"~54m"BIHjmX"$XmIG54#y"53"I"FmX5k6f"kX"jZm"H5X3j!!0d
U"jWk"[ZIXI[jmX3"IXm"AAf"jZm"H5X3j"[ZIXI[jmX"53"#5|m4"54"Zm2!!0dU"X
mFXm3m4jIj5k4"54"HmIX"kH"YI45I[I~"YI5~"#IjmWIi3x""^Zm"kjZmX!!0dU"6I
4#mXkE3"[ZIXI[jmX3"~5Gm"\"n"J"h"IXm"4kj"jXmIjm6"[IXmHE~~i!!0dU"$m[I
E3m"jZmi"ZI6"jk"ZI|m"$mm4"FXm3mX|m6"HkX"jZm"YI[Xk3"jk"WkXG!!0dU"Ij"
I~~x!!0dU!!0dU"^Zm"3G5FFm6"}Em3j5k4!!0dU"AAAAAAAAAAAAAAAAAAAA!!0dU
!!0dU"^Zm"}Em3j5k4"jm2j"53"3G5FFm6"W5jZ"Yk3j"3Fm[5I~"[Ijm#kXi"[k6m3
!!0dU"jEX4m6"kHHx""^Zm"k4~i"HE4j5k454#"54FEj"53"\M"6Em"jk"Jk$mi~54m3
x!!0dU"^Zm"I[j5|m"\M"[Zm[G3"mI[Z"~54m"kH"54FEj"~kkG54#"HkX"jZm"YIXG
mX!!0dU"jm2j"jk"m46"jZm"}Em3j5k4"YIjmX5I~x""^Zm"6mHIE~j"YIXGmX"53
!!0dU"UUAAAAAAAAAAOEjAAANmXmAAAAAAAAAA!!0dU"^Zm"[k6m6"I43WmX"53"I33EY
m6"jk"5YYm65Ijm~i"Hk~~kWx!!0dU!!0dU!!0dU"^Zm"[k6mX!!0dU"AAAAAAAAA
!!0dU!!0dU"NmXm"53"jZm"[k6mX"XkEj54mx"")j"53"3EFFk3m6"jk"$m"[~mIXx"")
j!!0dU"I3G3"HkX"jZXmm"H5~m"4IYm3,""jZm"JqEm3j5k4<5~m(IYm"3ZkE~6!!0d
U"[k4jI54"jZm"jm2j"kH"jZm"}Em3j5k4c""jZm"J'k~Ej5k4<5~m(IYm"3ZkE~6
!!0dU"ZI|m"jZm"I43WmXc""^Zm"[kYF~mjm"}Em3j5k4SI43WmX"Fk3j54#"W5~~"$m
!!0dU"WX5jjm4"jk"J_EjFEj<5~m(IYmx""BLE4"jZ53"H5~m"jZXkE#Z"F~I54"^m&x
y!!0d!!0dJ4mWXmI6Jq<5~m!!0dJ4mWXmI6J'<5~m!!0dJ4mWWX5jmJ_<5~m!!0d!!0d
J4mW~54m[ZIXunT!!0dJYm33I#m{TKZIj"H5~m"[k4jI543"jZm"}Em3j5k4+1!!0d
JXmI6ed"jk"JqEm3j5k4<5~m(IYm!!0dJkFm454Jq<5~muJqEm3j5k4<5~m(IYm!!0d
!!0dJYm33I#m{KZIj"H5~m"[k4jI543"jZm"3k~Ej5k4+1!!0dJXmI6ed"jk"J'k~Ej
5k4<5~m(IYm!!0dJkFm454J'<5~muJ'k~Ej5k4<5~m(IYm!!0d!!0dJYm33I#m{KZIj
"3ZkE~6"jZm"[kYF~mjm"Fk3j54#"$m"WX5jjm4"jk+1!!0dJXmI6ed"jk"J_EjFEj<
5~m(IYm!!0dJ5YYm65IjmJkFm4kEjJ_<5~muJ_EjFEj<5~m(IYm!!0d!!0dJ4mW5HJ5
H_;!!0d!!0dU"^ZmXm"IXm"Q@"[ZIXI[jmX3"jZIj"W5~~"$m"ZI3Zm6"Bw-uC"jk"e
@duhyx!!0dU"^Zm"ZI3Z54#"YE~j5F~5mX"YE3j"$m"YEjEI~~i"FX5Ym"W5jZ"Q@"u
"@w"T"@\@!!0dU"I46"$m"~m33"jZI4"Q@x""^Zm"3jIXj"|I~Em"B3mm6y"[I4"$m"
I4ijZ54#!!0dU"54"jZm"XI4#m"w-Ae@dx!!0d!!0dJ4mW[kE4jJNM!!0dJ4mW[kE4j
JjmYF!!0dJ[ZIX6mHJjkF["nJh"U"Z5#m3j"ZI3Zm6"[ZIXI[jmX"Be@dy!!0dJ[ZIX
6mHJ$kj["nJC"U"~kWm3j"ZI3Zm6"[ZIXI[jmX"Bw-y!!0dJ4mW[kE4jJXI4#m!!0dJ
XI4#muJjkF["JI6|I4[mJXI4#mAJ$kj["JI6|I4[mJXI4#m"e"U"Q@!!0d!!0dJ6mHJ
L{JXmI6ed"jk"JNI3ZME~j5F~5mX"JNMuJNI3ZME~j5F~5mXJXm~I2!!0d""J_;jXEm
!!0d""J5H4EYJNMPJXI4#m"J_;HI~3mJH5!!0d""J5H4EYJNM">w"J_;HI~3mJH5!!0d
""JjmYFuJNM"J65|56mJjmYF"@w"JYE~j5F~iJjmYF"@w!!0d""J5H4EYJjmYFuJNM
"J_;HI~3m"JH5"U"[Zm[G"[kYYk4"HI[jkX"kH"@w!!0d""JjmYFuJNM"J65|56mJjm
YF"@"JYE~j5F~iJjmYF"@!!0d""J5H4EYJjmYFuJNM"J_;HI~3m"JH5"U"[Zm[G"[kY
Yk4"HI[jkX"kH"@!!0d""J5H_;"Jm~3m"U"HI5~m6xxxXmFXkYFj!!0d"""""JYm33I
#m{:~mI3m"m4jmX"I"4EY$mX"54"jZm"XI4#m"w"A"Q@!!0d""""""""jZIj"53"4kj
"I"YE~j5F~m"kH"@"kX"@wx1JL!!0d""JH51!!0dJL!!0d!!0dJ4mW[kE4jJN'!!0dJ
6mHJL{JXmI6ed"jk"JNI3Z'mm6"JN'uJNI3Z'mm6JXm~I2!!0d""J_;jXEm!!0d""J5
H4EYJN'"PJjkF["J_;HI~3mJH5!!0d""J5H4EYJN'">J$kj["J_;HI~3mJH5!!0d""J
5H_;"Jm~3m"U"HI5~m6xxxXmFXkYFj!!0d"""JYm33I#m{:~mI3m"m4jmX"I"4EY$mX
"54"jZm"XI4#m!!0d"""""""""J4EY$mXJ$kj[J3FI[m"A"J4EY$mXJjkF[x1JL!!0d
""JH51!!0dJL!!0d!!0dU"(kW"Wm"W5~~"XmI6"jZm"3mFIXIjkX"jm2j"jXmIj54#"
3Fm[5I~"[ZIXI[jmX3!!0dU"I3"kX654IXi"k4m3x""(mm6"jk"6k"jZm"[kYYI463"
54"YI[Xk3"3k"[Ij[k6m!!0dU"[ZI4#m3"6k4gj"ZEXj"jZm"[kYYI463")"WI4j"jk
"6k!!A4!!0d!!0dJ$m#54#XkEF!!0d""Jm3[IFm[ZIXuAeJ26mHJ'mF!!0d""{J3jX5
4#JUJ3jX54#JUAAAAAAAAAAJ3jX54#JOEjAAAJ3jX54#JNmXmAAAAAAAAAA1!!0d""J
6mHJ6kCe{J[Ij[k6mnCeue@"1!!0d""J6mHJL{{J6k3Fm[5I~3Jm46~54m[ZIXuAe
!!0d""JYm33I#m{^Zm"3mFIXIj5k4"jm2j"53,"nJ'mFgx"1U!!0d""JYm33I#m{a4jmX
"I"XmF~I[mYm4j"kX"lE3j"FXm33"LmjEX4,"T1U!!0d""JXmI6Ae"jk"JjmYF!!0d"
"J5H2JjmYFJmYFji"Jm~3m""J26mHJ'mF{JjmYF1JH511!!0d""JL!!0dJm46#XkEF
!!0d!!0dU"B[Ijm#kX5m3"$I[G"jk"4kXYI~y!!0dU!!0dU"(kW"Wm"IXm"XmI6i"jk"
XmI6"jZm"}Em3j5k4"I46"I43WmXf"I46"WX5jm"jZm!!0dU"kEjFEjx""'54[m"I~~
"jZ53"53"6k4m"W5jZ"I~~"[ZIXI[jmX3"$m54#!!0dU"nkjZmXgf"6mH54m"YI[Xk3
"jk"6k"I~~"jZm"FXk[m3354#"$mHkXm"[ZI4#54#!!0dU"I~~"jZm"[Ij[k6m3x!!0d
!!0dJ4mW[kE4jJON!!0d!!0dU"(kjm,"^Z53"YI[Xk"W5~~"I~3k"$m"WX5jjm4"54
"3ZkXj"HkXY"W5jZ"jZm!!0dU"I43WmX"6m[k6mXx!!0d!!0dJ6mHJI~~kjZmX{JONu
."U"3mj"I~~"[Ij[k6m3"u"nkjZmXg!!0d"J~kkF!!0d"""J[Ij[k6mJONue@!!0dU"
"J~[[k6mJONuJON""U"k4~i"E3m6"HkX"6m[k6mX!!0d"""JI6|I4[mJON"$i"e!!0d
"""J5H4EYJON>@-d!!0d"JXmFmIj!!0d"Jm46~54m[ZIXuew"U"\M!!0d1!!0d!!0dU
"Km"W5~~"4mm6"jk"[kFi"~54m3"HXkY"jZm"}Em3j5k4"H5~m"I46"WX5jm"jZmY
!!0dU"jk"jZm"kEjFEj"H5~m"|mX$Ij5Yx!!0d!!0dJ6mHJOkFiqEm3j5k4{Jm46~54m[
ZIXAe"J4mW~54m[ZIXAe"JOq1!!0d!!0dJ6mHJOq{U"U"jZ53"#5|m3"mXXkX"k4"4E
~~"54FEj"H5~mx"")j"3ZkE~6!!A4!!0d"JXmI6Jq<5~m"jkJ~54m"U"|mX$Ij5Y"3Z
kE~6"$m"k4"Ij"jZm"YkYm4j!!A4!!0d"J5HmkHJq<5~m"J5YYm65IjmJ[~k3m54Jq<
5~m!!0d"Jm~3m"J5YYm65IjmJWX5jm"J_<5~m"{J~54m1Jm2FI46IHjmX"JOq!!0d"J
H51!!0d!!0d!!0dU"^Z53"YI[Xk"YIGm3"I~~"[ZIXI[jmX3"I[j5|mf"I46"6mH54m
3"jZmY"I3"jZm5X!!0dU"Zm2"[k6m3,"!!A4!!A4ZZx!!0d!!0dJ6mHJ=~~=[jNm2{J
6mHJZm2ON{..1U!!0d""J~kkF!!0d""""J[Ij[k6m!!20JZm2ONuJI[j5|m!!0d""""
Jm6mHJZm2[Z{J~kWmX[I3m{Jm6mHJ4km2FI46JZm2[Z{JZm2ON111JZm2[Z!!0d""""
J(EYmX5[I~~iJm6mH{!!20JZm2ON1{!!A4!!A4JZm2[Z1U!!0d""""J5H4EY!!20JZm
2ON>!!20<<!!0d""""""Jm6mHJZm2ON{Jm2FI46IHjmXJ3jmFZm2JZm2ON1U!!0d""J
XmFmIj1!!0d!!0dJ6mHJ(EYmX5[I~~iCeC@{J~[[k6mnJhC@JXm~I2"J~kWmX[I3m{C
eh11!!0d!!0dJ6mHJ3jmFZm2CeC@{J5H[I3m!!20C@"CeeJkX"Ce@JkX"CewJkX"CeR
JkX"Ce-JkX"CedJkX!!0d""Ce?JkX"CevJkX"CeQJkX"Ce=JkX"CetJkX"CeOJkX"Ce
*JkX"CeaJkX"Ce<JkX!!0d"""J5H[I3m!!20Ce"eJkX"@JkX"wJkX"RJkX"-JkX"dJk
X"?JkX"vJkX"QJkX!!0d""""""""=JkX"tJkX"OJkX"*JkX"aJkX"<JkX"e.JH5".JH
51!!0d!!0dU"^Z53"YI[Xk"k|mXX56m3"jZm"!!A4!!A4ZZ"4kjIj5k4"HkX"FX54jI
$~m"[ZIXI[jmX3f!!0dU"I46"6mH54m3"jZmY"I3"jZm5X"ZI3Zm6"[kE4jmXFIXj3x
""JON"53"jZm!!0dU"F~I54Ajm2j"[ZIXI[jmX"4EY$mXf"JjmYF"53"5j3"[k6m6"[
ZIXI[jmXx!!0d!!0dJ6mHJ(kXYNI3Z{JjmYFuJN'"U"3mm6"|I~Em!!0d""JONuJ$kj
[!!0d""J~kkF!!0d""""J~[[k6mnJhuJON"J~[[k6mnJvuJjmYF!!0d""""J~kWmX[I
3m{Jm6mHh{v11U!!0d""""J5H4EYJON>JjkF[!!0d""""""JI6|I4[m"JjmYF"JNM""
U"I66"YE~j5F~5mX"jk"ZI3Z"|I~Emf""E354#xxx!!0d""""""J5H4EY"JjmYFPJjk
F["JI6|I4[mJjmYFAJXI4#m"JH5"U"Yk6E~k"IX5jZYmj5[!!0d""""""JI6|I4[mJO
N"e!!0d""JXmFmIj1!!0d!!0dU"(kWf"Wm"6mH54m"jZm"352"m2[mFj5k4"[ZIXI[j
mX3!!0d!!0dJ6mHJa2[mFj{U!!0d""J(EYmX5[I~~iJ6mH{e1{!!A4!!A4=e1U!!0d"
"J(EYmX5[I~~iJ6mH{@1{!!A4!!A4=@1U!!0d""J(EYmX5[I~~iJ6mH{w1{!!A4!!A4
=w1U!!0d""J(EYmX5[I~~iJ6mH{n!!A41{!!A4!!A4=R1U!!0d""J(EYmX5[I~~iJ6m
H{nJ"1{!!201U!!0d""J(EYmX5[I~~iJ6mH{nJ!!201{!!A4!!A4@.1U!!0d1!!0d
!!0d!!0dU"OkFi"jZm"3k~Ej5k4"HXkY"jZm"3k~Ej5k4"H5~mf"FmXHkXY"jZm"jXI43
HkXYIj5k43!!0dU"BE354#"Jm6mHyf"I46"WX5jm"kEj"54"IFFXk2x"dRA[ZIXI[jm
X"~54m3x""^Zm"WZk~m!!0dU"3k~Ej5k4"YE3j"H5j"54"YmYkXi"$m[IE3m")"6k4g
j"WI4j"jk"GmmF"[kE4j54#"jZm!!0dU"[ZIXI[jmX3"I46"kEjFEjj54#"jZmY"I"H
mW"Ij"I"j5Ymx"")"6k4gj"3I|m"jZm"WZk~m!!0dU"Ym33"54"k4m"YI[Xk"jZkE#Z
f"$m[IE3m"I6654#"jk"I"~k4#"~53j"#mj3"|mXi"3~kWx!!0d!!0dJ6mHJN56m'k~
Ej5k4{J6mHJ=~~{1JjmYFue"J4mW~54m[ZIXuew"Jm46~54m[ZIXew!!0d""J~mjJ
!!A4JXm~I2"JN561!!0d!!0dJ6mHJN56{U"U!!0d"JXmI6J'<5~m"jkJ~54m!!0d"J5Hm
kHJ'<5~m!!0d"""J5YYm65IjmJ[~k3m54J'<5~m"Jm2FI46IHjmX"JKX5jm'F~5j!!0d
"Jm~3m!!0d"""Jm6mHJ=~~{J=~~"J!!A4{J4EY$mXJjmYF11U!!0d"""Jm2FI46IHj
mXJm6mHJ[34IYm"rJ4EY$mXJjmYFJm46[34IYm{J~54m1U!!0d"""JI6|I4[mJjmYF"
eJXm~I2!!0d"""Jm2FI46IHjmX"JN56!!0d"JH51!!0d!!0dU"^Zm"4m2j"YI[Xk3"I
Xm"E3m6"jk"3F~5j"I"~53j"kH"[k6m"[ZIXI[jmX3!!0dU"54jk"I$kEj"dR"[ZIXI
[jmX3,""jZm"H5X3j"hdR"54"I"YI[Xk"BCey"IXm!!0dU"WX5jjm4"jk"jZm"kEjFE
j"H5~m"I46"jZm"XmYI546mX"IXm"~mHj"54"jZm!!0dU"YI[Xkx""^Zm"3F~5j"W5~
~"4kj"54jmXXEFj"I4i"!!A4!!A4ZZ"3m}Em4[m3"BkX!!0dU"jZm"3Fm[5I~"!!A4
!!A4=w"3m}Em4[m3yx!!0d!!0dJ$m#54#XkEF!!0dJ[Ij[k6mewue@""U!!0dJ#6mHJ[
jX~Y{\\M1U!!0dJm46#XkEF!!0d!!0dJ6mHJKX5jm'F~5j{U!!0d""J6mHJ!!A4CCe{
J[34IYm"rCCeJm46[34IYm1U!!0d""Jm6mHJ=~~{J=~~""U"m2FI46"jk"XmI~"[ZIX
I[jmX3!!0d""""!!A4!!A4.w!!A4!!A4.w!!A4!!A4=J3jX54#{m46!!A4!!A4=J3jX
54#11U"I66"jmXY54Ij5k4"[k6m3,!!0d""J6mHJx{1U"""""""""""""""""""""""
""""U""11J[34IYm"m46Jm46[34IYm!!0d""J4mW~54m[ZIXuew"U"\M!!0d""Jm6mH
J=~~{Jm2FI46IHjmXJK'J=~~"JxJxJxJxJxJxJxJxJxJxJm461!!0d""J5YYm65IjmJ
WX5jmJ_<5~m{J=~~1U!!0d""J5YYm65IjmJ[~k3mkEjJ_<5~m!!0d1!!0d!!0dJ6mHJ
K'{JfJfJfJfJfJfJfJfJOEjJXm~I21U"FI33"k|mX"v"T"v"u"dR"[ZIX!!0d!!0dJ6
mHJfCeJXm~I2"C@CwCRC-CdC?CvCQ{C@CwCRC-CdC?CvCQU"FI33"v"[ZIX!!0d""J5
H2CQJxJm2FI46IHjmXJm46m6mHJH5CeJXm~I21!!0d!!0dJ6mHJOEjJXm~I2CeC@Cw{
U")43mXj"~54mHmm6"[ZIXI[jmXf!!0d""J5H2Ce!!A4U"""""""""""U"$Ej"6k4gj
"54jmXXEFj"I4i"!!A4!!A4ZZ!!0d""""J5H2C@!!A4J[jX~Y"CeC@CwJm~3m"CeC@C
wJ[jX~Y"JH5!!0d""Jm~3m!!0d""""J5H2C@!!A4CeJ[jX~YC@CwJm~3m"CeC@J[jX~
YCwJH5!!0d""JH5"JK'1!!0d!!0dJ6mHJm46m6mHCeJm46{1U"m46"kH"jm2jf"3k"m
46"Jm6mH"I46"#k$$~m"XmYI5454#"lE4G!!0d!!0dU"=~~"jZIjg3"~mHj"jk"6mH5
4m"IXm"jZm"3G5FFmX"Yk6E~m"I46"jZm"6m[k6mX!!0dU"Yk6E~mx""^Zmi"$kjZ"I
Xm"WX5jjm4"54jk"jZm"Fk3j54#"jk"$m"m2m[E6m6!!0dU"$i"jZm"Xm[m5|mXx""^
Zmi"IXm"[kYFXm33m6"I46"k$HE3[Ijm6f"$Ej"jZm!!0dU"k$HE3[Ij5k4"53"Yk3j
~i"lE3j"[kYFXm335k4,"E354#"[kYYI46"3iY$k~3!!0dU"~5Gm"Jf"HkX"~k4#mX"
[kYYI46"WkX63f"I46"E354#"$E5~jA54"Xm#53jmX3!!0dU"543jmI6"kH"I~~k[Ij
54#"Xm#53jmX3x""'kYm"kH"jZm"I$$Xm|5Ij5k43"I46!!0dU"jZm"[Zk5[m3"kH"X
m#53jmX"IXm"YmI4j"jk"$m"[k4HE354#"I46SkX"35~~ix!!0dU":~I54Ajm2j"|mX
35k43"kH"jZm"Yk6E~m3"IXm"#5|m4"ZmXmf"I3"Wm~~"I3!!0dU"I"#~k33IXi"kH"
jZm"k$HE3[Ij5k4x!!0dU!!0dU"NmXm"53"jZm"3G5FFmX"Yk6E~mx"")j"53"E3m6"
54"jZm"HkXY,!!0dU"JqEm3j5k4,!!0dU"I"3Fm[5I~"~54m"kH"jm2j!!0dU"I4ijZ
54#"jZIj"53"3G5FFm6"m4j5Xm~if!!0dU"E4j5~"I#I54"3mm54#!!0dU"I"3Fm[5I
~"~54m"kH"jm2j!!0dU!!0dU"J6mHJqEm3j5k4,{J$#XkEF!!0dU"""JIHjmX#XkEFJ
m46!!0dU"""JI~~kjZmX!!0dU"""J'G5FFmX1!!0dU!!0dU"J'G5FFmX"3jIXj3"jZm
"3G5FF54#"$i"XmI654#"jZm"6m~5Y5jmX"jm2j"I46!!0dU"6mH5454#"jZm"YI[Xk
"nJ'G5Fr54mg"jk"3G5F"I"~54mf"jm3j54#"HkX"jZm!!0dU"m46"jm2jx""^Zm"jm
3j"53"6k4m"$i"[k43jXE[j54#"I"[kYYI46"4IYm"HXkY!!0dU"jZm"3m4j54m~"jm
2j"I46"HXkY"mI[Z"~54mf"I46"[kYFIX54#"jZmY"BW5jZ!!0dU"J5H2yx!!0dU!!0d
U"{J[Ij[k6mnJ\\Mue@"U"kjZmX!!0dU"J#6mHJ'G5FFmXCe\\MC@\\M{U"XmI6"jZ
53"~54m"AP"Cec"4m2j"~54m"AP"C@!!0dU"U""6mH54m"3m4j54m~"YI[Xk,!!0dU"
""Jm2FI46IHjmXJ6mHJ[34IYmC@Jm46[34IYmJSJJSSJS{=+\\Mf8DJk|mX1U!!0dU"
U"6mH54m"YI[Xk"jk"XmI6"~54m"I46"[kYFIXm"5j"W5jZ"3m4j54m~,!!0dU"""J6
mHJ'G5Fr54mCCe\\M{Jm2FI46IHjmXU!!0dU"""""J5H2J[34IYmCCeJm2FI46IHjmX
Jm46[34IYmJ[34IYmC@Jm46[34IYmU!!0dU"""""""Jm2FI46IHjmX"J*m[k6m=43Wm
X"U"H5453Zm6"3G5FF54#!!0dU"""""Jm~3mU!!0dU"""""""Jm2FI46IHjmX"J'G5F
r54m"U"GmmF"3G5FF54#!!0dU"""""JH51U!!0dU"1!!0dU!!0dU"J*m[k6m=43WmX"
E4ZI3Zm3"jZm"I43WmX"jm2j"I46"WX5jm3"5j"jk"jZm!!0dU"3[Xmm4x"^Zm"E4FX
54jI$~m"[ZIXI[jmX3"XmFXm3m4jm6"I3"!!A4!!A4ZZ"IXm"~mHj!!0dU"I3"jZmi"
IXm"B5xmxf"Fk335$~i"E4FX54jI$~m!!A4y"Ok4jXk~AM"B!!A4!!A4.6y"W5~~!!0d
U"$XmIG"jZm"jm2j"54jk"~54m3"k4"jZm"3[Xmm4c"jZm"~54m$XmIG3"54"jZm
!!0dU"ZI3Zm6"jm2j"IXm"5#4kXm6x""JN'"53"3mj"jk"jZm"3mm6"|I~Em"$mHkXm
!!0dU"J*m[k6m=43WmX"53"54|kGm6x!!0dU!!0dU"J6mHJ*m[k6m=43WmX{U"B[kYFIX
m"H5X3j"FIXj"W5jZ"J(kXYNI3Zy!!0dU"""JONuJ$kj["U"H5X3j"[ZIXI[jmX"BF~
I54"jm2jy!!0dU"""J~kkF"U"k|mX"ZI3Zm6"[ZIXI[jmX3!!0dU"""""J~[[k6mJN'
uJON"U"YIF"[k654#"jk"F~I54"jm2j!!0dU"""""J5H4EYJON>JjkF[!!0dU""""""
"JI6|I4[m"JN'"JNM""U"I66"YE~j5F~5mX"jk"ZI3Z"|I~Emf""E354#xxx!!0dU""
"""""JI6|I4[mJON"e"U"jZ53"ZmXm"FXm|m4j3"JN'"HXkY"$m54#"jm3jm6"FXmYI
jEXm~i!!0dU"""""""J5H4EY"JN'PJjkF["JI6|I4[mJN'AJXI4#m"JH5"U"Yk6E~k"
IX5jZYmj5[!!0dU"""JXmFmIj!!0dU"U"*mH54m"m2[mFj5k43x""OkYFIXm"jZ53"F
IXj"W5jZ"Ja2[mFj!!0dU"""J[Ij[k6mnJ\\=u."U"nm3[IFmgf"J!!0dU"U"J[Ij[k
6mnJ\\tue"U"nkFm4gf"{"AA"E44m[m33IXi!!0dU"""J[Ij[k6mnJ\\Ou@"U"n[~k3
mgf"1!!0dU"""J[Ij[k6mnJ!!A4u?"""U"n3EFmX3[X5Fjgf"\"BHkX"Zm2"54FEjy
!!0dU"""J~[[k6mnJ"unJ!!20!!0dU"""J~[[k6mnJ!!20unJ!!0dU"U!!0dU"""Jm46
~54m[ZIXuAe"U"5#4kXm"~54m"$XmIG3"54"[k6m6"jm2j!!0dU"""J4mW~54m[ZIXu
nJ\\M!!0dU"""J~kWmX[I3mJ$#XkEFJ5YYm65IjmJWX5jmJN'J$#XkEF!!0dU"1!!0d
U!!0dU"s~k33IXi"kH"I$$Xm|5Ij5k43"I46"k$HE3[Ij5k43!!0dU"AAAAAAAAAAAA
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAA!!0dU!!0dU!!0dU"J~mj"""""""""""""""""
h!!0dU"JN'"""BJjk~mXI4[my"""Jf"""""BJZI3Z"|I~Emy""""""""Ce!!0dU"JNM
""""""""""""""""""Jx"""""BZI3Z"YE~j5F~5mXy""""C@!!0dU"JjkF["""BJYk4
jZy"""""J0"""""B~I3j"ZI3Zm6"[ZIX,"e@d"hy!!0dU"J$kj[""""""""""""""""
w-"""""BH5X3j"ZI3Zm6"[ZIX,"w-"Cy!!0dU"JXI4#m"""""""""""""JimIX""""B
JjkF[AJ$kj[/e,"Q@y!!0dU"JON"""""BJE[ZiFZy""""Jc"""""BI"[ZIXI[jmX"[k
6my!!0dU"J[Ij[k6m"""""""""""""J,!!0dU"J6mH"""""""""""""""""JC!!0dU"
Jm2FI46IHjmX"""""""""J\!!0dU"J[34IYm""""""""""""""J{!!0dU"Jm46[34IY
m"""""""""""J1!!0dU"J~[[k6m""""""""""""""J7!!0dU"JI6|I4[m""""""""""
"""JA!!0dU"J$#XkEF""""""""""""""JT!!0dU"J5H4EY"""""""""""""""J5H!!0d
U"Jm46~54m[ZIX"""""""""Jb!!0dU"J5H2"""""""""""""""J5H\\v!!0dU"J'G5
FFmX"""""""""""""JK!!0dU"J'G5Fr54m""""""""""""J&!!0dU"J*m[k6m=43WmX
""""""""J]!!0dU"J~kWmX[I3mJ$#XkEFJ5YYm65IjmJWX5jmJN'J$#XkEF""""""J8
!!0dU!!0dU"^Zm3m"I335#4Ym4j3"IXm"WX5jjm4"jk"jZm"kEjFEj"$i"JKX5jm_Ej
FEj"BI[jEI~~i!!0dU"$i"JK'9y"I46"jZm4"jZm"k$HE3[Ijm6"FXk[m3354#"[k6m
"53"WX5jjm4"B#5|m4!!0dU"I3"I"FIXIYmjmX"54356m"B"y"IHjmX"JKX5jm_EjFE
jyx"JKX5jm^ZmLm3j"53!!0dU"IEjkYIj5[I~~i"54|kGm6"jk"[kFi"jZm"}Em3j5k
4"I46"jZm4"jZm"I43WmXx!!0d!!0dJ6mHJKX5jm_EjFEjCeC@{J$m#54#XkEF!!0d"
"J[Ij[k6mnJ\ue@"J4mW~54m[ZIXuew!!0d""Jm6mHJJ{J4km2FI46JK'9{Ce1{C@11
JJ1!!0d!!0dJ6mHJK'9CeC@{J~mjJJJ3jX54#"U"|mX$Ij5Y53Zf"I|k56"3FI[m3"I
HjmX"[kYYI463!!0d""J5YYm65IjmJWX5jmJ_<5~m!!0d"""{JJJ~mjJJhJJJ~mjJJh
JJJCJJJ6mHJJJCJJJx{C@1JJhJJJfJJJjk~mXI4[mJJJfCe1U!!0d""JI~~kjZmXJ[I
j[k6mnBueJ[Ij[k6mnyu@!!0d""JIHjmXI335#4Ym4jJKX5jm^ZmLm3j!!0d""JjkG3
.u1!!0d!!0dJ6mHJKX5jm^ZmLm3j{J5YYm65IjmJWX5jmJ_<5~m{JjZmJjkG3.1!!0d
""J[Ij[k6mnBue@J[Ij[k6mnyue@!!0d""J5YYm65IjmJWX5jmJ_<5~m{J3jX54#JqE
m3j5k4,1U!!0d""J5YYm65IjmJWX5jmJ_<5~m{J'mF1U!!0d""JOkFiqEm3j5k4!!0d
""J5YYm65IjmJWX5jmJ_<5~m{J'mF1U!!0d""{J=~~=[jNm2J(kXYNI3ZJa2[mFjJN5
6m'k~Ej5k41!!0d""Jm46#XkEF1!!0d!!0d!!0dJKX5jm_EjFEj{JjZmJN'1{JjZmJN
MU!!0d1BhJ0JYk4jZhJcJE[ZiFZhJ,J[Ij[k6mhJ\Jm2FI46IHjmXhJ{J[34IYm{hJC
J26mHhhJ3jX54#!!0dJCJe{h\\=1JCJw{h\\O1JCJR{h\\I11hJ1Jm46[34IYmhJT{h
J7J~[[k6mJCJ8{J4mW~54m[ZIX!!20*!!0dJ~kWmX[I3mJTJ5YYm65IjmJWX5jmJfJT
1hJAJI6|I4[mJimIXQ@hJ5HJ5H4EYhJbJm46~54m[ZIX!!0dJ0!!20?aJCJ\\-eEm3\
\R5k\\dm,{Jc."J~kkFJ,Jc!!20OJAJce"J5HJc>@-d"JXmFmIjJb!!20*JK1{J,!!20
\end{comment}
\begin{lcode}
*!!20O!!0dJ#6mHJKCe\\MC@\\M{J\JCJ{C@J1JSJJSSJS{=+\\Mf8DJk|mX1JCJ&C
Ce\\M{J\J5H\\vJ{CCeJ\U!!0dJ1J{C@J1J\J]Jm~3mJ\J&JH51J&11JCJ]{Jcw-J~k
kFJ7JfJcJ5HJc>J0JAJfJxJAJceJ5HJfPJ0!!0dJAJfAJimIXJH5JXmFmIjJ,eg.J,w
!!20@J,wwg?J7!!20@.n!!20J7n!!20!!20@.JbAeJ81!!0dy!!0d!!0dJm46!!0d
!!0d!!03!!03!!A{end!!A}
\end{lcode}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%[Addendum 2: TeX encoder for my decoder. (mjd,18-Aug-1994)]
\subsection{Addendum 2}

  TeX encoder for my decoder. (mjd,18-Aug-1994)


\begin{lcode}
%    Source character set: 13,32-126 = 96
%
%    (Note exclusion of tab. Assumption: Text to be translated will
%    always be untabified first.)
%
%    Target character set: 33-126.
%
%    Carriage return (13) cannot be included in the target set because
%    of the constraint to have a maximum line length of 72 in the
%    encoded text. If 13 (carriage return) were included in the
%    encoding, then the end of the current line would only occur at
%    the next instance in the ciphered text of the character that
%    translates to 13. And depending on what that character is, who
%    knows how long the encoded line could be? Perhaps as long as the
%    entire text.
%
%    Space (32) are not included in the target set for a subtler
%    reason. If spaces in the encoded text happen to fall at the end
%    of a line, they will be dropped by TeX during the decoding
%    process, instead of decoded. So we either must exclude them from
%    the target set, or make sure that they never fall at the end of a
%    line.
%
%    By excluding space from the target set, we make it possible for
%    the decoder to use a space as its argument delimiter. If we have
%    only one space, at the end of the encoded text, it is not so hard
%    to ensure that it does not fall at the end of a line. But note
%    that the decoder must make sure to change the catcode of space to
%    something other than 10, so that it will not disappear if it
%    falls at the *beginning* of a line.

\def\colon{:}\def\arrow{->}%
\let\isx\message
%\def\isx#1{}

\iffalse
%    OK, here is how the encoding works. Start with  \mag = random (in
%    the target range 33-125), first encoding value. Handle two
%    special cases first: ^^M encodes to \mag, space encodes to \mag
%    +1. Then start normal encoding at \fam = 35 (char 35 = ! encodes
%    to \mag +2, and so forth). When \mag reaches 126, we wrap it
%    around to 33 (don't want to encode any character to space).
%    Finally, when \fam reaches 126, we must handle the last three
%    characters (126,33,34: ~!") as digraphs: encode them as ~x~y~z,
%    where xyz are obtained by continuing to increment \mag.

@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_  ! "#$%&'()*+,-./0123456789:;<=>?
             R                  S~S~TTUVWXYZ[\]^_`abcdefghijklmnop
@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|} ~
qrstuvwxyz{|}!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQ~R
\fi %       ^^^

\def\setup{%
  \def\notilde{}% later will be defined to include a tilde
  \def\encodeone{%
    \catcode\fam\active\lccode126\fam\lccode 48\mag
    \lowercase{\edef~{\notilde 0}%
\isx{[\string~\colon \notilde 0\space\number\fam\arrow\number\mag]}%
    }%
    \advance\mag7 \ifnum\mag>125\advance\mag-93 \fi
    \advance\fam1
  }%
  \def\do{\encodeone \csname do\ifnum\fam>125 stop\fi\endcsname
  }%
%    ASSUMPTION: \mag initialized before the call of \setup
%    Encode ^^M -> \mag
  \fam13 \encodeone
%    Encode space -> next \mag
  \fam32 \encodeone
%    Now encode the rest
  \fam35 \let\dostop\relax \do
%    Now \fam = 34, \mag = ?. We need to define encoding for
%    characters 34,33,126 ("!~) as ~z ~y ~x. But what are convenient
%    values for x y z? Why, just the next \mag's in sequence
  \edef\notilde{\string ~}
  \encodeone \fam33 \encodeone \encodeone
}

\def\outwrite{\immediate\write15{\outline}%
%    If a digraph occurred at the end of the line, carry over the
%    second character to the beginning of the next line.
  \expandafter\ifx\csname 73\endcsname\relax
  \else
    \expandafter\let\expandafter\1\csname 73\endcsname
    \expandafter\let\csname 73\endcsname\relax
    \charnum 1
  \fi
  \checkeof}
%    For fast looking on screen:
%\def\outwrite{\immediate\write16{\outline}\checkeof}

\begingroup
\let\0\catcode \0`\0 11 \0`\2 11 \0`\3 11 \0`\4 11 \0`\5 11
\0`\6 11 \0`\7 11 \0`\8 11 \0`\9 11 \0`\1 11
\gdef\outline{\1\2\3\4\5\6\7\8\9\10\11\12\13\14\15\16\17\18\19
  \20\21\22\23\24\25\26\27\28\29\30\31\32\33\34\35\36\37\38\39
  \40\41\42\43\44\45\46\47\48\49\50\51\52\53\54\55\56\57\58\59
  \60\61\62\63\64\65\66\67\68\69\70\71\72}
\endgroup

\newcount\charnum

\def\checkeof{\futurelet\next\encodemore}

\def\tildecheck#1#2{\if \string~#1%
  \expandafter\def\csname\number\charnum\endcsname{#1}%
  \advance\charnum 1
  \expandafter\def\csname\number\charnum\endcsname{#2}%
\fi}

\def\encodemore{\ifx\next\EOF
    \let\next\outwrite \let\checkeof\relax
\global\tracingcommands2\global\tracingmacros2\global\tracingonline0
%    At end of file, assume that there was a ^^M at the end,
%    translated to the digraph ~|. Remove it, to reduce the number of
%    blank lines that will be produced on screen during decoding.
%    BUT, if \charnum = 72, leave the ^^M there to avoid having the
%    space at the end of the line.
    \ifnum\charnum<72
      \expandafter\def\csname\number\charnum\endcsname{ }%
    \else
      \def\1{ }%
    \fi
  \else
    \advance\charnum 1
    \ifnum\charnum>72
      \charnum 0 \let\next\outwrite
    \else
      \let\next\getnextchar
    \fi
  \fi
  \next}

\def\getnextchar#1{%
  \edef\0{#1}%
  \expandafter\let\csname\number\charnum\endcsname\0\relax
  \expandafter\tildecheck\0\relax\relax
  \checkeof}%

%    For this we need just a unique no-op value for \ifx comparison.
\def\EOF{\relax\relax}

\def\writefile#1{\expandafter\checkeof\input#1 \EOF}%

\begingroup
%    Define \0 to read in the text for \writepreamble.
\def\0#1XXX#2^^JZZZ^^J{\endgroup
  \def\writepreamble##1{\begingroup
%    Convert ##1 into a hex number.
    \newlinechar=10 \chardef\0=##1\def\1####1"{"}%
    \immediate\write15{#1\expandafter\1\meaning\0#2}\endgroup}}%
%    Now change all special catcodes to 12. We don't use \dospecials
%    because we want to do backslash last, in conjunction with
%    \afterassignment.
\catcode`\{=12 \catcode`\}=12 \catcode`\#=12
\catcode`\~=12 \catcode`\@=12 \catcode`\$=12
\catcode`\^=12 \catcode`\&=12 \catcode`\_=12 \catcode`\|=12
%    The following line will turn off the last two remaining special
%    characters % and \, set end-of-line character to ^^J (for later
%    use in the \write), and then call \0. ^^M still has category 5 at
%    this point and the new value of \endlinechar won't get applied
%    until the *next* line is read, so the catcode assignment for \
%    will get terminated properly by the space from ^^M, thus \0 will
%    get called before TeX attempts to read the % at the beginning of
%    the subsequent line.
\catcode`\%=12 \endlinechar=10 \afterassignment\0 \catcode`\\=12
%%%% Self-decoding answer: run the following text through plain TeX %%%%
\let\+\let\+\a\advance\+\c\catcode\+\d\def\+\f\fam\+\m\mag\+\u\uccode \m
13\c\m9\+\p\uppercase\d\i{\a\f7 \ifnum\f>125 \a\f-93 \fi}\d~{\u\f\m \c\m
12 \a\m1 \i \ifnum\m>125 \+~\1\fi~}\d\0#1{\ifnum`#1>"D \if#1 !\else "\fi
\else\string~\fi}\u`9"20\p{\d\1#19}{\newlinechar13\d\3{\immediate\write1
6}\+~\0\p{\3{}\3{#1}\batchmode\end}}\fXXX\u\f\m\i\m32\u\f\m\c\m12\i\m35~
ZZZ

\def\encodefile#1{%
  \immediate\openout15=encode.out \relax
  \begingroup
%    Get a random number from \time, normalize it to fall in the range
%    33--125. First set \mag = \time mod 93, then add 33 to make it
%    fall in the proper range.
  \fam\time \mag\time \divide\fam93 \multiply\fam 93 \advance\mag-\fam
  \advance\mag 33
  \message{======= Code shift: time \number\time\space -->
    mag \number\mag\space ============================}%
  \writepreamble{\number\mag}%
%    \setup uses \mag.
  \setup \charnum=0
  \immediate\write16{Starting to create file encode.out . . .}%
  \writefile{#1}%
  \endgroup
  \immediate\closeout15 \relax
  \immediate\write16{The encoded output is in the file encode.out.}%
}

\immediate\write16{Enter the name of the file you want to encode:}
{\catcode\endlinechar=9 \global\read-1 to\filnam}
\encodefile{\filnam}

\end
\end{lcode}
%$
%%\endinput


\chapter{Defining new control sequences}

\section{Exercise}

%%\input{ex012}
% ex012.tex
\begin{comment}
Date: 24 Sep 1993 16:11:36 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #12
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
========================================================================
*** Exercise 12:
\end{comment}
\ed{\oposted{1993/09/24}. \arch{exercise.012}.}

How many commands are there in plain TeX that can be used to define a
new (i.e., previously undefined) control sequence?

\begin{comment}
========================================================================

E-mail answers to my address, below. A summary will be posted circa
October 15, 1993.

Michael Downes ---------------------------------------------------------
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
\end{comment}
%$
%%\endinput

\section{Answers}

%%\input{ans012}
% ans012.tex
\begin{comment}
[The addendum was not included in the original post but added in my
archives later ---mjd]

Date: 25 Oct 1993 16:36:43 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #12, answer
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

\ed{\oposted{1993/10/25}. \arch{answer.012}.}

%Exercise 12 asked `How many commands are there in plain TeX that can
%be used to define a new (i.e., previously undefined) control
%sequence?'. 
This exercise has latent ambiguities. The parenthetical
remark `(i.e., previously undefined)' was intended as a hint towards
the most comprehensive possible answer.

There are three main criteria that could be used for `new' status of a
control sequence:
\begin{enumerate}
\item  If executed, the control sequence causes an `\texttt{Undefined control
sequence}' error.

\item The control sequence is \piif{ifx}-equivalent to \cmd{\relax} when constructed
with \cmd{\csname} \texttt{\ldots} \cmd{\endcsname}. This is the basis of the LaTeX
\cmd{\@ifundefined} test.

\item The control sequence has not yet been entered into the hash table.
\end{enumerate}

Criterion (3) doesn't work for one-character control sequences (\cmd{\a},
\cmd{\0}, \cmd{\:}) since they have space reserved for them separate from the
hash table whether or not they are defined in any sense.

Criterion (2) obviously gives a spurious true result if applied to
\cmd{\relax} or to something like LaTeX's \cmd{\protect} command that spends much
of its time being equivalent to \cmd{\relax}.

Criterion (1) therefore seems best. Notice that control sequences can
enter into the hash table without becoming defined anywhere along the
way, so a control sequence can be `old' by criterion (3) but still
new by criterion (1). In all of the following examples the control
sequence \cmd{\foo} will get added to the hash table but remain undefined.
\begin{lcode}
  \def\x{\foo}
  \toks0{\foo}
  \string\foo
  \noexpand\foo
  \gobble\foo (assuming \def\gobble#1{})
  \uppercase{\iffalse\foo\fi}
  \show\foo
  \meaning\foo
\end{lcode}
Two notable cases where tokenization, but not hash-table-ization, of
\cmd{\foo} occurs are in an \piif{ifx} comparison or on the false branch of an 
\piif{if}:
\begin{lcode}
  \ifx\foo\something...
  \iffalse\foo\fi
\end{lcode}
(\emph{TeXbook}, Appendix D, p384).

The straightforward answer to Exercise 12 is to count up the various
kinds of def'ing and let'ing functions (table~\ref{tab:deflet}):
\begin{comment}
\begin{lcode}
Primitive:          Nonprimitive:

\def                \newcount
\edef               \newdimen
\gdef               \newskip
\xdef               \newmuskip
\let                \newfam
\futurelet          \newwrite
\chardef            \newread
\mathchardef        \newbox
\countdef           \newtoks
\dimendef           \newinsert
\skipdef            \newlanguage
\muskipdef          \newif
\toksdef            \newhelp
\font               
\read
\csname
\end{lcode}
\end{comment}
\begin{table}
\centering
\caption{The def'ing and let'ing functions}\label{tab:deflet}
\begin{tabular}{ll} \toprule
Primitive          &  Nonprimitive       \\ \midrule
\cmd{\def}         &  \cmd{\newcount}    \\
\cmd{\edef}        &  \cmd{\newdimen}    \\
\cmd{\gdef}        &  \cmd{\newskip}     \\
\cmd{\xdef}        &  \cmd{\newmuskip}   \\
\cmd{\let}         &  \cmd{\newfam}      \\
\cmd{\futurelet}   &  \cmd{\newwrite}    \\
\cmd{\chardef}     &  \cmd{\newread}     \\
\cmd{\mathchardef} &  \cmd{\newbox}      \\
\cmd{\countdef}    &  \cmd{\newtoks}     \\
\cmd{\dimendef}    &  \cmd{\newinsert}   \\
\cmd{\skipdef}     &  \cmd{\newlanguage} \\
\cmd{\muskipdef}   &  \cmd{\newif}       \\
\cmd{\toksdef}     &  \cmd{\newhelp}     \\
\cmd{\font}        &  \\             
\cmd{\read}        &  \\
\cmd{\csname}      &  \\
  \bottomrule
\end{tabular}
\end{table}

The reason for including \cmd{\csname}? After 
\begin{lcode}
    \csname foobar\endcsname
\end{lcode}
\cmd{\foobar} is no longer undefined; the change in its status is
indistinguishable from the change effected by the statement
\verb?\let\foobar\relax?. \cmd{\endcsname} is not counted separately because
\cmd{\csname} and \cmd{\endcsname} can only be used together.

So: 16 primitive, 13 non-primitive make 29 total. But to those should
be added two more, since the statement of the Exercise didn't exclude
`private' macros: (i) the internal function \cmd{\alloc@} of plain.tex
that is shared by all the \cmd{\newxxx} macros (except for \cmd{\newif} and
\cmd{\newhelp}), and (ii) the internal function \cmd{\@if} used by \cmd{\newif}.

That brings the total to 31.

Beyond that there can be added another, less obvious, class of
commands, if we paraphrase the exercise as follows:
\begin{quote}
  Find all commands such that executing command \cmd{\xxx}, with its normal
  arguments (if any), causes at least one control sequence to pass
  from undefined status to defined status, where undefined status
  means that executing the control sequence would generate the error
  `Undefined control sequence'.
\end{quote}
For example, the first use of \cmd{\loop} causes \cmd{\body} and \cmd{\next} to become
defined. As it turns out, there are many of these in plain TeX 
(table~\ref{tab:user} and~\ref{tab:internal} as well as \verb?'? or \cmd{\rq}
in math mode only).

\begin{comment}
User functions:
\begin{lcode}
\loop, \t, \smash, \vfootnote, \settabs, \phantom,
\vphantom, \hphantom, \footnote, \multispan, \longleftarrow,
\longrightarrow, \mathstrut, \longmapsto, \matrix, \pmatrix;
\end{lcode}
\verb?'? or \cmd{\rq} (math mode only)
\end{comment}

\begin{figure}
\freetabcaption{User functions}\label{tab:user}
\autorows{c}{4}{l}{%
\cmd{\footnote}, 
\cmd{\hphantom}, 
\cmd{\longleftarrow},
\cmd{\longmapsto}, 
\cmd{\longrightarrow}, 
\cmd{\loop}, 
\cmd{\mathstrut}, 
\cmd{\matrix}, 
\cmd{\multispan}, 
\cmd{\phantom},
\cmd{\pmatrix},
\cmd{\settabs}, 
\cmd{\smash}, 
\cmd{\t}, 
\cmd{\vfootnote}, 
\cmd{\vphantom} 
}
\end{figure}


\begin{comment}
Internal functions: 
\begin{lcode}
\iterate, \relbar, \sett@b, \s@tt@b, \prim@s,
\ph@nt, \fo@t, \f@@t, \pr@m@s, \pr@@@s, \s@tcols
\end{lcode}
\end{comment}

\begin{figure}
\freetabcaption{Internal functions}\label{tab:internal}
\autorows{c}{6}{l}{%
\cmd{\f@@t},
\cmd{\fo@t},
\cmd{\iterate},
\cmd{\ph@nt},
\cmd{\pr@@@s},
\cmd{\pr@m@s},
\cmd{\prim@s},
\cmd{\relbar},
\cmd{\s@tcols},
\cmd{\s@tt@b},
\cmd{\sett@b}
}
\end{figure}

Adding these 18 user functions and 11 internal functions to the
previously cited 31 gives a total of 60 functions available in
\pfile{plain.tex} that satisfy a strict interpretation of the exercise
statement.

Credit for the best answer goes to Dan Luecking\index{Luecking, Dan}, 
who found 29 of the
primary 31, and did not miss the other two (\cmd{\csname}, \cmd{\@if}) by
overlooking them but by considering them and believing they didn't
satisfy the requirements.

My own score in that part was 28: I overlooked \cmd{\read}, \cmd{\alloc@}, and
\cmd{\@if} until Luecking and Peter Schmitt\index{Schmitt, Peter} 
brought them to my notice.

Ian Collier\index{Collier, Ian} also submitted a good answer, including 
identification of
the secondary class of functions that define scratch macros as a side
effect.

%%========================================================================

Notes:
\begin{itemize}
\item \cmd{\iterate}, \cmd{\settabs}, \cmd{\sett@b}, \cmd{\s@tt@b}, 
      \cmd{\t}, \cmd{\prim@s}, \cmd{\ph@nt}, \cmd{\smash},
      \cmd{\vfootnote}, \cmd{\fo@t}, \cmd{\f@@t} all define \cmd{\next}.
\item \cmd{\loop} defines \cmd{\body}.
\item \cmd{\pr@m@s} defines \cmd{\nxt}.
\item \cmd{\prim@s} is called by active \verb?'? (mathcode \verb?"8000?) 
      and by \cmd{\pr@@@s}.
\item \cmd{\iterate} is called by \cmd{\loop}.
\item \cmd{\sett@b} is called by \cmd{\settabs}.
\item \cmd{\s@tt@b} is \emph{conditionally} called by \cmd{\sett@b}.
\item \cmd{\smash} is called by \cmd{\relbar}.
\item \cmd{\ph@nt} is called by \cmd{\phantom}, \cmd{\vphantom}, and 
      \cmd{\hphantom}.
\item \cmd{\vfootnote} is called by \cmd{\footnote}.
\item \cmd{\fo@t} is called by \cmd{\vfootnote}.
\item \cmd{\f@@t} is \emph{conditionally} called by \cmd{\fo@t}.
\item Active \verb?'? is produced by \cmd{\rq} if used in math mode.
\item \cmd{\pr@@@s} is called by \cmd{\pr@m@s}.
\item \cmd{\loop} is called by \cmd{\multispan} and \cmd{\s@tcols}.
\item \cmd{\relbar} is called by \cmd{\longleftarrow} and \cmd{\longrightarrow}.
\item \cmd{\vphantom} is called by \cmd{\mathstrut}.

\item \cmd{\pr@m@s} is called by \cmd{\prim@s}.
\item \cmd{\s@tcols} is *conditionally* called by \cmd{\sett@b}.
\item \cmd{\longrightarrow} is called by \cmd{\longmapsto}.
\item \cmd{\mathstrut} is called by \cmd{\matrix}.

\item \cmd{\matrix} is called by \cmd{\pmatrix}.

\item \cmd{\prim@s} won't necessarily define \cmd{\next} because it does 
a \cmd{\futurelet}
which will leave \cmd{\next} undefined if the next thing happens to be an
undefined control sequence (rather unlikely, however).

\item \cmd{\vfootnote} and \cmd{\settabs} also do a \cmd{\futurelet} but it is followed by
another macro that ensures that \cmd{\next} does not end up undefined.
\end{itemize}

\begin{comment}
Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\end{comment}
%$

\section{Addendum}
\enlargethispage{3\onelineskip}
\begin{comment}
Addendum: From comp.text.tex
===========================================================================
Archive-Date: Wed, 29 Sep 1993 13:21:40 CST
From: cet1@cus.cam.ac.uk (Chris Thompson)
Subject: Re: Managing Large LaTeX Files. How ??
Date: Wed, 29 Sep 1993 16:36:23 GMT
To: tex-news@SHSU.EDU
\end{comment}

From \texttt{comp.text.tex}
\begin{lcode}
From: cet1@cus.cam.ac.uk (Chris Thompson)
Subject: Re: Managing Large LaTeX Files. How ??
Date: Wed, 29 Sep 1993 16:36:23 GMT
To: tex-news@SHSU.EDU
In article <93265.121206SPIT@EVALUN11.BITNET>, Werenfried Spit <SPIT@EVALUN11.BITNET>
writes:
|> In article <1993Sep20.130331.16568@vax.oxford.ac.uk>, kaye@vax.oxford.ac.uk
|> (Richard Kaye) says:
|> >Has anyone else had save stack overflow when LaTeX read the .aux files?
|> >
|> >[Will a TeX guru please explain it to me?  I thought \global\def's could not
|> >cause save stack overflow until I found this problem.  If it's a general
|> >problem, it seems a bit silly that LaTeX should try to input so much
|> >information in this way.]
|> >
|> >I fixed it so that the data was read {\it outside} the group (as part of one
|> 
|> Could someone explain it to me too? I'm even more puzzled after I tried
|> out Richards solution and played a bit with it. When you put in
|> your input file directly after the \documentstyle command the line
|>  \input \jobname.aux
|> LaTeX reads the aux file without its memory getting overflowed; then
|> at \begin{document} it reads the aux file again (as expected), but
|> the memory doesn't overflow this time either. (If you leave out the
|> \input \jobname.aux LaTeX only reads the aux file during \begin{document}
|> and then chokes on an exceedence of the save size.)
\end{lcode}
[Chris Thompson] This was a hard one to track down. I could claim that it was all my fault...

The entries on the save stack are not the result of the 
\cmd{\global}\cmd{\@namedef},
which as suggested above never needs to use such a thing. They come from
the earlier \cmd{\@ifundefined} call in \cmd{\newlabel}.

Change \#337 in \pfile{tex82.bug} numbering, applied in TeX 2.9, changed the implicit
setting of an undefined control sequence referenced via \cmd{\csname}...\cmd{\endcsname}
to \cmd{\relax} (\emph{TeXbook}, page 213) from being (sort of) global to being local to
the current group. Don made this change as a direct result of my posting to
TeXhax (year 1987, digest 103) pointing out that the TeXbook didn't correctly
describe what happened.

The change was a potent source of new bugs, because TeX was not originally
designed to cope with token expansion have side-effects of modifying the
save stack (see in particular change \#371 in tex82.bug). I have more than
once wondered whether I should have kept quiet about the whole business\ldots

In an ideal world, the problem wouldn't arise because the implicit setting
to \cmd{\relax} wouldn't occur at all (IMNSHO). But everything (especially LaTeX)
relies on it now, so it's (far) too late to change it. Something to be got
right in the next incarnation.

\begin{lcode}
Chris Thompson
Cambridge University Computing Service
\end{lcode}

%%\endinput


\chapter{\cs{endlinechar} and \cs{par}}

\section{Exercise (fast)}

%%\input{ex013}
% ex013.tex
\begin{comment}
Date: 13 Oct 1993 12:31:56 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #13
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

\ed{\oposted{1993/10/13}.\arch{exercise.013}.}

\begin{lcode}
%%%% Three lines of overhead for the self-decoding answer; see below %%%
\let\+\let\+\a\advance\+\c\catcode\+\d\def\+\f\fam\+\m\mag\f"20\d~{\c\f9
\a\f1 \ifnum\f>125\f002\d~{\a\f-1 \ifnum\f<1\egroup\fi}\fi~}\c`\^^M="9{~
\end{lcode}

%%========================================================================
%%*** Exercise 13 (fast):

(a) If \cmd{\endlinechar} does not have category 5 do you still get a \piif{par}
from a blank line?

(b) If \cmd{\endlinechar}=-1 do you still get a \piif{par} from a blank line?

\begin{comment}
========================================================================

Michael Downes =========================================================
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
\end{comment}
%$

Self-decoding answer given below. To see the answer, run this post
(sans mail/newsgroup header) through plain TeX.
\begin{lcode}
\d~{\u\f\m\c\m12 \a\m1\a\f1 \ifnum\f>125\f33 \fi\ifnum\m>125\+~\1\fi~}\+
\u\uccode\+\p\uppercase\d\0#1{\ifnum`#1>"D \if#1 !\else"\fi\else\string~
\end{lcode}
\ed{There are sixteen lines like this, all of which are in the archived
version if you need them. The last line is:}
\begin{comment}
\fi}\u`9"20\p{\d\1#19}{\newlinechar13 \d\3{\immediate\write16}\+~\0\p{\3
{}\3{#1}\batchmode\end}}\f"6C\m"0D\u\f\m\a\f"1\m32\u\f\m\c\m12\a\f1\m35~
/\aeS`amb]m/`]c\RmbVSm0S\Rmn|!(llsOtm<]ymsPtm<]ymm7\m]bVS`me]`RawmOmPZO\
YmZW\SmeWZZm^`]RcQSmOmJ^O`mWTlO\Rm]\ZgmWTmS\RZW\SmQVO`OQbS`amO`Sm^`SaS\b
mO\RmVOdSmQObQ]RSm#ym7bmWalW\bS`SabW\Umb]m\]bSmbVObmbe]mQ]\aSQcbWdSmS\RZ
W\SmQVO`OQbS`amO`Sm\]blb`O\aZObSRmaW[^Zgmb]mJ^O`wmPcbmb]m*a^OQS,J^O`ymms
BVSma^OQSmeWZZlRWaO^^SO`mW\ma][SmQW`Qc[abO\QSawmSyUywmOTbS`mOmQ]\b`]Zme]
`RwmOQQ]`RW\Ulb]mBSFram\]`[OZmaQO\\W\Um`cZSaytmBVWamWambVSm`SOa]\ms]`mOb
mZSOabm]\Sl`SOa]\tmbVObmOmJ^O`m]^S`ObW]\m[cabm^S`T]`[mO\mW[^ZWQWbmJc\aYW
^l]^S`ObW]\ymBVS`SmeOamOZa]mOm`SQS\bm^]abmb]mQ][^ybSfbybSfmPgm2]\OZRl/`a
S\SOcmb]m^]W\bm]cbmbVSm^`]PZS[meWbVma][S]\SramRSZW[WbSRxO`Uc[S\bl[OQ`]mR
STW\WbW]\(llmmJRSTJa][SbVW\Un|yJ^O`i*R]ma][SbVW\UmeWbVmn|,kllBVSmRSZW[Wb
S`mab`W\Um~nyJ^O`~nmRWRm\]bm[ObQVmbVSmOQbcOZmbSfbllmmyyyma][SmbSfbylmm*P
\end{comment}
\begin{lcode}
ZO\YmZW\S,llPSQOcaSm]TmbVSma^OQSmb]YS\mT]ZZ]eW\UmbVSm^S`W]Ry mbSfbylmm*P
\end{lcode}

%%\endinput

\section{Answers}

%%\input{ans013}
% ans013.tex
\ed{\arch{answer.013}.}

[This was included as a self-decoding answer in the posting of Exercise
\#13 which is archived as \pfile{exercise.013}.]


Answers to Around the Bend \#13:

(a) No. (b) No.  In other words, a blank line will produce a \piif{par} if
and only if endline characters are present and have catcode 5. It is
interesting to note that two consecutive endline characters are not
translated simply to \piif{par}, but to \meta{space}\piif{par}.  (The space will
disappear in some circumstances, e.g., after a control word, according
to TeX's normal scanning rules.) This is the reason (or at least one
reason) that a \piif{par} operation must perform an implicit \cmd{\unskip}
operation. There was also a recent post to \pfile{comp.text.tex} by Donald
Arseneau\index{Arseneau, Donald} to point out the problem with someone's 
delimited-argument macro definition:
\begin{lcode}
  \def\something#1.\par{<do something with #1>}

The delimiter string ".\par" did not match the actual text

  ... some text.
  <blank line>

because of the space token following the period..
\end{lcode}

%%\endinput


\chapter{TeX's stomach}

\section{Exercise}

%%\input{ex014}
% ex014.tex
\begin{comment}
Date: 26 Oct 1993 09:29:08 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #14
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

\ed{\oposted{1993/10/26}. \arch{exercise.014}.}

\begin{lcode}
%%%%% Two lines of overhead for the self-decoding answer; see below %%%%
\let\+\let\+\a\advance\+\c\catcode\+\d\def\+\f\fam\+\m\mag\c13 9{\c32'16
\end{lcode}

%% =======================================================================
\begin{quote}
 *** Exercise 14 [proposed by Jonathan Fine]:

 Which character code/category code pairs can actually reach TeX's
 `stomach'?
\end{quote}

%% =======================================================================

 This is a refinement of The \emph{TeXbook}'s Exercise 7.3. You need to be a
 little careful about your answer. I didn't get it right on my first
 try \ldots

 To make the notion of `reaching TeX's stomach' more precise: A token
 is said to `reach TeX's stomach' if it produces a token report when
 \cmd{\tracingcommands} = 1. And a `token report' is a phrase in braces,
 e.g.,
\begin{lcode}
   {the letter A}
 \end{lcode}
 as produced by TeX in the log file when tracing commands.

\begin{comment}
 Michael Downes ========================================================
 mjd@math.ams.org          ASCII 32--55,56--126: !"#$%&'()*+,-./01234567
 89:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
\end{comment}
%$

 Self-decoding answer given below. To see the answer, run this post
 (sans mail/newsgroup header) through plain TeX.

\begin{lcode}
}\d~{\u\f\m\c\m12\a\m1\a\f1 \ifnum\f>125\f33 \fi\ifnum\m>125\+~\1\fi~}\+
\u\uccode\+\p\uppercase\d\0#1{\ifnum`#1>"D \if#1 !\else"\fi\else\string~
\end{lcode}
\ed{In the archived form there are 20 lines like this, the last being:}
\begin{comment}
\fi}\u`9"20\p{\d\1#19}{\newlinechar13 \d\3{\immediate\write16}\+~\0\p{\3
{}\3{#1}\batchmode\end}}\f"39\m"0D\u\f\m\a\f"1\m32\u\f\m\c\m12\a\f1\m35~
Y).2}-:/*:Y-*0)|:/#}:Z})|:;ILR99::[y/{*|}:[#y-:[*|}.::::[y/{*|}:[#y-:[*|
}.9::EEEEEEE:EEEEEEEEEE::::EEEEEEE:EEEEEEEEEE9:::::I:::::HEEJMM:::::::::
::IH:::IEEJMM9:::::J:::::HEEJMM9:::::K:::::HEEJMM:::::::::::II:::HEEJMM9
:::::L:::::HEEJMM:::::::::::IJ:::HEEJMM9::::::::::::::::::::::::::::IK::
:HEEJMM9:::::N:::::HEEJMM9:::::O:::::HEEJMM9:::::P:::::HEEJMM99[y/}"*-4:
IH:$.:/#}:}3{}+/$*)y':{y.}F:[y/{*|}EIH:{#y-y{/}-.:2$/#:{#y-y{/}-9{*|}:TV
:KJ:{y):*)'4:z}:+-*|0{}|:z4:t0++}-{y.}Gt'*2}-{y.}:/-${&.:@l}pz**&D9Y++})
|$3:\AF:k*:/#}:+y$-:{#y-y{/}-:HD:{y/{*|}:IH:$.:)*/:+*..$z'}R:t0++}-{y.}9
y)|:t'*2}-{y.}:{y))*/:+-*|0{}:y:{#y-y{/}-:H:!-*(:y:)*)EH:{#y-y{/}-F99Y{/
$1}:{#y-y{/}-.:2$'':/}./:/-0}:!*-:{y/}"*-4:IH:2$/#:t$!{y/:$!:/#}4:y-}9t'
}/:},0y':/*:y:.+y{}:/*&})F:Z0/:$!:/#}:~9:{#y-y{/}-:@.y4A:#y.:z}}):.*9|}!
$)}|D:$/:2$'':)*/:(y/{#:y:.+y{}:$):/#}:|}'$($/}-:/}3/:*!:y:(y{-*:2$/#9|}
'$($/}|:y-"0(})/.F:Y)|:y{{*-|$)":/*:t/-y{$)"{*((y)|.:/#}:(}y)$)":*!:y)9y
{/$1}:/$'|}:/#y/:#y.:z}}):t'}/:},0y':/*:y:.+y{}:$.:~;z'y)&:.+y{}::~;D92#
\end{comment}
\begin{lcode}
}-}y.:/#}:(}y)$)":*!:y:{y/}"*-4EIH:/$'|}:$.:~;z'y)&:.+y{}:~9~;F ::~;D92#
\end{lcode}

%%\endinput


\section{Answers}

%%\input{ans014}
% ans014.tex

\ed{\arch{answer.014}.}

[This was included as a self-decoding answer in the posting of Exercise
\#14, which is archived as \pfile{exercise.014}.]

\begin{lcode}
Answer to Around the Bend #14:

  Catcode Char Codes    Catcode Char Codes
  ------- ----------    ------- ----------
     1     0--255           10   1--255
     2     0--255                       
     3     0--255           11   0--255
     4     0--255           12   0--255
                            13   0--255
     6     0--255
     7     0--255
     8     0--255
\end{lcode}
Category 10 is the exceptional case. Catcode-10 characters with character
code $<>$ 32 can only be produced by \cmd{\uppercase}/\cmd{\lowercase} tricks 
(\emph{TeXbook}, Appendix D). So the pair character 0, catcode 10 is not 
possible: \cmd{\uppercase}
and \cmd{\lowercase} cannot produce a character 0 from a non-0 character.

Active characters will test true for category 10 with \piif{ifcat} if they are
\cmd{\let} equal to a space token. But if the \verb?~? character (say) has been so
defined, it will not match a space in the delimiter text of a macro with
delimited arguments. And according to \cmd{\tracingcommands} the meaning of an
active tilde that has been \cmd{\let} equal to a space is 
\verb?`blank space  '?
whereas the meaning of a category-10 tilde is \verb?`blank space ~'?.

%%\endinput


\chapter{Space removal}

\section{Exercise}

%%\input{ex015}
% ex015.tex
\begin{comment}
Date: 05 Nov 1993 16:34:28 -0500 (EST)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #15
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

\ed{\oposted{1993/11/05}. \arch{exercise.015}.}

(a) Write a macro \cmd{\trimspace} that takes another macro as its argument and
removes a trailing space from the replacement text of the macro, if one
is present, and otherwise leaves it unchanged.

(b) Write a macro \cmd{\trimspaces} that removes a leading space, if
present, and then calls \cmd{\trimspace} to remove a trailing space.

%%========================================================================

Motivation: If a user inadvertently includes an extra space
in a text argument, such as a section heading:
\begin{lcode}
  \section{Title of the section }
\end{lcode}
then you must usually take care to remove the space when typesetting
the text. The simple way is to perform an \cmd{\unskip} at the end (if the
text is immediately followed by \piif{par}, the \cmd{\unskip} operation is
built-in) and an \cmd{\ignorespaces} at the beginning, but various
complications can arise, so it would be preferable to be able to apply
a \cmd{\trimspaces} function when an argument is first read, and then have
the information in proper form for all subsequent uses.

\begin{comment}
Send answers to the address below. A summary will be posted
November 23, 1993 or thereabouts.

Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
\end{comment}
%$
%%\endinput


\section{Answers}

%%\input{ans015}
% ans015.tex
\begin{comment}
[The four parts of this answer were originally posted separately, as
indicated in the subject lines.]

Date: 16 Dec 1993 16:34:45 -0500 (EST)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #15, answers
To: info-tex@shsu.edu
\end{comment}

\ed{\oposted{1993/12/16}. \arch{answer.015}.}

Exercise 15 asked for a function \cmd{\trimspace} to trim a trailing space
from the replacement text of a macro, and a function \cmd{\trimspaces} to
trim both a leading and a trailing space. At the time of posting the
exercise I had no prepared solution; as luck would have it the problem
was rife with latent complications (including some hard questions
about limiting the domain of application), which propagated an
unusually diverse crop of approaches among the submitted solutions,
and which made the task of preparing a good summary extraordinarily
difficult. Even after breaking down the `summary' into two or three
pieces, to avoid a too formidably large monolith of a posting, I'll
have to leave out some material that I would otherwise have included.

I'd say Donald Arseneau\index{Arseneau, Donald} deserves credit for 
the best analysis,
including an accurate survey of brace-stripping problems. Nearly
everyone, including myself, had missed a lurking flaw of that kind in
the first submitted version of their solution. Another good idea of
Donald's that caught my fancy was to use TeX's built-in scanning
procedures for \meta{optional space} to strip the leading space in
\cmd{\trimspaces}. I managed to work that into my own best solution, much to
my satisfaction.

Peter Schmitt\index{Schmitt, Peter} came up with perhaps the most 
aerodynamic solution, on his second go-round. A solution by 
Ian Collier\index{Collier, Ian} differed notably from
the others by using \cmd{\meaning} to look for a leading space. Another
submission, from 
Gary McGary\index{McGary, Greg}\index{McGary, Gary|see{McGary, Greg}} 
\ed{I think this is a typo for Greg McGary}, contained some 
original syntactic ideas,
and explored the more general problem of removing an arbitrary token
pattern at the end of a token list.

A careless, off-the-cuff remark of mine in the statement of Exercise
15 that after removing a leading space, \cmd{\trimspaces} should call
\cmd{\trimspace} to remove a trailing space, was probably a mistake. In most
cases, at least, \cmd{\trimspaces} can be more elegantly written by letting
the two different space-removal procedures share a few tokens at a
lower level.

From Donald's\index{Arseneau, Donald} analysis:
\begin{quote}
 When I first read the question, I thought `why isn't there an answer
 with the question, because that one is easy?'  As I started to type
 my answer `cold', I realized that what I had used previously to
 ignore leading spaces 
\begin{lcode}
   \def\something#1#2\weird{#1#2}
\end{lcode}
 had the bad
 side-effect of stripping braces if the parameter began with `\verb?{?'.
\end{quote}

I append below Peter Schmitt's\index{Schmitt, Peter} 
solution, more or less as he wrote it.
The commentary refers to earlier correspondence in a place or two but I
believe there is sufficient context to make everything intelligible.
Test \#5 in the test suite traps the insidious brace-stripping problem
that infested most of the solutions in their first incarnation.

\begin{comment}
More on Exercise 15 to follow, some time in the next few days.

Michael Downes, mjd@math.ams.org

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\end{comment}

\begin{solution}{Solution 1 (Peter Schmitt)}
%%>>Solution 1 (Peter Schmitt, a8131dal@awiuni11.edvz.univie.ac.at)

Since I wanted to stay with delimited arguments it was clear that one
has to add a token (or tokens) in order to hide braces, which finally
have to be removed again. First I came up with using \cmd{\empty}, as you
did, but then I switched to a not expandable token because this can
more efficiently be used as a parameter delimiter.

\cmd{\trimspaces} and \cmd{\trimspace} are just used to expand the argument and
add delimiting tokens in front and at the end of it, and set up the
delimiting tokens for \cmd{\Trimspace} and \cmd{\Trimspaces}, too.

As Donald does, I do not call \cmd{\trimspace} by \cmd{\trimspaces} but rather
\cmd{\Trimspace} by \cmd{\trimspaces}. It would be easy to offer \cmd{\TrimLeft}
\cmd{\TrimRight} and \cmd{\TrimBoth} and also \cmd{\TrimLeftS} \cmd{\TrimRightS} and
\cmd{\TrimBothS} which iterate in the (very unlikely!) case that there are
several consecutive space tokens.

\cmd{\Trimspaces} and \cmd{\Trimspace} remove leading, respectively trailing,
spaces of the argument, but they both leave the delimiting tokens in
place. These (and outside tokens) are removed by \cmd{\TrimSpace} in the
process of redefining the initial controlsequence.
\begin{lcode}
\catcode`\<=3 \catcode`\>=3

\def\trimspace  #1{\expandafter\expandafter\expandafter
                   \Trimspace\expandafter <#1> >\\#1}
\def\trimspaces #1{\expandafter\expandafter\expandafter
                   \Trimspaces\expandafter <#1>< <\\#1}

%% \Trimspaces  < text>< <\\                |< text>|  ==>
%%          ->  || + |text> + | <|
%%          =>  ||+| <|+|text>|          == | <text>|
%%
%% \Trimspaces  <text>< <\\                 |<text>|   ==>
%%          ->  |<text>| + || + ||
%%          =>  |<text>|+||+||           == |<text>|

%% \Trimspace   <text > >\\                 |<text >|  ==>
%%         ->   |<text| + | >|
%%         =>   |<text|+>\\              == |<text>\\|
%%
%% \Trimspace   <text> >\\                   |<text>|  ==>
%%         ->   |<text>| + ||
%%         =>   |<text>|+>\\              == |<text>>|

\def\Trimspaces #1< #2<#3\\{\Trimspace #1#3#2 >\\}
\def\Trimspace  #1 >#2\\{\TrimSpace #1>\\}
\def\TrimSpace  #1>#2\\#3{%
            \expandafter\expandafter\expandafter\expandafter\expandafter
            \def \expandafter\expandafter\expandafter #3\expandafter
            {\Remove#1}}
            \def\Remove#1{}

\catcode`\<12 \catcode`\>=12

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\def\Test#1{\def\test{#1}\immediate\write0{|\test|}%
            \trimspaces\test
            \immediate\write0{|\test|}%
           }
\let\trim\trimspace
\let\trim\trimspaces

%%%%%%%%%%%%%%%%%%%%%%%%%

\Test{}
\Test{ }
\Test{ a }
\Test{ {}{} }
\Test{{braces}}
\Test{ {braces} }
\Test{ { braces } }
\Test{no space and no space}
\Test{no space and a space: }
\Test{ :a space and no space}
\Test{ :a space and a space: }

\def\test{ \ifx/ }\trimspace\test\show\test
\def\test{ \ifx }\trimspaces\test\show\test

\end %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\end{lcode}

\end{solution}

%%\endinput
\begin{comment}
Date: 23 Dec 1993 16:21:21 -0500 (EST)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #15, answers, 2nd installment
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

Some exposition seems called for here in order to lay out various
considerations running through my mind and the minds of the other
solution-submitters.


\subsection{Trimming a trailing space}

There are two possible ways to remove a trailing space. The first one
is to step through the given text one token at a time, and construct a
new token list in parallel by adding the tokens one by one at the end.
If the next token is a space, delay adding it until the subsequent
token is checked, and if it turns out the text is exhausted, discard
the space instead of adding it. The hard part about this approach is
dealing with braces (character tokens with catcode 1 or 2) because a
lone brace cannot be passed as a macro argument. A recent posting by
\'Eamonn McManus to comp.text.tex on a different sort of problem
showed that the braces can indeed be dealt with, it's just not easy.

The second, simpler approach is to use TeX's scanning of delimited
macro arguments to scan for the ending space and discard it. If you
merely scan for a space token, however, you end up scanning through
the given text `word' by `word' (word = sequence of non-space
characters or brace-delimited groups) instead of token by token, which
is perhaps if anything even more awkward than the first method above,
since you still must deal with brace complications.

The key refinement, therefore, is to scan for a pair of tokens: a
space token and some well-chosen bizarre token that can't possibly
occur in the scanned text. If you put the bizarre token at the end of
the text, and if the text has a trailing space, then TeX's delimiter
matching will match at that point and not before, because the earlier
occurrences of space don't have the requisite other member of the pair.

Next consider the possibility that the trailing space is absent: TeX
will keep on scanning ahead for the pair \meta{space}\meta{bizarre} until either
it finds them or it decides to give up and signal a `Runaway
argument?' error. So you must add a stop pair to catch the runaway
argument possibility: a second instance of the bizarre token, preceded
by a space. If TeX doesn't find a match at the first bizarre token, it
will at the second one.

Now all that's left is to test somehow where the hit occurred in order
to fork properly. This can be done in various clever ways, as
exhibited in the solutions.

%%\endinput

\subsection{Trimming a leading space}

More analysis from Donald Arseneau:
\begin{quote}
 There are two safe, expandable ways to eat `one optional space':
 `\piif{ifnum}' using an ascii code (\texttt{`c}) as the second number, and
 `\piif{ifdim}' using a literal unit of measure like `pt'.  Oh, yes,
 it could also be done with parameter syntax too, but more on
 that later.
\end{quote}

%%\endinput

In other words, one way to remove a leading space would be
\begin{lcode}
  \expandafter\def\expandafter\foo\expandafter{\ifdim0pt=0pt\foo \fi}
\end{lcode}
The \cmd{\expandafter}'s would cause the \piif{ifdim} to be executed first.
Execution of the \piif{ifdim} will not terminate until the scanning of the
second `0pt' is finished; therefore TeX will start expanding \cmd{\foo} as
part of the scanning of the `0pt'. Then if a space is the first thing
inside the expansion of \cmd{\foo}, it will be removed by TeX as denoting
the end of the dimension. Otherwise the first non-space token will
terminate the dimension scanning and will be left in place (well, I am
glossing over the problem of an expandable token at the beginning of
\cmd{\foo}, which can be handled by further refinements).

Notice that as written the trailing \piif{fi} will be included in the
redefinition of \cmd{\foo}. No problem---just rewrite it with the \piif{fi} 
after the closing brace:
\begin{lcode}
  \expandafter\def\expandafter\foo\expandafter{\ifdim0pt=0pt\foo}\fi
\end{lcode}

[Now for a sharp little question: will that work with \cmd{\edef} instead of
\cmd{\def}?
\begin{lcode}
\edef\foo{\ifdim0pt=0pt\foo}\fi 
\end{lcode}
See if you can guess before
testing it.]

%%\endinput

%%\begin{verbatim}
Other ways of removing a leading space include using \cmd{\futurelet} to
look at the first token in the scanned text, or using TeX's argument
delimiter scanning to scan for a space. The latter method is perhaps
most straightforwardly done as a mirror-image of the method for
removing a trailing space: make the delimiter \meta{bizarre}\meta{space}, and
then call the macro (let's say \cmd{\trimx}) by putting \meta{bizarre} before 
the
scanned text and a stop pair \meta{bizarre}\meta{space} after it, in case a
leading space is not present:
\begin{lcode}
  \trimx<bizarre>#1<bizarre> \endtrimx
\end{lcode}
It would be possible to do without the bizarre token and have the
delimiter consist only of a space, but with some ensuing
complications, I think, that would make it scarcely worthwhile.

\subsection{Some remarks about the domain of the problem}

The application I had in mind was, generally speaking, to remove
unwanted spaces at the beginning and end of a piece of text supplied
by the user, such as a section title or other heading.

Typical situation: A user command \cmd{\title} takes an argument
\begin{lcode}
  \title{ Some Article Title }
\end{lcode}
with the definition of \cmd{\title} being
\begin{lcode}
  \def\title#1{\def\savedtitle{#1}\trimspaces\savedtitle}
\end{lcode}

Thereafter we may use \cmd{\savedtitle} in any number of ways: print it; put
it in a \cmd{\mark} for running heads; write it to an auxiliary file for
table of contents use, or for adding to a BibTeX database; or write it
on screen to show progress when typesetting a collection of articles.
For the last two examples in particular trimming spaces with
\cmd{\ignorespaces} or \cmd{\unskip} is undesirable.

Notice also that \cmd{\unskip} will remove \emph{any} trailing glue, including
\cmd{\leader}'s or explicit \cmd{\hskip}'s that might sometimes be added by
users for their own inscrutable purposes and whose unexpected
removal could be (indeed, has been in true life) the cause of
much consternation.

If we call \cmd{\trimspaces} in the definition of \cmd{\title}, then leading and
trailing spaces are removed once and for all, and none of the many
functions that later use \cmd{\savedtitle} need to worry about that task.

With this restricted domain of use in mind for \cmd{\trimspaces}, I screened
the submitted solutions through the following conditions.

\begin{description}
\item[Condition 1] The text has been stored in a macro. The result of
\cmd{\trimspaces} is a redefinition of the macro.

This is not exactly a necessary condition, but removal of this
condition would suggest that constructions like
\begin{lcode}
  \def\foo#1{...
    \message{Your argument "\trimspaces{#1}" makes me laugh}%
    ...}
\end{lcode}
should be supported. The full expansion done by \cmd{\message} or other such
commands, however, can't be applied carelessly to arbitrary
user-supplied text. You would need to deactive problematic elements
(by changing catcodes, adding \cmd{\protect}'s, whatever). So supporting
full expansion for the operand of \cmd{\trimspaces} is of low relevance for
the envisioned normal applications.

\item[Condition 2] It suffices to remove a single space before and after the
text.

In almost any other programming language, a typical space-trimming
function would need to handle the possibility of multiple consecutive
spaces. But in text supplied by an average user through the normal TeX
lexical conventions, consecutive spaces will be reduced to a single
space before our trimming functions are ever called.

The next installment of this `summary' will include a recently arrived
solution by Jonathan Fine\index{Fine, Jonathan} 
that handles multiple trailing spaces as
easily as a single one, without any extra implementation cost.

\item[Condition 3] For both the trailing space and the leading space, we
don't know whether or not they are present.

If we knew for certain that a given space was present, of course, the
procedure for removing it would be easier.
\end{description}

%%========================================================================
%%>>Solution 2 (Ian Collier) [Ian.Collier@prg.oxford.ac.uk]
%\begin{description}
\begin{solution}{Solution 2 (Ian Collier)}\index{Collier, Ian}

\ldots I used \cmd{\meaning} to find out whether or not the
first character of the argument is a space (because spaces are usually
ignored and this seems to be the only way to make the space visible).
I'm fairly sure that `blank space' is the only \cmd{\meaning} beginning with
`bl'.  I had rather a lot of trouble with braces, because if the first
character is a brace then \cmd{\meaning} removes it and leaves an unmatched
right brace.  However I finally realised that \verb?\iffalse...\fi? could be
used to remove it.
\begin{lcode}
{\catcode`Q=3 \catcode`@=11
 \gdef\trimspace#1{\expandafter\trimspac@a#1QAA QB}
 \gdef\trimspac@x#1{\trimspac@a#1QAA QB}
 \gdef\trimspac@a#1 Q#2{\if#2A#1\expandafter\trimspac@b
  \else\trimspac@c#1\fi}
 \gdef\trimspac@b A QB{}
 \gdef\trimspac@c#1QAA{#1}

 \gdef\trimspaces#1{\expandafter\expandafter\expandafter\tr@a
  \expandafter\meaning#1A\fi{#1}}
 \gdef\tr@a#1#2{\if#1b\if#2l\expandafter\expandafter\expandafter\tr@c
  \else\expandafter\expandafter\expandafter\tr@b\fi\else
  \expandafter\tr@b\fi}
 \gdef\tr@b{\expandafter\trimspace\iffalse}
 \gdef\tr@c{\expandafter\tr@d\iffalse}
 \gdef\tr@d#1{\expandafter\tr@e#1Q}
 \def\:{\gdef\tr@e}\: #1Q{\trimspac@x{#1}}
}

\def\test#1{\edef\text{#1}\immediate\write16 {"\trimspaces\text"}}
\test{ Leading space}
\test{Trailing space }
\test{ Leading and trailing spaces }
\test{Nospaces}
\test{ {braces}Leading space{braces}}
\test{{braces}Trailing space{braces} }
\test{ {braces}Leading and trailing spaces{braces} }
\test{{braces} Nospaces {braces}}
\test{}
\test{ }
\test{\space\space{two spaces}\space\space}

\end
\end{lcode}
%%========================================================================

Comments: Some extra work would be necessary to handle the possibility
\begin{lcode}
  \def\text{\iftrue a\else b\fi}
  \trimspaces\text
\end{lcode}
because removal of the \piif{iftrue} by \cmd{\meaning} will leave the 
\piif{else} and \piif{fi} unmatched, confusing the later \piif{iffalse} 
step done by \cmd{\tr@b}, \cmd{\tr@c}.
But such a value for \cmd{\text} is rather unlikely in ordinary
user-supplied arguments.
%\end{description}
\end{solution}

\begin{comment}
Some more solutions to Exercise 15 will follow in a few days.

Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~

Date: 30 Dec 1993 17:07:17 -0500 (EST)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #15, answers, 3rd installment
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}
%$

I have done some slight condensing in the answers, indicated by
\verb?[...]?.

Solution 3 by Greg McGary contains an interesting idea for an
alternative syntax of the \cmd{\trimspaces} function: Instead of writing
\begin{lcode}
  \def\savedtitle{#1}\trimspaces\savedtitle
\end{lcode}
you would write
\begin{lcode}
  \trimmed\def\savetitle{#1}
\end{lcode}
%%========================================================================
%%>>Solution 3 (Greg McGary, gkm@tmn.com)
%\begin{description}
\begin{solution}{Solution 3 (Greg McGary)}\index{McGary, Greg}
\begin{lcode}
%%% preliminaries: (Mad about those abbreviations!)
\catcode`@=11
\let\ea=\expandafter
\let\nx=\noexpand
\let\ag=\aftergroup
\def\agg{\ag\ag\ag}
\let\bg=\begingroup
\let\eg=\endgroup

[...]

%%% The underlaying tool I use is \trimmed, which is used as a modifier for
%%% macro definitions to trim the trailing space from the body:
%%% \trimmed\def\foo{foo } will set \foo to {foo}
%%% Notice that any form of \def modifier may be interposed between \trimmed
%%% and \def, as in \trimmed\global\long\outer\def\foo{foo }
%%% 
%%%   As an aside, TeX has no \expanded modifier.  Expanded definitions
%%%   must be accomplished through use of \edef or \xdef (equivalent to
%%%   \global\edef)  This is annoying, as we might like to use \trimmed with
%%%   expanded definitions and don't want to write a separate \etrimmed.
%%%   Luckily, we can easily roll our own \expanded modifier, like so:

\def\expanded#1\def{#1\edef}

%%%   Other modifiers may optionally be inserted between \expanded and
%%%   \def, like so: \def\foo{foo} \outer\expanded\long\def\bar{\foo}

%%% Here's the definition of \trimmed:

\long\def\trimmed#1\def#2#3{\bg
    \long\def\!##1##2 \!##3\trimmed@{\eg
	\ifx\relax##3\relax
	    \trimmed@{##1}##2%
	\else
	    ##1{##2}%
	\fi}%
    \!{#1\def#2}#3\! \!\trimmed@}

\long\def\trimmed@#1#2\!{#1{#2}}

%%% Notice the use of \begingroup...\endgroup to make the definition of \!
%%% temporary so as not to disturb any previous definition, and so that the
%%% temporary will disappear once we're done with it.  Notice that the
%%% \endgroup appears right away in the body of \!, so that the ensuing \def
%%% will occur in the proper group.  \! was chosen as a name for the temporary
%%% macro because it is a non-alphabetic (non-catcode-11) character; any other
%%% non-alphabetic would suffice as well.  Non-alphabetic macro-names have the
%%% desirable property of preserving any trailing space token.
%%%
%%% If we are really fastidious about keeping clutter out of the global name
%%% space, we can also define \trimmed@ as a temporary alongside \!.  We would
%%% also want to use a name that's already defined, to avoid entering a new
%%% name into TeX's hashtable.  A non-alphabetic name like \: seems like a
%%% good (though cryptic) choice:

\long\def\trimmed#1\def#2#3{\bg
    \long\def\:##1##2\!{\eg##1{##2}}
    \long\def\!##1##2 \!##3\:{%
	\ifx\relax##3\relax
	    \:{##1}##2%
	\else
	    \eg##1{##2}%
	\fi}%
    \!{#1\def#2}#3\! \!\:}

%%% Notice that we've had to delay the \endgroup until after our new
%%% temporary \: has been used.
%%%
%%% Anyway, we may now define \trimspace as follows:

\def\trimspace#1{\ea\trimmed\ea\def\ea#1\ea{#1}}

%%% Notice that the replacement definition is a normal \def, whereas the
%%% macro we started with could have had any number of modifiers attached,
%%% such as \long, \outer, or \global.  A further exercise might be to fix
%%% this problem.
%%%
%%% A more generalized trim might allow any list of tokens to be trimmed off
%%% the tail of another list of tokens.  Here, we add an initial argument to
%%% \trimmed specifying those tokens.  In order to strip off trailing ".\par"
%%% for instance, we could write: \trimmed{.\par}\outer\long\def\foo{foo.\par}
%%%
%%% Here's the general definition of \trimmed:

\long\def\trimmed#1#2\def#3#4{\bg
    \long\def\:##1##2\!{\eg##1{##2}}
    \long\def\!##1##2#1\!##3\:{%
	\ifx\relax##3\relax
	    \:{##1}##2%
	\else
	    \eg##1{##2}%
	\fi}%
    \!{#2\def#3}#4\!#1\!\:}

%%% The auxiliary \trimmed@ remains unchanged.  Notice that we no longer really
%%% need a non-alphabetic macro name for the temporary macro, since we don't
%%% have to preserve the literal space token following the macro.
%%%
%%% Unfortunately, the literal space token problem doesn't disappear, it's just
%%% pushed up a level.  Now we have to give that space as an argument to \trimmed
%%% in the definition of \trimspace, and hop over it with \expandafter!

\edef\trimspace#1{\nx\ea\nx\trimmed\nx\ea
    {\nx\ea\space\nx\ea}\nx\ea\def\nx\ea#1\nx\ea{#1}}

%%% N.B., The curly braces, "\nx\ea{...\nx\ea}" around the "\nx\ea\space"
%%% are necessary.
%%%
%%% This approach of defining \trimspace in terms of an underlaying \trimmed
%%% \def'inition facility has the advantage of reusing code, but the
%%% disadvantage of forcing a macro redefintion even if there is no trailing
%%% space to remove.  We could modify \trimmed to produce a new macro, \trim,
%%% that redefines a macro only if it has the trailing pattern of interest.
%%% (It also happens to be simpler!)

\long\def\trim#1#2{\bg
    \long\def\!##1#1\!##2\:{\eg
	\ifx\relax##2\relax \else
	    \def#2{##1}%
	\fi}%
    \ea\!#2\!#1\!\:}

%%% Now, we can define \trimspace in terms of \trim like so:

\edef\trimspace#1{\nx\ea\nx\trim\nx\ea{\nx\ea\space\nx\ea}\nx\ea#1}

%%% Ok, let's test it:

\def\HasTrailingSpace{has trailing space }
\def\NoTrailingSpace{no trailing space}

\trimspace\HasTrailingSpace \show\HasTrailingSpace
\trimspace\NoTrailingSpace  \show\NoTrailingSpace

%%% While we're at it, let's test another pattern:

\def\HasTrailingDotPar{has trailing dot par.\par}
\def\NoTrailingDotPar{no trailing dot par}

\trim{.\par}\HasTrailingDotPar \show\HasTrailingDotPar
\trim{.\par}\NoTrailingDotPar  \show\NoTrailingDotPar

%%% ### Exercise 15(b)
%%% Write a macro \trimspaces that removes a leading space, if
%%% present, and then calls \trimspace to remove a trailing space.

%%% I'm going to solve this in a quick and dirty way, as it's getting
%%% late and I'm running out of gas!  Just use \futurelet sequestered
%%% in a \vbox to inspect the first token. If it's a \space, gobble
%%% the first token and subject the remaining tokens to \trimmed.

\def\redefSansSp@ce#1 #2\redefSansSp@ce{\def#1{#2}}
\def\redefSansSpace#1{\ea\redefSansSp@ce\ea#1#1\redefSansSp@ce}
\def\trimspaces#1{\bg\setbox0=\vbox{%
    \def\maybeRedefSansSpace{\ea\ifx\space\@\agg\redefSansSpace\agg#1\fi}%
    \ea\futurelet\ea\@\ea\maybeRedefSansSpace#1}\eg
    \trimspace#1}

%%% \futurelet won't work for the more general case of trimming an
%%% arbitrary leading pattern, as it only looks at one token.
%%% I'll leave solving the general case as an exercise for the reader ;-)
%%% 
%%% This is also not the most efficient solution, since we redefine the macro
%%% twice if there is a leading space.  Notice that we put the \setbox0
%%% inside a group, to keep any previous definition of \box0 safe.  This
%%% is probably overkill, since \box0 is a temporary register and users
%%% should be aware that it's fair game, but it doesn't hurt to be
%%% courteous...  Also note the abbreviation \agg, which pushes its argument
%%% out two groups.

[...]

%%% Testing...

\def\foo{ foo }
\trimspaces\foo \show\foo
\end{lcode}

\end{solution}

%%========================================================================

In the previous posting I discussed the method of removing a trailing
space by scanning for a token pair \meta{space}\meta{bizarre}. In Schmitt's
solution, for example, the bizarre token was a greater-than character
with catcode 3. And in my solution, I used a letter Q with catcode
3. Solution 4 from Jonathan Fine takes the approach of using a second
\meta{space} token for the \meta{bizarre} token. In practice this works for
typical user-supplied text, as discussed before, since TeX's normal
reduction of multiple spaces to single spaces makes the pair
\meta{space}\meta{space} sufficiently bizarre. I have to admit I like this idea;
those who attempted a solution for this exercise and struggled with
various other delimiter possibilities will, I think, appreciate the
humor of it as I did.

As I mentioned last week, I found some theoretical interest in the
fact that if multiple space tokens were present at the end of the text
being trimmed, Fine's solution would remove them all, without needing
to use recursion. But another correspondent pointed out since then
that if multiple spaces were present at the end they might also be
presumed possible in the middle of the scanned text, and an occurrence
of multiple spaces in the middle would cause \cmd{\trim} to fail.



\begin{solution}{Solution 4 (Jonathan Fine)}\index{Fine, Jonathan}
\begin{lcode}
%% NOTE:  I have benefited from Michael Downes posting of answers, dated
%% 16 December, particularly for stripping the leading space, and the
%% discussion of the hazards of grouped arguments

\catcode`\@=11
%% The Solution
\def\trim #1{\expandafter\trim@\expandafter{#1 }#1}
\def\trim@ #1{\trim@@ @#1 @ #1 @ @@}
\def\trim@@ #1@ #2@ #3@@{\trim@@@\empty #2 @}
\def\unbrace#1{#1}
\unbrace{\def\trim@@@ #1 } #2@#3{\expandafter\def
  \expandafter #3\expandafter {#1}}

%% Test Code
\def\Test{\afterassignment\Test@ \def\test}
\def\Test@{\trim\test \afterassignment\Test@@ \def\test@}
\def\Test@@{\message{\ifx\test\test@ Y\else FAIL:|\meaning\test|\fi}}
\catcode`\@=12

%% Testing The Solution
\Test{}{}
\Test{ }{}
\Test{ a }{a}
\Test{ {}{} }{{}{}}
\Test{{braces}}{{braces}}
\Test{ {braces} }{{braces}}
\Test{ { braces } }{{ braces }}
\Test{no space and no space}{no space and no space}
\Test{no space and a space: }{no space and a space:}
\Test{ :a space and no space}{:a space and no space}
\Test{ :a space and a space: }{:a space and a space:}
\Test{ \ifx }{\ifx}
\Test{ \ifx/ }{\ifx/}
\end{lcode}
\end{solution}

\begin{comment}
Since my solution got rather long after I added some commentary I'll
post it separately in a couple of days, rather than double the size of
this post.

Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~

Date: 03 Jan 1994 17:14:14 -0500 (EST)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #15, answers, 4th (last) installment
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}
%$

My solution here is the result of weeks of incremental refinement,
ending only last week, and consequently benefits from analysis of the
other solutions.

%%========================================================================
\begin{solution}{Solution 5 (Michael Downes)}
\begin{lcode}
%    Here I only solve part (b) of Exercise 15, in an attempt to make
%    a solution of utmost compactness (3 control sequences, 45 tokens).
%    Also, it seems likely that in actual use \cmd{\trimspaces} can be
%    applied without harm whenever \trimspace might be needed.
%
%    The method for pausing after each test might be of ancillary
%    interest to some readers; unlike the alternative of setting
%    \pausing=1, the \test's aren't required to be on separate lines.

\catcode`\Q=3

%    \cs{trimspaces}\x redefines \x to have the same replacement text sans
%    leading and trailing space tokens.
%
\def\cs{trimspaces}#1{%
%    Use grouping to emulate a multi-token afterassignment queue.
  \begingroup
%    Put `\toks 0 {' into the afterassignment queue.
  \aftergroup\toks\aftergroup0\aftergroup{%
%    Apply \trimb to the replacement text of #1, adding a leading
%    \noexpand to prevent brace stripping and to serve another purpose
%    later.
  \expandafter\trimb\expandafter\noexpand#1Q Q}%
%    Transfer the trimmed text back into #1.
  \edef#1{\the\toks0}%
}

%    \trimb removes a trailing space if present, then calls \trimc to
%    clean up any leftover bizarre Qs, and trim a leading space. In
%    order for \trimc to work properly we need to put back a Q first.
%
\def\trimb#1 Q{\trimc#1Q}

%    Execute \vfuzz assignment to remove leading space; the \noexpand
%    will now prevent unwanted expansion of a macro or other expandable
%    token at the beginning of the trimmed text. The \endgroup will feed
%    in the \aftergroup tokens after the \vfuzz assignment is completed.
%
\def\trimc#1Q#2{\afterassignment\endgroup \vfuzz\the\vfuzz#1}

\catcode`\Q=11

\def\test#1{\errhelp{#1}\message{[\the\errhelp]}%
  \edef\x{\the\errhelp}%
  \global\tracingcommands2\global\tracingmacros2\global\tracingonline0
  \cs{trimspaces}\x
  \global\tracingcommands0\global\tracingmacros0\global\tracingonline0
  \errhelp\expandafter{\x}\message{-> [\the\errhelp]}%
  \read16 to\PressReturnToContinue
}

\test{ x }      \test{ xy z }   \test{}         \test{{}}
\test{{}{}}     \test{ {x} }    \test{ }        \test{{ }}
\test{\AA}      \test{\fi}      \test{\space x\space}
\test{ #1 }

\end
\end{lcode}

Commentary

Suppose we have a macro \cmd{\x} with replacement text \verb?" {xyz} "?. 
The task of
\cmd{\trimspaces} is to construct a statement of the form
\begin{lcode}
  \def\x{{xyz}}
\end{lcode}
i.e., to redefine \cmd{\x} with the same replacement text except for removal
of a leading or trailing space. However, a similar statement
\begin{lcode}
  \toks0{{xyz}}\edef\x{\the\toks0}
\end{lcode}
is more robust if the replacement text might contain \# tokens. For
example,
\begin{lcode}
  \def\x{\def\y##1{}}
\end{lcode}
works OK but after thus defining \cmd{\x}, the statements
\begin{lcode}
  \def\trimx#1{\expandafter\def\expandafter\x\expandafter{#1}}
  \trimx\x
\end{lcode}
fail with an error message because the `\#1' in the definition of \cmd{\y} is
misinterpreted as a parameter token for the redefinition of \cmd{\x}.

Although \# tokens seem highly unlikely in average user-supplied text, I
aimed for a statement of the second, robuster kind, as if I were writing
\cmd{\trimspaces} for use in a major macro package with thousands of
prospective users.

The basic structure of \cmd{\trimspaces} is therefore: First remove a trailing
space, then remove a leading space, then put the remaining text into
\cmd{\toks}\texttt{0}, then transfer the text to \cmd{\x} with \cmd{\edef}.

For removing the trailing space, I apply a macro scan with delimiter
\verb?<space,10><Q,3>? Here the notation \verb?<c,n>? means the character token
consisting of character code \texttt{c} with catcode \texttt{n}.

The leading space is removed by executing the assignment
\verb?\vfuzz=\the\vfuzz? at the beginning of the operand text, in order to use
a side effect of the assignment: removal of a following space.  (Credit
to Donald Arseneau for this good idea.) The main reason for using
\verb?\the\vfuzz? instead of 0pt is that it's slightly shorter (one token),
although if we did not have the group structure to localize the `change'
to \cmd{\vfuzz}, then using \verb?\the\vfuzz? would also be a good idea for the
sake of preserving the variable's previous value.

The statement \verb?\vfuzz=\vfuzz? (sans \cmd{\the}), by the way, would not gobble a
following space: when TeX recognizes a suitable variable on the
right-hand side of an assignment, it copies the value directly into the
left-hand side and skips the scanning process entirely.

Here's a step-by-step breakdown of the operation of \cmd{\trimspaces} through
two possibilities, one where both a leading and a trailing space are
present, and one where neither are present.
\begin{lcode}
------------------------------------------------------------------------
Case 1 (spaces present)              Case 2 (no spaces to be removed)
------------------------------------------------------------------------
\def\x{ {xyz} } \cs{trimspaces}\x        \def\x{{xyz}} \cs{trimspaces}\x

Step 1:                              Step 1:
\begingroup...                       Same as for Case 1.
\expandafter\trimb
\expandafter\noexpand\x Q Q}...

Step 2:              ||              Step 2:              ||
\trimb\noexpand {xyz} Q Q...         \trimb\noexpand{xyz}Q Q...
      ^^^^^^^^^^^^^^^                      ^^^^^^^^^^^^^^^
Here the row of ^^^ indicates the    In this case the first Q is taken
material that is taken as argument   up as part of #1, which is passed
#1 of \trimb, and || indicates the   to \trimc. The second Q added by 
tokens that match the macro          \trimb therefore falls after the 
delimiter. #1 is now passed to       leftover Q instead of before.    
\trimc, with another Q token added;
the leftover <space>Q token pair
follows.

Step 3:              |               Step 3:             |
\trimc\noexpand {xyz}Q Q...          \trimc\noexpand{xyz}QQ...
      ^^^^^^^^^^^^^^^ ^^                   ^^^^^^^^^^^^^^ ^
Here we have #1, delimiter token Q,  The situation at the end of the
and #2. The space before the second  trimmed text ends up being the same
Q is skipped by TeX because it's     as in Case 1, except for the
looking for a nondelimited argument  absence of a space between the Qs.
for #2.

Step 4:                              Step 4:
\afterassignment\endgroup            \afterassignment\endgroup
\vfuzz\the\vfuzz\noexpand {xyz}}...  \vfuzz\the\vfuzz\noexpand{xyz}}...
                         ^
Here the ^ marks the leading space
that is to be removed.

Step 5: \endgroup{xyz}}...           Step 5: \endgroup{xyz}}...

\endgroup is from \afterassignment.

Step 6:                              Step 6:
\toks0{{xyz}}                        \toks0{{xyz}}
^^^^^^^---from \aftergroup           ^^^^^^^---from \aftergroup
\edef\x{\the\toks0}                  \edef\x{\the\toks0}
\end{lcode}

\end{solution}

\begin{comment}
========================================================================

That's a wrap on Exercise 15.

Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
\end{comment}
%$
%%\endinput


\chapter{Assorted numbers, skips, and modes}

\section{Exercise}

%%\input{ex016}
% ex016.tex
\begin{comment}
Date: 13 Jan 1994 16:42:27 -0500 (EST)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #16
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>

************************************************************************
*** Exercise 16:
\end{comment}

\ed{\oposted{1994/01/13}. \arch{exercise.016}.}

Predict the messages that will be produced by plain TeX for the
following test file.
\begin{lcode}
\catcode`\@=11 \newcount\m
\def\msg#1{\advance\m 1 \message{(\number\m): #1}}
\def\T{\msg{T}}\def\F{\msg{F}}
\mag=1728 \hfuzz=1pt \tabskip=1pt \baselineskip=12pt
\topskip=10pt \lineskiplimit=1pt \lineskip=1pt

\setbox0\vbox{%
\mag=\time \ifnum\mag>1500 \T\else\F\fi                         %  (1) 
\mag=\number\year \ifnum\mag>1500 \T\else\F\fi                  %  (2) 
\hfuzz=99pt \ifdim\hfuzz=99pt \T\else \F\fi                     %  (3) 
\tabskip=\z@ \ifdim\tabskip<\p@\T\else\F\fi                     %  (4) 
\tabskip=\p@ minus2pt \ifdim\tabskip>\z@\T\else\F\fi            %  (5) 
\baselineskip=-\prevdepth \ifdim\baselineskip=12pt \T\else\F\fi %  (6) 
\advance\baselineskip 2\topskip                                 %  (7) 
  \ifdim\baselineskip>\@m\p@ \T\else\F\fi                       %        
\lineskiplimit=\z@ \ifnum\lineskiplimit>0 \T\else\F\fi          %  (8) 
\lineskip=\z@skip \ifdim\lineskip>\lineskiplimit \T\else\F\fi   %  (9) 
\kern2pc\ifdim\lastkern=2pc \T \else\F\fi                       % (10)
\hskip1em
  \ifvmode\T\else\ifdim\lastskip>\z@\msg{FT}\else\msg{FF}\fi\fi % (11) 
\font\cmrtest=cmr10 \ifx\cmrtest\tenrm \T\else\F\fi             % (12)
}
\end
\end{lcode}
Where should \cmd{\relax} be inserted?

\begin{comment}
************************************************************************

Answers will be posted circa January 27, 1994.

Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
\end{comment}
%$
%%\endinput


\section{Answers}

%%\input{ans016}
% ans016.tex
\begin{comment}
[There was an error in the first posted version: \twelverm instead of
the first \tenrm in the statement

  \font\tenrm = \fontname\tenrm scaled 1200

The posting containing this correction is appended below.]

Date: 27 Jan 1994 11:59:48 -0500 (EST)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #16, answers
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

\ed{\oposted{1994/01/27}. \arch{answer.016}.}

Here is my commentary on Around the Bend \#16.
\begin{lcode}
% \mag=1728 \hfuzz=1pt \tabskip=1pt \baselineskip=12pt
% \topskip=10pt \lineskiplimit=1pt \lineskip=1pt

% \mag=\time \ifnum\mag>1500 \T\else\F\fi                         %  (1) 
\end{lcode}
(1): F --- At the time of the \piif{ifnum}, \cmd{\mag} is in the range [0,1440)
depending on what time it was when you ran TeX.

\begin{lcode}
% \mag=\number\year \ifnum\mag>1500 \T\else\F\fi                  %  (2) 
\end{lcode}
(2): F --- At the time of the \piif{ifnum}, \cmd{\mag} still has its previous value
because TeX is still scanning for digits to add on after `1994'.

\begin{lcode}
% \hfuzz=99pt \ifdim\hfuzz=99pt \T\else \F\fi                     %  (3) 
\end{lcode}
(3): T --- Everything fine, dimension scanning terminated with the
space after `99pt'.

\begin{lcode}
% \tabskip=\z@ \ifdim\tabskip<\p@\T\else\F\fi                     %  (4) 
\end{lcode}
(4): F --- \cmd{\z@} is a dimension register, therefore it serves only as the
first part of the glue value that TeX is looking for. At the time of the
\piif{ifdim}, TeX is still looking for `plus' or `minus' and hasn't yet
finished the assignment of \cmd{\tabskip}.

\begin{lcode}
% \tabskip=\p@ minus2pt \ifdim\tabskip>\z@\T\else\F\fi            %  (5) 
\end{lcode}
(5): T --- Glue value scanning terminated properly. \cmd{\p@} is a dimension
register like \cmd{\z@} but the additional clause `minus 2pt' fills out the
glue value to the required three parts.  TeX assumes `plus 0pt' when it
finds a `minus' clause without a preceding `plus' clause. Note that TeX
does \emph{not} continue scanning for a possible `plus' after reading a minus
component. Unlike the height, depth, and width components of a \cmd{\vrule} or
\cmd{\hrule}, the components of a glue value have a required order and each
part can only occur once.

\begin{lcode}
% \baselineskip=-\prevdepth \ifdim\baselineskip=12pt \T\else\F\fi %  (6) 
\end{lcode}
(6): T --- At the beginning of a vbox or at the beginning of a TeX run
\cmd{\prevdepth} = -1000pt. So it would seem that \cmd{\baselineskip} should get set
to +1000pt and the test should be False; but \cmd{\prevdepth} is a dimension
register, not a glue register, so following stretch or shrink components
are still possible, and \cmd{\baselineskip} does not yet have its new value at
the time of the test.

\begin{lcode}
% \advance\baselineskip 2\topskip                                 %  (7) 
%   \ifdim\baselineskip>\@m\p@ \T\else\F\fi                       %        
\end{lcode}
(7): F --- Without the factor 2 in front of \cmd{\topskip}, the test would
be True: \cmd{\topskip} is a glue register so TeX would copy each component of
\cmd{\topskip} to the corresponding component of \cmd{\baselineskip}; then, having plus
and minus components already in hand, TeX would not scan ahead for
`plus' or `minus'. However, a preceding factor for a glue register
causes TeX to use only the first component of the glue register,
multiplied by the given factor, which means that additional scanning is
then attempted for possible stretch or shrink components.

\begin{lcode}
% \lineskiplimit=\z@ \ifnum\lineskiplimit>0 \T\else\F\fi          %  (8) 
\end{lcode}
(8): F --- Normal termination of dimension scanning. \cmd{\lineskiplimit}
is a dimen register, not a glue register, so the dimen constant \cmd{\z@} is
sufficient to complete the assignment and TeX scans no further.

\begin{lcode}
% \lineskip=\z@skip \ifdim\lineskip>\lineskiplimit \T\else\F\fi   %  (9) 
\end{lcode}
(9): F --- Normal termination of glue scanning. \cmd{\z@skip} is a glue
register so it suffices to complete the assignment of \cmd{\lineskip}. Compare
to the \cmd{\tabskip} assignments above.

\begin{lcode}
% \kern2pc\ifdim\lastkern=2pc \T \else\F\fi                       % (10)
\end{lcode}
(10): F --- At the time of the \piif{ifdim}, TeX is still looking for
an optional final space at the end of the dimension value `2pc'. If it
were \verb?2\p@? instead of \verb?2pc?, the test would evaluate to True.

\begin{lcode}
% \hskip1em
%   \ifvmode\T\else\ifdim\lastskip>\z@\msg{FT}\else\msg{FF}\fi\fi % (11) 
\end{lcode}
(11) FF --- TeX enters horizontal mode as soon as the \cmd{\hskip} command
comes along, before it finishes scanning the skip amount. So the
\piif{ifvmode} test is false. The \piif{ifdim} test is also false because scanning
is not yet complete (TeX is looking ahead for a plus or minus component)
so the glue has not yet been entered into the horizontal list, so it is
not accessible to \cmd{\lastskip}.

For more on the switch into horizontal mode, see `TeX from \cmd{\indent} to
\piif{par}', Marek Ry{\'c}ko and Bogus{\l}aw Jackowski, TUGboat 14/3, October
1993 (1993 Annual Meeting Proceedings), pp. 171--176.

\begin{lcode}
% \font\cmrtest=cmr10 \ifx\cmrtest\tenrm \T\else\F\fi             % (12)
\end{lcode}
(12) F --- Interestingly, the following versions of the \piif{ifx} test are
also false at that point:
\begin{lcode}
 \ifx\cmrtest\undefined, \ifx\cmrtest\relax.
\end{lcode}
The reason is that after `\verb?\font\cmrtest?' TeX immediately sets 
\verb?\cmrtest = \nullfont?, before scanning the rest of the font assignment. So the test
\verb?\ifx\cmrtest\nullfont? would yield True. According to the \emph{TeXbook},
the reason for this behavior is to allow statements of the form
\begin{lcode}
  \font\cmrtest=cmr10 \cmrtest
\end{lcode}
for switching to the font \cmd{\cmrtest} immediately after it is defined. TeX
does a bit of boomeranging in such a case:
\begin{lcode}
  \font\cmrtest          % set \cmrtest = \nullfont
  =cmr10                 % space terminates font name, start looking for
                         % "at" or "scaled"
  \cmrtest               % \cmrtest = \nullfont = nonexpandable, not
                         % "a", not "s"; terminate the font assignment
                         % and put back the \cmrtest token to be read
                         % again:
  \cmrtest               % Now \cmrtest selects the given font
\end{lcode}
Although I sympathize with Knuth's desire to smooth out a potential
problem for naive users, I wonder if it only encourages users to pay
less attention to the nitty-gritty details of scanning and expansion,
and therefore lay themselves open to greater confusion later on when
something similar fails (inconsistently!) to work. I'd have thought it
better to require, and document, proper termination of font assignment
scanning by \cmd{\relax} or whatever. Users would have to be a little more
knowledgeable but they would be rewarded with a more consistent language
to work with.  As it stands TeX unnaturally forbids certain
constructions that are perfectly colloquial to anyone who has an ear for
the TeX language, such as
\begin{lcode}
  \font\tenrm = \fontname\tenrm\space scaled 1200
\end{lcode}
I hold a similar opinion for the way \cmd{\chardef} and \cmd{\mathchardef} set their
arguments to \cmd{\relax} before scanning the number on the right-hand-side of
the assignment. Occasionally I would \emph{like} to be able to write
something like
\begin{lcode}
 \chardef\foo=\ifcase\foo 1\or 2\else 3\fi
\end{lcode}
 but TeX doesn't allow that.

One could argue that the \cmd{\chardef} behavior should for consistency be
imitated by \cmd{\edef}, \cmd{\xdef} so that if \cmd{\foo} is undefined then
\begin{lcode}
  \edef\foo{a\foo}
\end{lcode}
should not give an undefined control-sequence error for the \cmd{\foo} in the
replacement text, but make it temporarily equivalent to \cmd{\relax} and leave
it there.  (Of course, this means that executing \cmd{\foo} will then start up
an infinite loop, but my point was that it's the behavior of \cmd{\chardef}
that should be changed to achieve consistency, not the behavior of
\cmd{\edef}.)

%%%========================================================================

At the end of Exercise \#16 there was the question `Where should \cmd{\relax}
should be inserted?'

\cmd{\relax} should be inserted just before the \piif{if}... in statements (2), (6),
(7), (11), and (12). In statement (4) \cmd{\z@skip} should be used instead of
\cmd{\z@}; then \cmd{\relax} is unnecessary. A space suffices instead of \cmd{\relax} in
(10). I would also tend to put a \cmd{\relax} at the end of the preliminary
assignments to \cmd{\baselineskip} and \cmd{\lineskip}, as a matter of principle; I
like to make sure that scanning is definitely terminated at the end of a
line, so that if any error occurs during the scanning, TeX will show the
line containing the assignment statement and not a later line. This is
particularly relevant for font assignments: If \pfile{foo10.tfm} does not exist
on your system, then the assignment
\begin{lcode}
  \font\foo=foo10
  <blank line>
\end{lcode}
will cause TeX to show you the blank line instead of the preceding line
in the error context:
\begin{lcode}
  ! Font \foo=foo10 not loadable: Metric (TFM) file not found.
  <to be read again>
                     \par
  l.2
\end{lcode}
And if the following material is some complicated macro instead of a
blank line, TeX will go into the replacement text of the macro, looking
for `at' or `scaled', before giving the error message!

\begin{comment}
Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~

Date: 28 Jan 1994 08:01:12 -0500 (EST)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #16, answers, correction
To: info-tex@shsu.edu

Instead of

  \font\twelverm = \fontname\tenrm\space scaled 1200

read

  \font\tenrm = \fontname\tenrm\space scaled 1200

The latter line is what I originally wrote but I changed it in an obtuse
moment a day later, forgetting the very point it was supposed to
illustrate.
\end{comment}
%$
%%\endinput


\chapter{Missing \cs{input} file}

\section{Exercise}

%%\input{ex017}
% ex017.tex
\begin{comment}
Date: 14 Jan 1994 12:44:13 -0500 (EST)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #17
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

\ed{\oposted{1994/01/15}. \arch{exercise.017}.}

%%************************************************************************
%%*** Exercise 17:
When TeX cannot find an input file it prompts with `Please enter another
input file name:'. On some systems you can enter `nul' in response to
this prompt to have TeX input a null file and continue processing. On
most systems TeX also allows you to enter a system-dependent end-of-file
character (Control-Z (DOS, VMS), Control-D (Unix), ...?), to which it
responds with an "Emergency stop" instead of continued processing.

An alternative would be to maintain a file called `\pfile{.tex}' containing an
error message so that merely pressing RETURN would cause TeX to read
`\pfile{.tex}' and issue the error message. Unlike the null file case or
EOF-character case, this would allow normal access to the full menu of
error recovery options, including e.g., exiting to an editor, inserting
or deleting tokens, or changing the interaction mode. It would probably
be nice to have the file also accessible under various aliases `\pfile{h.tex}',
`\pfile{help.tex}', `\pfile{?.tex}', `\pfile{q.tex}', `\pfile{quit.tex}', 
`\pfile{x.tex}', `\pfile{exit.tex}', or
`\verb?@#&@%$.tex?' corresponding to typical responses from stumped users.

But making a robust `\pfile{.tex}' file for input error recovery is not so
simple a task as might first seem. One needs to take into account, for
example, the possibility that an \cmd{\input} might be attempted when normal
catcodes or normal \cmd{\endlinechar} are not in effect.

Given the programmability of TeX, an all-encompassing solution is
probably not possible, so this exercise has two parts: consider what
would be a reasonable minimal set of assumptions for an input error
recovery file; and write a \pfile{.tex} file containing a suitable 
error message and satisfying the assumptions.
%%************************************************************************

Motivation: From \url{comp.text.tex}:
\begin{lcode}
> From: wayne@csri.toronto.edu (Wayne Hayes)
> Subject: Why does TeX ignore interupts???
> Message-ID: <1993Dec24.000935.2007@jarvis.csri.toronto.edu>
> Date: 24 Dec 93 05:09:35 GMT
> 
> If there's ONE thing that annoys me more than anything about a program,
> it's when it refuses to die on command, and for no good reason.  The
> absolute worst case is when it's waiting for input and you don't know
> what to tell it, and would like to quit for now.
> 
> Thus my extreme annoyance every time I mistype an \input command to TeX
> and it asks me on the terminal "Please input another file name: ", and
> I usually just want to exit and re-edit my file to fix the \input
> error.  But TeX refuses to die when I press ^C at this moment, and will
> only die if I send a QUIT (^\), at which point it dumps a
> multi-megabyte core file into the current directory.  ARGGGHHHH!!  Why
> does it do this?  I can't see any good reason why it ignores interupts
> at this point.  Is this intended?  Is it a bug?  Does it drive anyone
> else as nuts as it drives me??  Can it be changed in the next release???
\end{lcode}

It's puzzling that most of the implementations of TeX I know of don't
check for the interrupt key possibility at this prompt [Textures notably
cuts clean through the problem by popping up a dialog box if an input
file is not found]. Seems as if interrupt-key checking at that point
would be a desirable addition to the set of system-dependent changes for
each system.

\begin{comment}
A summary will be posted circa February 17, 1994.

Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
\end{comment}
%$
%%\endinput

\section{Answers}

%%\input{ans017}
% ans017.tex
\begin{comment}
[The TUGboat article mentioned below appeared as [info not yet
available--18-Aug-1994]]

Date: 17 Mar 1994 13:04:36 -0500 (EST)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #17, answers
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>
\end{comment}

\ed{\oposted{1994/03/13}. \arch{answer.017}.}

Exercise 17 (posted January 14) asked for an error recovery file to
provide better recovery from file input errors: When TeX cannot find an
input file, it prompts for an alternative file name and refuses to
continue until a valid file name is entered or the user presses some
(system-dependent) abort key. This can be rather unfriendly, especially
for novice users.

At the request of Barbara Beeton\index{Beeton, Barbara} (TUGboat's editor) 
I wrote up the
results of this exercise as an article for publication in TUGboat, so
this posting will be largely redundant with that article.


%%%-------------------------------------
%%DON'T BOTHER, REDEFINE \cmd{\input} INSTEAD
\subsection{Don't bother, redefine \cs{input} instead}

Interestingly, both of the answers I received 
(from Victor Eijkhout\index{Eijkhout, Victor} and 
Donald Arseneau\index{Arseneau, Donald}) recommended redefining input 
instead of trying to
make an input error recovery file. Donald summed it up thus:
\begin{quotation}
 Since verbatim file input is an important mainstream application,
 the task is hopeless.
 
 The right approach is to redefine \cmd{\input} and check for the file's
 existence at the macro level.
\end{quotation}

I.e., consider the way a typical \cmd{\verbfile} commands works: first, start
a group; next, deactivate all special characters such as \verb?\ { } # % }? by
changing their catcodes; then input the desired file; and finally close
the group to restore normal catcodes. If the desired file is not found
and an input error recovery file is read instead, the IERF will not be
able to do anything because of the deactivation of \verb?\ { }? etc.


%%----------------------------------------------
%%DIFFICULTIES ASSOCIATED WITH REDEFINING \cmd{\input}
\subsection{Difficulties associated with redefining \cs{input}}

Generally speaking I am in favor of redefining input (for instance,
to make up for the deficiency in TeX that the current input file name
is not accessible like \cmd{\jobname} or \cmd{\inputlineno}), but there are
some practical problems:

\begin{itemize}
\item In order to serve all users, the redefinition of \cmd{\input} would
have to go into plain TeX, LaTeX, and any other major macro
packages that are not layered on top of plain TeX or LaTeX.

\item The most commonly used approach to test for the existence of an input
file is
\begin{lcode}
  \openin N=file.name \ifeof N ...
\end{lcode}
but for
some TeX implementations \cmd{\openin} will only open a file in the
current directory, and not search through the entire `TeX inputs'
path. I believe that this restriction is canonical in \pfile{TeX.web}
therefore only overridden by the system-dependent changes of each TeX
implementation according to the judgment of the individual implementor.

\item The details of how to redefine \cmd{\input} are nontrivial. If you
redefine \cmd{\input} to take an argument delimited by a space, for
example, there is some risk of bombing on existing files with
statements like
\begin{lcode}
  \input x.y\relax
\end{lcode}
It becomes especially nontrivial if you want to use some method other
than simple \verb?\openin ... \ifeof? to test for file existence, so that
the method will be reliable across all systems.

It is worth noting that in LaTeX2e the \cmd{\input} command has
been dramatically overhauled so that it solves, among other things,
some of the problems mentioned here. Anyone doubting the claim that
the work is nontrivial is invited to look at the LaTeX2e definitions.

\item Redefining \cmd{\input} will (generally speaking) not help for the
jobname file itself. When the file name is given on the command line, or
following a ** prompt, the input operation is done directly by
TeX instead of through invoking the control sequence \cmd{\input}.

\item When a non-existing file is called for by a verb-file command,
TeX will prompt the user for a file name, and then if a \pfile{.tex} recovery
file exists, pressing \meta{return} will typeset the contents of that file;
but this is at least as good as inputting a null file, in that you are
not stuck at the prompt with no obvious way to quit.
\end{itemize}

%%----------------------------------------------------------
%%SOMEBODY ALREADY PUBLISHED SOME INPUT ERROR RECOVERY FILES
\subsection{Somebody already published some input error recovery files}

Coincidentally, reading through one of my books a few days after posting
Around the Bend \#17, I found that someone had already written and
published a suite of input error recovery files: 
Frank Mittelbach\index{Mittelbach, Frank}, \emph{The
LaTeX Companion}, section 14-4 \ed{First edition}.


%%------------------------------------------------------
%%BUT WHAT THE HECK, HERE ARE MY SLIGHTLY DIFFERENT ONES
\subsection{But what the heck, here are my slightly different ones}

The basic idea is to create a file named \pfile{h.tex} that will produce an
\cmd{\errmessage}\verb?{...}? statement. Copies (or links) of this file will be made
under several different names corresponding to the typical user
responses to an input file error, to the extent that the operating
system permits.

So a first attempt would be something like this:
\begin{lcode}
  \errmessage{Enter x to exit or ? to see other options}
\end{lcode}
Suppose we test this with a simple test file:
\begin{lcode}
  %   This is line 1
  %   This is line 2
  \input fzrg \relax % This is line 3
  %   This is line 4
  \end
\end{lcode}

The on-screen result looks like this:
\begin{lcode}
  ! I can't find file `fzrg.tex'.
  l.3 \input fzrg 
                  \relax % This is line 3
  Please type another input file name: h
  (h.tex
  ! Enter x to exit or ? to see other options.
  l.1 ... to exit or ? to see other options}

  ?
\end{lcode}

Then if the user enters \texttt{?} they will see
\begin{lcode}
  Type <return> to proceed,
  S to scroll future error messages,
  R to run without stopping,
  Q to run quietly,
  I to insert something,
  E to edit your file,
  1 or ... or 9 to ignore the next 1 to 9 tokens of input,
  H for help, X to quit.
  ? x
\end{lcode}

Now let's examine this solution a little more closely, to ask what are
the potential problems, and what assumptions can be done away with?

One problem is the possibility of an unusual catcode for space, question
mark, left brace, right brace, backslash, or \cmd{\endlinechar}. For the
backslash (and the letters) we don't have much choice; if they don't
have normal catcodes, \pfile{h.tex} cannot issue an \cmd{\errmessage} command, or even
try to fix up the catcodes. (This is why the problem of verbatim file
input is insoluble, if primitive \cmd{\input} is used.) Note that for users of
a macro package such as texinfo, which has \verb?@? for the escape character
instead of backslash, a different IERF would be required.

The \cmd{\endlinechar} problem can be solved by adding a percent sign at the
end of the line:
\begin{lcode}
  \errmessage{...}%
\end{lcode}
but at the cost of a new assumption: percent must have catcode 14. This
and some of the other catcode assumptions can be removed with a bit of
extra work:
\begin{lcode}
  \begingroup\chardef\%37\catcode\%14\chardef\ 32\catcode\ 10\relax%
  \catcode123 1\catcode125 2\catcode63 12 %
  \errmessage{%
  Enter x to exit or ? to see other options}%
  \endgroup\endinput%
\end{lcode}
This enforces the desired catcodes for \verb|space, %, {, }, and ?|; and
putting \% at the end of each line makes \cmd{\endlinechar} harmless, no matter
what its prevailing value and catcode might happen to be. The
\cmd{\begingroup} ... \cmd{\endgroup} pair of course keep the catcode changes local,
just in case (though I expect that the user will normally choose to exit
anyway). I write
\begin{lcode}
  \chardef\%37\catcode\%14
\end{lcode}
in preference to the alternatives
\begin{lcode}
  \catcode37 14
  \catcode37=14
  \catcode37'16
  \catcode37"E
  \catcode`\%14
\end{lcode}
which require assuming a usable catcode for one extra character (space
or = or ' or ...). Even using \cmd{\string}, as in
\begin{lcode}
  \catcode37\string"E
\end{lcode}
would fail if \texttt{"} had catcode 5, 9, 10, 11, 14, or 15.

Here now is the screen output produced by the above IERF:
\begin{lcode}
  ! I can't find file `fzrg'.
  l.3 \input fzrg 
                  \relax % This is line 3
  Please type another input file name: h
  (h.tex
  ! Enter x to exit or ? to see other options.
  l.5  Enter x to exit or ? to see other options}
                                                 %
  ? x
\end{lcode}

%%------------------
%%BEST FINAL VERSION
\subsection{Best final version}

There is one fairly obvious drawback of the above IERF: the error
message is repeated twice on screen, once by \cmd{\errmessage} and once in the
error context shown for line 5. There is a little trick that can be used
to fix that: Use only the error context for showing the message text, by
putting it in a comment rather than in the argument of \cmd{\errmessage}!
[Cf.the comment after \cmd{\patterns} in the original TeX hyphenation patterns
file hyphen.tex.]

\begin{lcode}
  \begingroup\chardef\%37\catcode\%14\chardef\?63\catcode\?12\relax%
  \chardef\{123\catcode\{1\chardef\ 32\catcode\ 2\relax%
  \errmessage{Input\string canceled\string ..%
   % Enter x to exit or ? to see other options %
  \endgroup\endinput%
\end{lcode}
I have thrown in some extra cleverness with the catcode of space to
clean up the screen output a tiny bit more. The result looks like this:
\begin{lcode}
  ! I can't find file `fzrg'.
  l.3 \input fzrg 
                  \relax % This is line 3
  Please type another input file name: h
  (h.tex
  ! Input canceled ...
  l.4  
       % Enter x to exit or ? to see other options %
  ? x
\end{lcode}

Frank Mittelbach's IERF solution differs from mine by providing a set of
files that attempt to mimic standard TeX error recovery according to
their name: The file \pfile{s.tex}, for example, arranges to switch into
\cmd{\scrollmode} and continue processing, as would happen if you entered `s'
at a normal error message prompt. And there are files named \pfile{e.tex},
\pfile{x.tex}, \pfile{q.tex} that mimic the corresponding error message actions. His
IERFs also don't bother to worry about possible odd catcodes for \{,
space, \}, etc.---an approach whose simplicity perhaps outweighs the
minor added robustness of my version.

%%-----------
%%CONCLUSIONS
\subsection{Conclusions}

It seems that it would be a worthy service to their users if the authors
of all TeX implementations took a second look at how input file errors
are handled and added suitable actions depending on the operating
system. For example, under DOS it is difficult to create a file named
\pfile{.tex}, so perhaps emTeX, PCTeX, TurboTeX, etc., should check for the case
when the user presses the \meta{return} key at the prompt, and automatically
exit instead of trying to input a highly improbable file! Similar
arguments would hold for an input file name of \pfile{?} or \pfile{?.tex} 
for operating
systems where \texttt{?} is an OS wild-card character.

And another part of improving the input error handling might be to add
to their standard distributions a set of IERFs in the TeX inputs area,
to help users who are using some macro package \emph{other} than LaTeX2e.
(Or, even for LaTeX2e users, to help in the case when it is the jobname
file itself that was not input-able.) I recommend of course my IERF
given above; my feelings would not be deeply wounded, however, if
Frank's version gets used instead. Installing either version would be
much better for end users than none at all.

\begin{comment}
Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
\end{comment}
%$
%%\endinput


\chapter{Page breaking}

\section{Exercise}

%%\input{ex018}
% ex018.tex
\begin{comment}
Date: 21 Apr 1994 09:48:48 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #18
To: info-tex@shsu.edu
X-ListName: TeX-Related Network Discussion List <INFO-TeX@SHSU.edu>

========================================================================
*** Exercise 18:
\end{comment}

\ed{\oposted{1994/04/21}. \arch{exercise.018}.}

On page 254 of the \emph{TeXbook} the following output routine is described:
\begin{lcode}
	\output={\unvbox255 \penalty\outputpenalty}
\end{lcode}
and in the ensuing text Knuth writes `If the \cmd{\vsize} hasn't changed, and
if no insertions have been held over, the same page break will be
found.' This claim is rather false. Why? How should the output routine
be rewritten to work as intended?
%%========================================================================

Thanks to William Baxter\index{Baxter, William}
%(web@superscript.com) 
for contributing this question.

\begin{comment}
Michael Downes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
mjd@math.ams.org (Internet) ASCII 32--54,55--126: !"#$%&'()*+,-./0123456
789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
\end{comment}
%$
%%\endinput


\section{Answers}

%%\input{ans018}
% ans018.tex
\begin{comment}
Date: 27 May 1994 08:19:39 -0400 (EDT)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Around the Bend #18, answer
To: info-tex@shsu.edu
\end{comment}

\ed{\oposted{1994/05/27}. \arch{answer.018}.}

I intended to post this sooner but in researching the answer it turned
out that in order to clear up a couple of nagging questions I had to
follow some side trails a long way.

%%Answer to Around the Bend #18:

Exercise 18 (21 April 1994) pointed out that the output routine
\begin{lcode}
	\output={\unvbox255 \penalty\outputpenalty}
\end{lcode}
described in the \emph{TeXbook} p 254 doesn't exactly work as intended: `If
the \cmd{\vsize} hasn't changed, and if no insertions have been held over, the
same page break will be found.'

The same pagebreak will be found only if the original page break
occurred at a penalty item. Otherwise (\emph{TeXbook}, p 125) TeX
sets \cmd{\outputpenalty}\texttt{=10000} before firing up the user's output
routine. Consequently, the output routine constructs a vertical list in
which the original break point has disappeared.

By an optimization found in section 890 of \emph{TeX: The Program}, the
penalty between two paragraph lines---the sum of all applicable
penalties from the set \cmd{\interlinepenalty}, \cmd{\clubpenalty}, 
\cmd{\widowpenalty}, \cmd{\displaywidowpenalty}, and \cmd{\brokenpenalty}---is 
not actually added to
the vertical list unless it is nonzero. Thus when \cmd{\interlinepenalty} =
0 (default from IniTeX/plain TeX) and hyphenated lines are not too
frequent, `most' pairs of lines in a paragraph have no intervening
penalty. And there is usually no penalty between ordinary text
paragraphs. Thus an \cmd{\outputpenalty} value of 10000 will occur fairly
often in practice.

W. E. Baxter\index{Baxter, William}\index{Baxter, W E|see{Baxter, William}}
(the submitter of this exercise) 
looked into the
possibility of recompiling TeX without the cited optimization, but found
that the resulting version fails the trip test.

In order for the example to work as intended it would have to be
rewritten as
\begin{lcode}
  \output={\unvbox255
    \ifnum\outputpenalty=10000 \else \penalty\outputpenalty\fi}
\end{lcode}
For completeness it should be pointed out that the output routine
could come even closer to the goal of `doing nothing' if the parameter
\cmd{\holdinginserts}, added in TeX version 3.0 (circa 1990), were set to
some value greater than 0, so that the state of floating inserts would
be preserved; but that has to be done before the output routine is
entered.

I would have said that such a do-nothing output routine is useless, but
as a matter of fact I wrote something rather close to it as one cycle of
a multi-cycle output routine a couple of years ago. The goal was to look
at the values of \cmd{\pagetotal}, \cmd{\pagestretch}, etc in order to print a
complete survey of the page contents in a marginal note, to help the
person dealing with page break decisions when the automatic breaks
turned out to be inadequate. Unfortunately, the values of \cmd{\pagetotal} etc
reported in the output routine are not exactly the values that are
needed, because if the page break did not occur at a forcing penalty 
($<=-10000$) then the values include material on the recent contributions
list, yet only the material up to the chosen page break is relevant. So
in order to get accurate values I had to insert a do-almost-nothing
cycle that merely inserted a forcing penalty at the break point after
dumping the contents of \texttt{box255} back on the main vertical list.

%%------------------------------------------------------------------------
\subsection{Some historical research}

If you have an older copy of the \emph{TeXbook} (pre-1990), as I do, the
above-mentioned section on p 125 about \cmd{\outputpenalty} says that it is
set to 0 (rather than 10000) if the break did not occur at a penalty
item. Thus the output routine example on p 254 seems to be another
case of a well-known phenomenon: documentation failing to keep up with
changes in the software. Make a note of it in your copy!

Excerpt from the \emph{TeXbook} errata files:
\begin{verbatim}
  \bugonpage A125, lines 13--29 (9/23/89)

  \ddanger \looseness=-1
  When the best page break is finally chosen, \TeX\ removes everything after
  the chosen breakpoint from the bottom of the ``current page,'' and puts it
  all back at the top of the ``recent contributions.'' The chosen
  breakpoint itself is placed at the very top of the recent contributions.
  If it is a penalty item, the value of the penalty is recorded in
  ^|\outputpenalty| and the penalty in the contribution list is changed
  to $10000$; otherwise |\outputpenalty| is set to 10000.
\end{verbatim}

It's not clear to me from a cursory examination of \pfile{tex82.bug},
\pfile{errata-five.tex}, and \pfile{tex.web} when this change occurred 
in \pfile{tex.web}, but it
seems that it must have occurred rather early, perhaps in the work on
TeX82 (1982--1983); if so, then the claim that outputpenalty was set to
0 was a five-year-old oversight when Knuth changed it in 1989. In
\pfile{tex82.bug} there is no reference to output\_penalty or even inf\_penalty
near 9/23/89, and tracing backwards from there didn't turn up anything
that seemed relevant to me. Furthermore, a copy of TeX version 2 (circa
1985) that I was able to dig up had outputpenalty 10000 instead of 0,
following the erratum, and my 1986 copy of \emph{TeX: The Program} (i.e.
the woven version of tex.web) agrees with that.

Thanks again to W. E. Baxter\index{Baxter, William} for contributing 
this exercise and several parts of the answer.

%%\endinput



\chapter{Author lists}

%%\input{bend019}
% bend019.tex

\section{Exercise (hard)}

\ed{\oposted{1994/08/23}}

First, an 
announcement: Archive copies of exercises and solutions in the 
Around the Bend series are now available over the network, thanks to the 
ongoing remarkably fine service of CTAN (\url{ftp.shsu.edu}, 
\url{ftp.dante.de}, \url{ftp.tex.ac.uk},\ldots). Look in the directory 
\url{tex-archive/info/aro-bend}.

%======================================================================== 
%%*** Exercise 19 (hard): 

In a multi-author LaTeX article, author names are normally given 
as a list with \cmd{\and} separating the names, for example
\begin{lcode}
Arthur B. Clark\and Damian Edlan\and Ferency G. van Hoep
\end{lcode}

The way the author names are laid out on the printed page may 
vary widely from one publication to another. The generic 
`article' documentclass provides a definition for \cmd{\and} to print 
the author names together with their addresses in an array form. 
But there is no support in basic LaTeX to print such a list of 
names in standard series form 
\begin{lcode}
A            (1 author)
A and B      (2 authors)
A, B, and C  (3+ authors)
\end{lcode}

\begin{enumerate}
\item Write a macro \cmd{\andlist} to convert a list of author names to 
    series form. Assume that the names reside in a macro \cmd{\@author}. 

            Suggested tests:
\begin{lcode}
\def\test#1{\def\@author{#1}% 
  % Convert contents of \@author, leave result in \@temp: 
  \andlist\@author\@temp
  % Examine the result
  \message{\@temp}}

\test{Arthur B. Clark}
\test{Arthur B. Clark\and Damian Edlan}
\test{Arthur B. Clark \and Damian Edlan \and Ferency G. van Hoep}
\test{Arthur B. Clark \and Damian Edlan 
      \and Ferency G. van Hoep \and Irene Jackson}
\end{lcode}

to produce

\begin{lcode}
Arthur B. Clark
Arthur B. Clark and Damian Edlan
Arthur B. Clark, Damian Edlan, and Ferency G. van Hoep
Arthur B. Clark, Damian Edlan, Ferency G. van Hoep and Irene Jackson
\end{lcode}


Extra credit:

\item discuss the relative merits of the following alternatives:
     
     \begin{enumerate}
\item \verb?\andlist\@authors\@temp? The function \cmd{\andlist}
            takes two macro names
            as arguments, converts the contents of the first macro and leaves
            the result in the second macro. 

\item \verb?\andlist\@authors? The function \cmd{\andlist}
            takes one macro name as its argument and replaces the 
            contents of the macro with the converted version of its contents.

\item \verb?\andlist\@authors? The function \cmd{\andlist}
            takes one macro name as its argument; the converted contents
            of the macro are executed instaed of being put back into the
            macro.

\item other?

\end{enumerate}

\item Extend your definition of \cmd{\andlist} to make it easy to change 
      the material placed between names, for example, to omit the last 
      comma in a list of three or more names, or to use small-caps for 
      the word `and', or to put each name in a box to prevent a line 
      break within a name, or to put a `good break' penalty after each 
      comma.

\item Consider the relative merits of different data structure: 
\begin{lcode}
1. A\and B\and C 
2. A,B,C 
3. \do{A}\do{B}\do{C}
\end{lcode}

    For example, if it were required that each author name must be 
given by a separate \cmd{\author} command, the third kind of data 
structure would be slightly simpler to produce, as compared to 
the first two. Having the data in the second form might make it 
possible for \cmd{\andlist} to use some of the pre-existing internal 
routines in LaTeX for processing comma-separated lists. And so forth.

\end{enumerate}

%%======================================================================== 

As usual, creative variations---such as using token registers 
instead of macros---are encouraged if their aptness is evident 
or explained. 


   Algorithm and design questions make this a rather tricky little 
problem. (Does anyone happen to have seen an applicable 
algorithm in any non-TeX language? I imagine it may be needed in 
some SGML applications.) 

Solutions will be posted circa September 12, 1994. 

%%Michael Downes 


\section{Editor's notes}

    I have not been able to find where, or even if, any answers were posted, 
which is unfortunate as I think that it is a useful exercise. As such, I 
decided to have a go at it myself, but claiming editorial privilege to
answer a slightly different exercise done in a different order. 

    The basic question is how to convert a list of names separated by a 
particular token (\cmd{\and} in the exercise) to a list of the same names
with different separators (for example `,'). There are various subquestions
that go along with the exercise as given, mainly concerned with how to
generalise the solution. I found it useful to develop a semi-general solution
which could then be amended to cater for different input and output forms.
Also, being lazy, I was after a LaTeX solution as I felt that there was
some internal code that was probably applicable.

    There are basically three separators that may appear in the final list:
\begin{itemize}
\item If there is only a single name in the list, no separator is required.
\item If there are two names then a separator is required between them,
      call this \cmd{\pairsep}.
\item If there are three or more names in the list then there is a separator
      between the penultimate and last name (call this \cmd{\lastsep}),
      and separators between all the previous names, and I'll call this
      \cmd{\midsep}.
\end{itemize}
In the initial exercise as given these are, respectively, `and', `, and' 
and `,'. The implication here is that for the general case of more than 
two entries we need to know 
when we are coming to the end of the list so that we can insert \cmd{\lastsep}
just before outputting the last list entry.

    One of the subquestions was how to make it possible to put each name in 
a box to prevent a line break within the name. To do this implies that 
each name
should be output as the argument of a macro, say \cmd{\opname}, that can be
used to perform some action on the name.

    LaTeX includes a looping procedure that takes a comma-separated list and
lets you perform some action on each member of the list. Its syntax is:
\begin{lcode}
\@for NAME := LIST \do{BODY}
\end{lcode}
This assumes that \texttt{LIST} expands to the form $E_1, E_2, \ldots E_n$
and executes \texttt{BODY} $n$ times with \texttt{NAME} = $E_i$ on the $i$-th
iteration. This is what I will use as the basis of my solution.

    Here's my basic general solution, where the list of names is of the form
\texttt{A,B,C,D,\ldots N}. I'm assuming that this is in a \pfile{.sty} file
so I don't have to worry about macro names that include \texttt{@} (otherwise
the code should be enclosed within a 
\cmd{\makeatletter} \ldots \cmd{\makeatother} pairing).

\begin{lcode}
%% these are in LaTeX kernel
\providecommand{\z@}{0}
\providecommand{\@ne}{1}
\providecommand{\tw@}{2}

\newcount\totalcnt % total number of names in list
\newcount\entrycnt % number of `current' name
\newcommand*{\opname}[1]{#1}
\newcommand*{\pairsep}{\space and} 
\newcommand*{\midsep}{\unskip,}
\newcommand*{\lastsep}{\unskip, and}
%% \commaed is the key part of the solution, converting
%% the separators in a comma-separated list to something else
\newcommand*{\commaed}[1]{%
%%% #1 is comma-separated list of names
  %% get number of names
  \totalcnt\z@%   zero \totalcnt
  \@for\@tempa:=#1\do{\advance\totalcnt\@ne}%
  %% process the list
  \entrycnt\@ne%  initialise \entrycnt to 1
  \@for\@tempa:=#1\do{%
    \advance\entrycnt\@ne%  increment \entrycnt
    \ifnum\totalcnt=\@ne
%% a single entry
      \opname{\@tempa}
    \else
      \ifnum\totalcnt=\tw@
%% just two entries
        \ifnum\entrycnt=\tw@
          \opname{\@tempa}\pairsep
        \else
          \opname{\@tempa}
        \fi
      \else 
%% More than two entries in list
       \ifnum\entrycnt<\totalcnt
       %% in the middle of the list
         \opname{\@tempa}\midsep
       \else 
         \ifnum\entrycnt=\totalcnt
           %% current name is the penultimate
           \opname{\@tempa}\lastsep
         \else
           %% this is the last name
           \opname{\@tempa}
         \fi
       \fi
     \fi
  \fi
  }% end of do
}% end of definition
\end{lcode}    

The macro \cmd{\commaed} takes a comma-separated list as its argument and 
outputs a revised list.
\newcount\totalcnt % total number of names in list
\newcount\entrycnt % `current' name
\newcommand*{\opname}[1]{#1}
\newcommand*{\pairsep}{\space and} 
\newcommand*{\midsep}{\unskip,}
\newcommand*{\lastsep}{\unskip, and}
\makeatletter
\newcommand*{\commaed}[1]{%
%%% #1 is comma-separated list of names
  %% get number of names
  \totalcnt\z@%   zero \totalcnt
  \@for\@tempa:=#1\do{\advance\totalcnt\@ne}%
  %% process the list
  \entrycnt\@ne%  initialise \entrycnt to 1
  \@for\@tempa:=#1\do{%
    \advance\entrycnt\@ne%  increment \entrycnt
    \ifnum\totalcnt=\@ne
%% a single entry
      \opname{\@tempa}
    \else
      \ifnum\totalcnt=\tw@
%% just two entries
        \ifnum\entrycnt=\tw@
          \opname{\@tempa}\pairsep
        \else
          \opname{\@tempa}
        \fi
      \else 
%% More than two entries in list
       \ifnum\entrycnt<\totalcnt
       %% in the middle of the list
         \opname{\@tempa}\midsep
       \else 
         \ifnum\entrycnt=\totalcnt
           %% current name is the penultimate
           \opname{\@tempa}\lastsep
         \else
           %% this is the last name
           \opname{\@tempa}
         \fi
       \fi
     \fi
  \fi
  }% end of do
}% end of definition
\makeatother

    The macro \cmd{\testcommaed} can be used to test \cmd{\commaed}. 
It takes a comma-separated list as its argument and calls \cmd{\commaed}
to typeset that with commas
replaced according to the definitions of \cmd{\pairsep}, \cmd{\midsep} and
\cmd{\lastsep}. The macro \cmd{\opname} is used to typeset the elements. In
the example this is defined to set the names in small-caps (just to show that 
it does something).

\begin{lcode}
\renewcommand*{\opname}[1]{\textsc{#1}}
\newcommand*{\testcommaed}[1]{%
  \def\alist{#1}%
  \commaed{\alist}}
\end{lcode}

\renewcommand*{\opname}[1]{\textsc{#1}}
\newcommand*{\testcommaed}[1]{%
  \def\alist{#1}%
  \commaed{\alist}}
\def\AL#1{\textit{Originally: \alist}}

    Some results are shown below.

\begin{itemize}
\item \verb?\testcommaed{Arthur B. Clark} ->? \\ 
  \testcommaed{Arthur B. Clark}

\item \verb?\testcommaed{Arthur B. Clark, Damian Edlan} ->? \\ 
  \testcommaed{Arthur B. Clark, Damian Edlan}

\item \verb?\testcommaed{Arthur B. Clark, Damian Edlan ,? \\
      \verb?Ferency G. van Hoep} ->?  \\ 
      \testcommaed{Arthur B. Clark, Damian Edlan , Ferency G. van Hoep}

\item \verb?\testcommaed{Arthur B. Clark, Damian Edlan,? \\
      \verb?Ferency G. van Hoep , Irene Jackson} ->? \\ 
      \testcommaed{Arthur B. Clark, Damian Edlan,
      Ferency G. van Hoep , Irene Jackson}
\end{itemize}

    The macro \cmd{\anded} is similar to \cmd{\commaed} execpt that the 
separator between list elements is \cmd{\and} instead of a comma. It is 
implemented using \cmd{\commaed}.
\begin{lcode}
\newcommand*{\anded}[1]{%
  \def\and{, }
  \edef\Alist{#1}
  \commaed{\Alist}}
\newcommand{\testanded}[1]{%
  \def\alist{#1}%
  \anded{\alist}}
\end{lcode}

\newcommand*{\anded}[1]{%
  \def\and{, }
  \edef\Alist{#1}
  \commaed{\Alist}}
\newcommand{\testanded}[1]{%
  \def\alist{#1}%
  \anded{\alist}}
The macro \cmd{\testanded} provides a means of testing \cmd{\anded} and some
results are given below.

\begin{itemize}
\item \verb?\testanded{Arthur B. Clark} ->? \\ 
      \testanded{Arthur B. Clark}
\item \verb?\testanded{Arthur B. Clark\and Damian Edlan} ->? \\ 
      \testanded{Arthur B. Clark\and Damian Edlan}

\item \verb?\testanded{Arthur B. Clark \and Damian Edlan\and? \\
      \verb?Ferency G. van Hoep} ->?  \\ 
      \testanded{Arthur B. Clark \and Damian Edlan\and
      Ferency G. van Hoep}

\item \verb?\testanded{Arthur B. Clark\and Damian Edlan\and? \\
       \verb?Ferency G. van Hoep \and Irene Jackson} ->? \\ 
       \testanded{Arthur B. Clark\and Damian Edlan\and
       Ferency G. van Hoep \and Irene Jackson}
\end{itemize}

    Finally, here is an answer to Michael's initial exercise (with a change 
in the names of macros to avoid the use of \texttt{@}). This is built on the
\cmd{\anded} macro. Test results are shown after the code definitions.

\begin{lcode}
\newcommand*{\andlist}[2]{
  \def\intermediate{\anded{#1}}
  \let#2=\intermediate}
\def\test#1#2{%
  \def\alist{#1}
  \andlist{\alist}{\Alist}}
\end{lcode}

\newcommand*{\andlist}[2]{
  \def\intermediate{\anded{#1}}
  \let#2=\intermediate}
\def\test#1#2{%
  \def\alist{#1}
  \andlist{\alist}{\Alist}}

\begin{itemize}
\item \verb?\test{Arthur B. Clark}{\Alist} \Alist ->? \\
      \test{Arthur B. Clark}{\Alist} \Alist 

\item \verb?\test{Arthur B. Clark\and Damian Edlan}{\Alist} \Alist ->? \\ 
      \test{Arthur B. Clark\and Damian Edlan}{\Alist} \Alist 

\item \verb?\test{Arthur B. Clark \and Damian Edlan\and? \\
      \verb?Ferency G. van Hoep}{\Alist} \Alist ->?  \\ 
      \test{Arthur B. Clark \and Damian Edlan\and Ferency G. van Hoep}{\Alist} \Alist 

\item \verb?\test{Arthur B. Clark\and Damian Edlan\and? \\
      \verb?Ferency G. van Hoep \and Irene Jackson}{\Alist} \Alist ->? \\
      \test{Arthur B. Clark\and Damian Edlan\and
       Ferency G. van Hoep \and Irene Jackson}{\Alist} \Alist 
\end{itemize}

    I think that I have shown enough for you to code answers
to the `extra credit' questions. By now, it should be obvious that I find
the \verb?A,B,C...? data structure to be advantageous compared with the
\verb?A\and B\and C...? structure because of the LaTeX \cmd{\@for} code I used.
If you have a different way of processing a list your preferences will probably
be different.

%%\endinput


\chapter{Math symbols}

%%\input{bend020}
% bend020.tex

\section{Exercise}

\ed{\oposted{1994/08/30}}

%%%*** Exercise 20: 

Why does plain.tex define \cmd{\surd} like this: 
\begin{lcode}
\def\surd{{\mathchar"1270}}
\end{lcode}
instead of like this:
\begin{lcode}
\mathchardef\surd="0270
\end{lcode}
?

%======================================================================== 
 
% Michael Downes

\begin{lcode}
%%%% Self-decoding answer: run the following text through plain TeX %%%% 
\let\+\let\+\a\advance\+\c\catcode\+\d\def\+\f\fam\+\m\mag\+\u\uccode\m 
13\c\m9\+\p\uppercase\d\i{\a\f7 \ifnum\f>125 \a\f-93 \fi}\d~{\u\f\m \c\m 
12 \a\m1 \i \ifnum\m>125 \+~\1\fi~}\d\0#1{\ifnum`#1>"D \if#1 !\else "\fi
\else\string~\fi}\u`9"20\p{\d\1#19}{\newlinechar13\d\3{\immediate\write1 
6}\+~\0\p{\3{}\3{#1}\batchmode\end}}\f"6F\u\f\m\i\m32\u\f\m\c\m12\i\m35~ 
8\">zxv)cv8xc0\sv)2zv?z\sv},{doo;sz$;"0xsZZ;U^)2l2^x~}%,O{hhvjxcs0lz"v^v 
U^)2cxsv^)cUv>9)2v)2zv"LUecNo7zx)9l^NNLvlz\)zxzsvc\v)2zvU^)2v^E9"mvFN^"" 
v%fff)2zv$9x")vs9+9)fffU^Gz"o^vU^)2cjv^)cU_v>2c"zvlc\)z\)"v^xzvlz\)zxzsv 
eLv`z|v9$v)2zLv^xzv\c)29\+oe0)v^v"9\+Nzv$c\)vl2^x^l)zxkv)2zvzE)x^v"z)vc$ 
vex^lz"v)2z\vl^0"zv`z|v)coj^lGv)2zvlz\)zxzsvl2^x^l)zxv9\)cv^vU^)2cxsv^)c 
U_vxz"0N)9\+v9\v)2zosz"9xzsvU^)2cxsv"j^l9\+vc\v)2zvNz$)v^\svx9+2)mv=\v)2 
zvc)2zxv2^\so;U^)2l2^xsz$;"0xsy~}{,O{_v>29Nzv")9NNvjxcs0l9\+v^vU^)2cxsv^ 
)cU_v>c0NsoL9zNsv^vxz^NNLv9\)zxz")9\+vjc"9)9c\vc$v)2zv"LUecNvCjxce^eNLv\ 
c)v>2^)vLc0o>c0Nsv+0z""kv)xLv9)v^\sv"zzJmvF$mvR0Nzv%%v9\v8jjz\s9Evbvc$v` 
2zv`z|eccGm >c0Nsv+0z""kv)xLv9)v^\sv"zzJmvF$mvR0Nzv%%v9\v8jjz\s9Evbvc$v` 
\end{lcode}

\section{Answer}

\begin{comment}
%%%% the result of TeXing the above
This is pdfTeXk, Version 3.141592-1.40.3 (Web2C 7.5.6)
 %&-line parsing enabled.
entering extended mode
(./codeans20.tex

Answer to Around the Bend #20:
\end{comment}

\ed{A ran the above through pdfTeX and it produced the following (less the formatting
that I added to the plain ASCII) as the answer. I suspect, though, that the command
\cs{ver} below is a typo and should not be there.}

\begin{lcode}
\def\surd{{\mathchar"1270}}
\end{lcode}
 produces a mathord atom with the symbol
vertically centered on the math axis. Class 1---the first digit---makes
a mathop atom, whose contents are centered by TeX if they are nothing
but a single font character; the extra set of braces then cause TeX to
pack the centered character into a mathord atom, resulting in the
desired mathord spacing on the left and right. On the other hand
\begin{lcode}
\ver\mathchardef\surd="0270
\end{lcode}
 while still producing a mathord atom, would
yield a really interesting position of the symbol (probably not what you
would guess; try it and see). Cf. Rule 11 in Appendix G of \emph{The TeXbook}.

%%\endinput


\chapter{Variable number of arguments}

%%\input{bend021}
% bend021.tex
\begin{comment}
\documentclass{memoir}
\usepackage{bend}
\usepackage{comment}
\usepackage{url}

\begin{document}
\end{comment}

\section{Remarks}

\ed{\oposted{2002/09/13}}

Back in the days when 
there existed an INFO-TeX mail list whose postings were 
automatically piped (by suitable arrangements) into 
\url{comp.text.tex}, I launched a thing called `Around the Bend'
with the following explanation:
\begin{quote}
[Date: Thu 10 Oct 91]

I would like to propose a regular department for INFO-TeX, 
called `Around the Bend'.
It will 
consist of macro-writing challenges on the level of the 
dangerous-bend exercises
in the \emph{TeXbook}, 
with interested parties invited to 
collaborate and/or compete to find the best solution. My 
motivation for doing this is partly selfish: to get more  
feedback from other macro writers about some of the interesting 
macro-writing problems that I run into.
\end{quote}

  There was never any attempt to establish a regular schedule for 
Around the Bend postings, I simply would do another one whenever I ran across an 
interesting problem, if I was able to spare some time to do so. The 
series is archived at \url{CTAN:pub/tex/info/aro-bend}
for anyone who has an interest in looking at it. I also noticed that the 
exercises and answers are available in \url{comp.text.tex} archives 
through \url{groups.google.com}.

    In response to a question on July 24, 2002 from Antoine 
Chambert-Loir\index{Chambert-Loir, Antoine} (with apologies for the delay in answering):
\begin{quote}
 \ldots why did 'Around the Bend' stop? 
There were nice challenges proposed there.
\end{quote}

I am tempted to say `Well, actually they didn't stop, there was 
just an unusually large gap in the aperiodic schedule'. 

But what I also wanted to say is that there are others quite as 
capable as I am of devising good Around the Bend 
exercises---I am thinking of a recent post by David Kastrup\index{Kastrup, David} 
about a completely expandable string comparison macro---and it 
occurred to me it might be better to invite interested parties 
to sign up for an informal `editorial board' to issue further 
exercises, so that other demands on my time do not have such a 
dampening effect on the rate of output. I don't have any desire 
to put restrictions on what goes out in continuation of the 
series apart from a (fairly crucial) one of striving for high 
quality and creativity. Send e-mail if you are interested, to 
the address below. There are only some obvious questions of 
coordination to address, such as trying (I think) to avoid two 
different people posting different exercises at the same time. 

Turning now to the next exercise, prompted by a recent 
\url{comp.text.tex} question from David Reitter\index{Reitter, David}: 

%======================================================================== 

%%*** Exercise 21: 
\section{Exercise}

    Define a macro that takes a variable number 
of arguments. Do it in the best way possible. For the sake of 
concreteness, consider this somewhat contrived example as a test 
case that your solution should be able to handle, though 
possibly using a different syntax:
\begin{lcode}
\printdate                         -> today's date in preferred form 
\printdate[Tuesday]                -> Tuesday
\printdate[Tuesday][17]            -> Tuesday the 17th
\printdate[Tuesday][17][9]         -> Tuesday, September 17th
\printdate[Tuesday][17][9][2002]   -> and so on
\printdate[Tuesday][17][9][2002][Gregorian calendar] -> and so forth 
\end{lcode}

    The lines above illustrate six different ways of calling the 
\cmd{\printdate} macro. The macro should print something appropriate 
in each case, but the exact form of the output is a matter of 
taste, it need not follow exactly what I have given here. 

    Part of a good solution will be a good analysis of why one way 
might be better than another. The solution that I came up with 
is based on the question from David Reitter\index{Reitter, David} that originally 
inspired this exercise, thus it assumes the context is LaTeX and 
tries to solve the problem in a way that is natural for LaTeX. 

    A straightforward solution based on existing examples of 
multiple-option commands in the LaTeX kernel would qualify as 
natural, but definitely not elegant since that would require 
defining a separate macro to handle each stage of the multiple 
option scanning. Non-LaTeX solutions are also considered to be 
of interest. 

%======================================================================== 

    I suggest posting your answers directly to comp.text.tex instead 
of mailing them to me (as was done in the past), though 
depending on how late you stayed up working on this entertaining 
exercise instead of writing your thesis or balancing your 
checkbook as you \emph{ought} to have been doing, you might want to 
beware of posting in haste and wait until you have had some 
sleep and a chance to reread what you wrote, to avoid 
embarrassing oversights [\ldots said he, speaking from experience]. 

    Please e-mail a copy in addition (or instead, if you like) to the 
Around the Bend Editorial Board ... hmm, that gives me an idea \ldots [pausing to 
consult the dictionary] make that the Supremely Honorable, 
Ingenious and, in Special Honor of Knuth, Around the Bend Editorial
Board---whose size will not long remain one I dare say, 
especially after the establishment of this glamorous name---at 
\url{<see acronym>@pobox.com}

%%Regards, Michael Downes 

\begin{comment}
            target=_parent>...</A>@ams.org (Michael J Downes) writes: 
            <P>
            <DIV class=qt id=qhide_741198 style="DISPLAY: block">&gt; 
            ======================================================================== 
            &gt; *** Exercise 21: &gt; Define a macro that takes a 
            variable number of arguments. Do it in the &gt; best way 
            possible. For the sake of concreteness, consider this somewhat 
            &gt; contrived example as a test case that your solution should 
            be able to &gt; handle, though possibly using a different 
            syntax: 
            <P>&gt; &nbsp; \printdate &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 
            &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; -&gt; today's date in 
            preferred form &gt; &nbsp; \printdate[Tuesday] &nbsp; &nbsp; 
            &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;-&gt; "Tuesday" &gt; 
            &nbsp; \printdate[Tuesday][17] &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 
            &nbsp;-&gt; "Tuesday the 17th" &gt; &nbsp; 
            \printdate[Tuesday][17][9] &nbsp; &nbsp; &nbsp; &nbsp; -&gt; 
            "Tuesday, September 17th" &gt; &nbsp; 
            \printdate[Tuesday][17][9][2002] &nbsp; -&gt; and so on &gt; 
            &nbsp; \printdate[Tuesday][17][9][2002][Gregorian calendar] -&gt; 
            and so forth 

\end{comment}

\section{Answers}


%\textbf{David Kastrup (2002/09/14)}
\begin{solution}{Solution 1 (David Kastrup)}\index{Kastrup, David}

\ed{\oposted{2002/09/14}}

\begin{lcode}
\def\printdate{\count@\z@\toks@{}\printdate@a} 
\def\printdate@a{\@ifnextchar[{\printdate@b}{\printdate@c}}
\def\printdate@b[#1]{\toks@\expandafter{\the\toks@{#1}}%
  \advance\count@\@ne\printdate@a} 
\def\printdate@c{\csname printdate@@\romannumeral\count@
  \expandafter\endcsname\the\toks@}
\end{lcode}

    You can now define the one-argument macro \cmd{\printdate@@i}, the 
5-argument macro \cmd{\printdate@@v} and so on.

    \cmd{\printdate@c} might also contain other stuff. For testing, 
we just define it as
\begin{lcode}
\def\printdate@c{\message{\number\count@\space arguments: \the\toks@}}
\end{lcode}

    This needs the LaTeX macro \cmd{\@ifnextchar}, of course. 

    If you want to have various defaults in sequence and just want to 
call \cmd{\printdate@@v}, you could write something like 
\begin{lcode}
\def\printdate@c{\let\gobble@x\relax\expandafter\newcommand
  \expandafter\gobble@x\expandafter[\number\count@]{}%
  \edef\next{{Tuesday}{17}{9}{2002}{Gregorian calendar}%
  \the\toks@}\expandafter\expandafter\expandafter
  \printdate@@v\expandafter\gobble@x\next}
\end{lcode}

    Ok, this latter proposal is ugly. Better ideas?
% -- David Kastrup, Kriemhildstr. 15, 44793 Bochum Email: 
\end{solution}


\begin{solution}{Solution 2 (mine)}

\ed{\oposted{2002/09/20}}

%\textbf{Michael J Downes (Sep 20, 2002)} 

Define a macro that takes a variable number of arguments. 
and gave the following example application: 
\begin{lcode}
\printdate                       -> today's date in preferred form 
\printdate[Tuesday]              -> Tuesday
\printdate[Tuesday][17]          -> Tuesday the 17th
\printdate[Tuesday][17][9]       -> Tuesday, September 17th
\printdate[Tuesday][17][9][2002] -> and so on 
\end{lcode}

My solution (see below), written with LaTeX in mind, has the 
            following characteristics: 
\begin{itemize}
\item The kernel of the solution is not specific to a particular 
            user-level command; for each user-level command, only two 
            command-specific macros are needed: the top-level one invoked by 
            the user, and the internal one that handles all the arguments. 
            By contrast, the standard LaTeX method of handling multiple 
            options requires a separate command-specific macro for each step 
            of the argument scanning. 
\item The number of optional arguments is quasi-limited. The number 
            of default values that you give in a command's definition 
            becomes an upper limit on the number of arguments that will be 
            scanned for. And if you supply twenty default values, the code 
            that ends up handling them will have to be more than a simple 
            TeX macro since macro arguments only go up to 9. 
\item Commands defined with this method can be nested, because the 
            delimiters for the optional arguments are regular curly braces \verb?{ }?,
            not square brackets [ ]. 
            
\end{itemize}

The choice of square brackets in LaTeX for optional arguments is 
            OK for arguments whose values are suitably restricted, but when 
            used for arguments that may contain arbitrary text---in 
            particular, other commands with optional arguments---it becomes 
            a pitfall that many users have fallen into over the years, and 
            generally costing them an amount of lost time in inverse 
            proportion to their understanding of catcodes. (I.e., its worst 
            effects are on the kind of users that LaTeX was intended to 
            serve in the first place.) The most common examples in practice 
            are perhaps \cmd{\twocolumn}\verb?[...]? and \verb?\begin{thm}[...]?, but it could 
            also happen in the optional arguments of \cmd{\section}, \cmd{\caption}, or 
            \cmd{\cite}. 

The chief argument against using braces for optional arguments 
            came out coincidentally in another thread only a couple of days 
            ago, as stated by Heiko Oberdiek\index{Oberdiek, Heiko} on \url{comp.text.tex} 
\begin{comment}
(&lt;am6mb5$a1<A 
            href="http://groups.google.com/groups/unlock?msg=b6e2e27a4e4413f7&amp;_done=/group/comp.text.tex/browse_thread/thread/cd0bd09362b1ac6c/b6e2e27a4e4413f7%3Flnk%3Dgst%26q%3Daround%2Bthe%2Bbend" 
            target=_parent>...</A>@n.ruf.uni-freiburg.de&gt; comp.text.tex 17 
            Sep 2002): 
\end{comment}
%$
\begin{quote}
How do you want to distinguish between a parameter and a 
group, both enclosed in \verb?"{}"? Example:
\begin{lcode}
\foo{bar}{\bfseries bla}
\end{lcode}
\end{quote}

But in practice it seems to me that this is not a significant 
            drawback. Savvy users would normally use the \verb?\textbf{...}? form 
            anyway (I hope). 

In fact the \verb?"{\whatever ...}"? form (called a \emph{declaration} in the 
            LaTeX book) is, in a certain sense, quite unnatural for a linear 
            language like TeX where the macro expansion works by simple 
            left-to-right substitution. At least, if used at document level 
            such a syntax makes it unnecessarily difficult to remap the 
            functions involved and therefore is a stumbling block in many 
            special applications. For example, it becomes feasible to add 
            italic corrections automatically only when we use the \cmd{\emph}\verb?{...}? 
            form rather than the \verb?{?\cmd{\em}\verb?...}? form. (There is an 
\cmd{\aftergroup} 
            trick that would sort of do the job but only by placing some 
            assumptions on the usage that do not hold in the real world.) 
            

%%%Regards, Michael Downes 

%            <P>------------------------------------------------------------------------ 
\begin{lcode}
\documentclass{article}
\usepackage{ifmtarg}
\makeatletter

% If \cmd{\MyCmd} is defined as 
%    \VariableArgs{\MyCode ...}{{Default1}{Default2}}
% then 
%     \MyCmd        -> \MyCode...{Default1}{Default2}
%     \MyCmd{aaa}   -> \MyCode...{aaa}{Default2}
%     \MyCmd{a}{bc} -> \MyCode...{a}{bc}
% In other words, \VariableArgs takes two arguments <code> and <defaults>
% and if the invocation via \MyCmd finds $n$ actual arguments, the first 
% $n$ default values are replaced by the actual arguments. 
%
% In principle the number of optional arguments is "whatever \MyCode is 
% able to handle" but if the number of defaults is $d$ then scanning 
% will stop as soon as $d$ arguments have been read, if not before. 
% In practice things will begin to get unwieldy after a dozen or so 
% arguments, because the process of scanning one more 
% actual argument involves rescanning the whole list of arguments 
% each time (actual arguments read previously plus any remaining defaults).

\newcommand{\VariableArgs}[2]{%
  \toks@{#1}%
  \@ifnextchar\bgroup{\AddArg #2{}@}{#1#2}}

\def\AddArg#1#2@#3{%
  \toks@\expandafter{\the\toks@{#3}}%
  \edef\RunIt{\the\toks@}%
  \@ifnextchar\bgroup{%
    \ifx @#2@%
      \begingroup
      \def\AddArg{\endgroup \expandafter\RunIt\@gobble}%
    \fi
    \AddArg #2@%
   }{%
     \RunIt #2%
   }%
  }

\newcommand{\printdate}{%
  % If zero args, use \today.
  \VariableArgs{\PrintDateFive}{{\today}{}{}{}{}}}

% This example is slightly more complicated than necessary because it 
% behaves differently depending on the number of arguments. 
\newcommand{\PrintDateFive}[5]{%
  % Always print #1, which might be \today (from the default value).
  #1%
  \@ifnotmtarg{#2#3#4#5}{%
    % If only #1 & #2 are given, use a slightly different form.
    \@ifmtarg{#3#4#5}{ the}{,}%
    % Args 2,3,4,5: Print each one if nonempty, but rearranging the
    % order slightly.
    \@ifnotmtarg{#3}{ \MonthName{#3}}%
    \@ifnotmtarg{#2}{ \OrdinalDay{#2}}%
    \@ifnotmtarg{#4}{, #4}%
    \@ifnotmtarg{#5}{ (#5)}%
   }}

\def\MonthName#1{%
  \ifcase 0#1 \number\month\or
    January\or February\or March\or April\or May\or June\or 
    July\or August\or September\or October\or November\or December% 
    \else Thirteen's Month\fi}

% If #2 is not a digit, use #1
\def\LastDigit#1#2{%
  \ifodd 0#21 \else #1\expandafter\@gobbletwo\fi\LastDigit #2}

\def\OrdinalDay#1{#1%
  \ifcase\LastDigit #1\space th\or st\or nd\or rd\else th\fi}

\begin{document} 
\noindent Testing:
\begin{enumerate}\setcounter{enumi}{-1}
\item \printdate
\item \printdate{Tuesday} 
\item \printdate{Tuesday}{17}
\item \printdate{Tuesday}{17}{9} 
\item \printdate{Tuesday}{17}{9}{2002} 
\item \printdate{Tuesday}{17}{9}{2002}{Gregorian calendar}
\end{enumerate}
\end{document}
\end{lcode}

\end{solution}


\begin{solution}{Solution 3 (Donald Arseneau)}\index{Arseneau, Donald}
%%\textbf{Donald Arseneau (2002/09/24)}

\ed{\oposted{2002/09/24}}

*** Exercise 21: \\
Define a macro that takes a variable number of arguments. 
\begin{lcode}
\printdate[Tuesday][17][9][2002][Gregorian calendar] -> and so forth
\end{lcode}

I did it (acually before MD posed the challenge) 
using \verb?{ }?, not \verb?[ ]?, and this answer does not match the challenge 
in other ways. But I haven't got around to working it in the last week or so.

    Two features notably missing are: error checking for a bad 
number when specifying the number of arguments, and provision 
of default values for omitted arguments (they are all null 
here). 
(I also think I could make do with one fewer 
\cmd{\MultiArgCollect} macros.)

    I think \verb?{}? delimiters really are the `best way' in regards to 
nesting macros. The one problem is confusion with 
non-explicit \verb?{?, and so I handle the most common case of \cmd{\bgroup}. 

\begin{lcode}
\makeatletter 
\let\MultiArgBgroup={

\def\MultiArg#1#2{\begingroup
  \let\bgroup\begingroup \let\egroup\endgroup
  \expandafter\MultiArgCollect\romannumeral\number#1001\delimiter{#2}}
 
\def\MultiArgCollect#1{\csname MultiArgCollect#1\endcsname}
\def\MultiArgCollectm#1\delimiter#2{%
  \@ifnextchar\MultiArgBgroup
    {\MultiArgCollectA#1\delimiter{#2}}%
    {\MultiArgCollect#1\delimiter{#2{}}}}

\def\MultiArgCollectA#1\delimiter#2#3{%
  \MultiArgCollect#1\delimiter{#2{#3}}}}

\def\MultiArgCollecti#1\delimiter#2{\endgroup#2}

\newcommand\DeclareMultiArgCommand[2]{\expandafter
  \Declare@MultiArg@ \csname MA\string_\string#1\endcsname{#1}{#2}}
\def\Declare@MultiArg@#1#2#3{%
  \DeclareRobustCommand{#2}{\MultiArg{#3}{#1}}
  \newcommand{#1}[#3]}

\DeclareMultiArgCommand {\printdate}{6}{...}
\end{lcode}

\end{solution}

%%\endinput

\indexintoc
\printindex

\end{document}