summaryrefslogtreecommitdiff
path: root/Build/source/libs/type1/t1io.c
blob: dd69bf6539ce18bdc614c9452ed0f121eec27caa (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
/* $XConsortium: t1io.c,v 1.4 91/10/10 11:19:41 rws Exp $ */
/* Copyright International Business Machines,Corp. 1991
 * All Rights Reserved
 *
 * License to use, copy, modify, and distribute this software
 * and its documentation for any purpose and without fee is
 * hereby granted, provided that the above copyright notice
 * appear in all copies and that both that copyright notice and
 * this permission notice appear in supporting documentation,
 * and that the name of IBM not be used in advertising or
 * publicity pertaining to distribution of the software without
 * specific, written prior permission.
 *
 * IBM PROVIDES THIS SOFTWARE "AS IS", WITHOUT ANY WARRANTIES
 * OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT
 * LIMITED TO ANY IMPLIED WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE, AND NONINFRINGEMENT OF
 * THIRD PARTY RIGHTS.  THE ENTIRE RISK AS TO THE QUALITY AND
 * PERFORMANCE OF THE SOFTWARE, INCLUDING ANY DUTY TO SUPPORT
 * OR MAINTAIN, BELONGS TO THE LICENSEE.  SHOULD ANY PORTION OF
 * THE SOFTWARE PROVE DEFECTIVE, THE LICENSEE (NOT IBM) ASSUMES
 * THE ENTIRE COST OF ALL SERVICING, REPAIR AND CORRECTION.  IN
 * NO EVENT SHALL IBM BE LIABLE FOR ANY SPECIAL, INDIRECT OR
 * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING
 * FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF
 * CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT
 * OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS
 * SOFTWARE.
 *
 * Author: Carol H. Thompson  IBM Almaden Research Center
 */
/*******************************************************************
*  I/O package for Type 1 font reading
********************************************************************/
 
#ifndef STATIC
#define STATIC static
#endif
 
#if defined(_MSC_VER)
# include <io.h>
# include <sys/types.h>
# include <sys/stat.h>
#else
# include <unistd.h>
#endif
#include <stdio.h>
#include <fcntl.h>
#include <string.h>
#include <ctype.h>
#include <stdlib.h>

#include "t1stdio.h"
#include "t1hdigit.h"

/* we define this to switch to decrypt-debugging mode. The stream of
   decrypted bytes will be written to stdout! This contains binary
   charstring data */
/* #define DEBUG_DECRYPTION */
/* #define DEBUG_PFB_BLOCKS  */

/* Constants and variables used in the decryption */
#define c1 ((unsigned short)52845)
#define c2 ((unsigned short)22719)
static unsigned short r;
static int asc, Decrypt;
static int extrach;
static int haveextrach;

static int starthex80=0;
static long pfbblocklen=0;
static long accu=0;
static unsigned long bytecnt=0;
static int eexec_startOK=0;
static int eexec_endOK=0;
static int in_eexec=0;

 
/* Our single FILE structure and buffer for this package */
STATIC F_FILE TheFile;
STATIC unsigned char TheBuffer[F_BUFSIZ];
 
/* Our routines */
F_FILE *T1Open(), *T1eexec();
int T1Close();
int T1Read(), T1Getc(), T1Ungetc();
void T1io_reset(void);
STATIC int T1Decrypt(), T1Fill();
 
/* -------------------------------------------------------------- */
/*ARGSUSED*/
F_FILE *T1Open(fn, mode)
  char *fn;    /* Pointer to filename */
  char *mode;  /* Pointer to open mode string */
{
  F_FILE *of = &TheFile;
  char c;
  
 
  Decrypt = 0;
  eexec_startOK=0;
  eexec_endOK=0;
 
#ifndef O_BINARY
#  define O_BINARY 0x0
#endif
 
  /* We know we are only reading */
  if ((of->fd=open(fn, O_RDONLY | O_BINARY)) < 0) return NULL;

  /* We check for pfa/pfb file */
  if (read( of->fd, &c, 1)!=1) {
    close( of->fd);
    return(NULL);
  }
  else
    if (c==(char)0x80){
      starthex80=1;
    }
  lseek( of->fd, 0, SEEK_SET);
  
  /* Initialize the buffer information of our file descriptor */
  of->b_base = TheBuffer;
  of->b_size = F_BUFSIZ;
  of->b_ptr = NULL;
  of->b_cnt = 0;
  of->flags = 0;
  of->error = 0;
  haveextrach = 0;
  return &TheFile;
} /* end Open */
 
/* -------------------------------------------------------------- */
int T1Getc(f)        /* Read one character */
  F_FILE *f;         /* Stream descriptor */
{
  if (f->b_base == NULL) return EOF;  /* already closed */
 
  if (f->flags & UNGOTTENC) { /* there is an ungotten c */
    f->flags &= ~UNGOTTENC;
    return (int) f->ungotc;
  }
 
  if (f->b_cnt == 0)  /* Buffer needs to be (re)filled */
    f->b_cnt = T1Fill(f);
  if (f->b_cnt > 0) return (f->b_cnt--, (int) *(f->b_ptr++));
  else {
    f->flags |= FIOEOF;
    return EOF;
  }
} /* end Getc */
 
/*  This function is added by RMz:
    T1Gets(): Read a line of the file and save it to string. At most,
    (size-1) bytes are read. The user *must* ensure (by making size large
    enough) that "eexec" does not get split between two calls because
    in this case, eexec-decryption does not set in.
    ------------------------------------------------------------ */
int T1Gets(char *string,
	   int size,
	   F_FILE *f) /* Read a line */
{
  int i=0;
  char *eexecP;
  
  if (string == NULL) {
    return( i);
  }
  if (f->b_base == NULL)
    return( i);  /* already closed */
  if (size<2)   /* no bytes to be read. For size = 1 we only had
		   room for the \0-character. */
    return( i);
  
  if (f->flags & UNGOTTENC) { /* there is an ungotten c */
    f->flags &= ~UNGOTTENC;
    string[i++]=f->ungotc;
    size--;
  }

  size--; /* we have to leave room for one \0-character */
  
  while ( size>0) {
    if (f->b_cnt == 0) { /* Buffer needs to be (re)filled */
      f->b_cnt = T1Fill(f);
    }
    if (f->b_cnt == 0) { /* no more bytes available. Put \0-char
			    and return. */
      if ( i==0) { /* we did not already store one single char to string */
	f->flags |= FIOEOF;
	return( i);
      }
      else {
	f->flags |= FIOEOF;
	string[i]='\0';
	return( i);
      }
    }

    /* do not skip white space as required by Adobe spec, because
       I have found fonts where the first encrypted byte was of
       white space type. */
    if ( (eexec_startOK==1) && (eexec_endOK==1)) {
      T1eexec( f);
      eexec_startOK=0;
      eexec_endOK=0;
      in_eexec=1;
      /* we are now in the encrypted portion. */
    }
    string[i]=*(f->b_ptr);
    
    /* Check whether eexec appears in the string just setup */
    if ( (Decrypt==0) &&
	 ((eexecP=strstr( string, "eexec"))!=NULL) ) {
      /* if eexec is an isolated token, start decryption */
      if ( (eexec_startOK==1) &&
	   (isspace( (int)string[i])!=0) ) {
	eexec_endOK=1;
      }
      if ( (eexec_startOK==0) &&
	   (isspace( (int)string[i-5])!=0) ) {
	eexec_startOK=1;
      }
    }
    i++;
    /* Under UNIX, '\n' is the accepted newline. For pfb-files it is also
       common to use '\r' as the newline indicator. I have, however, never
       seen a pfb-file which uses the sequence '\r''\n' as a newline
       indicator, as known from DOS. So we don't take care for this case
       and simply map both single characters \r and \n into \n. Of course,
       this can only be done in the ASCII section of the font.

       2002-10-26: Well, life life is teaching me better: There *are* fonts
       out there, ASCII encoded pfa's, that use the crappy DOSian 0x0d 0x0a
       sequence as line separation. In order to make it still work, we absorb
       the byte 0x0a. Failure to do so result in decryption failure. The
       workaround is implemented T1eexec():
       
    */
    if ( *(f->b_ptr)=='\n' || *(f->b_ptr)=='\r') {
      if (in_eexec==0)
	string[i-1]='\n';  
      string[i]='\0'; 
      f->b_cnt--;
      f->b_ptr++;
      return( i);
    }
    
    f->b_cnt--;
    f->b_ptr++;
    size--;
  } /* end of while (size>0) */
  
  string[i]='\0'; /* finish string */
  return( i);
  
} /* end of T1Gets() */



int T1GetDecrypt( void) 
{
  return( in_eexec);
}


/* Return the optional contents after the final cleartomark token.
   There might appear some PostScript code which is not important
   for t1lib, but which becomes important if subsetted fonts are
   embedded in PostScript files. */
int T1GetTrailer(char *string,
		 int size,
		 F_FILE *f)
{
  unsigned long off_save;
  char *buf;
  char *ctmP;
  int i=0, j;
  int datasize;
  int len;
  
  datasize=size;
  
  off_save=lseek( f->fd, 0, SEEK_CUR);
  if ((buf=(char *)malloc( size+1))==NULL ) {
    return( -1);
  }
  lseek( f->fd, -size, SEEK_END);
  read(f->fd, buf, size);
  buf[size]='\0';   /* to be ablo perform a strstr() on this memory */
  
  i=datasize;
  j=datasize-11;   /* length of "cleartomark" plus terminating white
		      space or newline */
   
  while ((j--)>-1) {
    if ((unsigned char)buf[i]==0x80) {
      datasize=i; /* we skip the segment marker of pfb-files */
    }
    if ((ctmP=strstr( &(buf[j]), "cleartomark"))!=NULL) {
      /* buf[i-1] now is the first character after cleartomark. Advance now
	 to the next non white character of EOF. */
      len = datasize - i;
      while ( (isspace( (int)(buf[i-1])) != 0) &&
	      (i < datasize) ) {
	++i;
      }
      memcpy( string, &(buf[i-1]), len);
      string[len]='\0';
      lseek( f->fd, off_save, SEEK_SET);
      free( buf);
      return len;
    }
    i--;
  }
  lseek( f->fd, off_save, SEEK_SET);
  free( buf);
  return( -1);
}



unsigned long T1GetFileSize( F_FILE *f) 
{
  unsigned long off_save;
  unsigned long filesize;
  
  off_save=lseek( f->fd, 0, SEEK_CUR);
  filesize=lseek( f->fd, 0, SEEK_END);
  lseek( f->fd, off_save, SEEK_SET);
  return( filesize);
}



/* -------------------------------------------------------------- */
int T1Ungetc(c, f)   /* Put back one character */
  int c;
  F_FILE *f;         /* Stream descriptor */
{
  if (c != EOF) {
    f->ungotc = c;
    f->flags |= UNGOTTENC;  /* set flag */
    f->flags &= ~FIOEOF;    /* reset EOF */
  }
  return c;
} /* end Ungetc */
 
/* -------------------------------------------------------------- */
int T1Read(buffP, size, n, f)  /* Read n items into caller's buffer */
  char *buffP;       /* Buffer to be filled */
  int   size;        /* Size of each item */
  int   n;           /* Number of items to read */
  F_FILE *f;         /* Stream descriptor */
{
  int bytelen, cnt, i;
  F_char *p = (F_char *)buffP;
  int  icnt;         /* Number of characters to read */
 
  if (f->b_base == NULL) return 0;  /* closed */
  icnt = (size!=1)?n*size:n;  /* Number of bytes we want */
 
  if (f->flags & UNGOTTENC) { /* there is an ungotten c */
    f->flags &= ~UNGOTTENC;
    *(p++) = f->ungotc;
    icnt--; bytelen = 1;
  }
  else bytelen = 0;
 
  while (icnt > 0) {
    /* First use any bytes we have buffered in the stream buffer */
    if ((cnt=f->b_cnt) > 0) {
      if (cnt > icnt) cnt = icnt;
      for (i=0; i<cnt; i++) *(p++) = *(f->b_ptr++);
      f->b_cnt -= cnt;
      icnt -= cnt;
      bytelen += cnt;
    }
 
    if ((icnt == 0) || (f->flags & FIOEOF)) break;
 
    f->b_cnt = T1Fill(f);
  }
  return ((size!=1)?bytelen/size:bytelen);
} /* end Read */
 
/* -------------------------------------------------------------- */
int T1Close(f)       /* Close the file */
  F_FILE *f;         /* Stream descriptor */
{
  if (f->b_base == NULL) return 0;  /* already closed */
  f->b_base = NULL;  /* no valid stream */
  return close(f->fd);
} /* end Close */
 

/* -------------------------------------------------------------- */
F_FILE *T1eexec(f)   /* Initialization */
  F_FILE *f;         /* Stream descriptor */
{
  int i;
  int H;
  
  unsigned char *p;
  int testchar;
  unsigned char randomP[8];
 
  r = 55665;  /* initial key */
  asc = 1;    /* indicate ASCII form */

#ifdef DEBUG_DECRYPTION
  printf("T1eexec(1): first 20 bytes=%.20s, b_cnt=%d\n", f->b_ptr, f->b_cnt);
#endif

  /* As the very first action we check the first byte against 0x0a.
     This mmight happen in context with the T1gets() function for
     pfa files that use DOSian linefeed style. If that character appears
     here, we absorb it (see also T1Gets()!). 
  */
  if ( ( testchar = T1Getc( f)) != 0x0a )
    T1Ungetc( testchar, f);
    
  /* Consume the 4 random bytes, determining if we are also to
     ASCIIDecodeHex as we process our input.  (See pages 63-64
     of the Adobe Type 1 Font Format book.)  */

  /* Skipping over initial white space chars has been removed since
     it could lead to unprocessable pfb-fonts if accindentally the
     first cipher text byte was of the class HWHITE_SPACE.
     Instead, we just read ahead, this should suffice for any
     Type 1 font program. (RMz, 08/02/1998) */

  /* If ASCII, the next 7 chars are guaranteed consecutive */
  randomP[0] = getc(f);  /* store first non white space char */
  fread(randomP+1, 1, 3, f);  /* read 3 more, for a total of 4 */
  /* store first four chars */
  for (i=0,p=randomP; i<4; i++) {  /* Check 4 valid ASCIIEncode chars */
    if (HighHexP[*p++] > LAST_HDIGIT) {  /* non-ASCII byte */
      asc = 0;
      break;
    }
  }
  if (asc) {  /* ASCII form, convert first eight bytes to binary */
    fread(randomP+4, 1, 4, f);  /* Need four more */
    for (i=0,p=randomP; i<4; i++) {  /* Convert */
      H = HighHexP[*p++];
      randomP[i] = H | LowHexP[*p++];
    }
  }
  
  /* Adjust our key */
  for (i=0,p=randomP; i<4; i++) {
    r = (*p++ + r) * c1 + c2;
  }

  /* Decrypt the remaining buffered bytes */
  f->b_cnt = T1Decrypt(f->b_ptr, f->b_cnt);
  Decrypt = 1;
  
#ifdef DEBUG_DECRYPTION
  printf("T1eexec(2): first 120 bytes=%.120s, b_cnt=%d\n", f->b_ptr, f->b_cnt);
#endif
  
  return (feof(f))?NULL:f;
} /* end eexec */
 
/* -------------------------------------------------------------- */
STATIC int T1Decrypt(p, len)
  unsigned char *p;
  int len;
{
  int n;
  int H=0, L=0;
  unsigned char *inp = p;
  unsigned char *tblP;
 
#ifdef DEBUG_DECRYPTION
  printf("T1_Decrypt(): called with len=%d\n",len);
#endif
  if (asc) {
    if (haveextrach) {
      H = extrach;
      tblP = LowHexP;
    }
    else tblP = HighHexP;
    for (n=0; len>0; len--) {
      L = tblP[*inp++];
#ifdef DEBUG_DECRYPTION
      printf("L=0x%X, %d, inp=%c (%d)\n", L,L, *(inp-1), *(inp-1));
#endif
      if (L == HWHITE_SPACE) {
#ifdef DEBUG_DECRYPTION	
	printf("continue\n");
#endif
	continue;
      }
      if (L > LAST_HDIGIT) {
#ifdef DEBUG_DECRYPTION
	printf("L=0x%X, --> break\n", L);
#endif
	break;
      }
      
      if (tblP == HighHexP) { /* Got first hexit value */
        H = L;
        tblP = LowHexP;
      } else { /* Got second hexit value; compute value and store it */
        n++;
        tblP = HighHexP;
        H |= L;
        /* H is an int, 0 <= H <= 255, so all of this will work */
        *p++ = H ^ (r >> 8);
        r = (H + r) * c1 + c2;
      }
    }
    if (tblP != HighHexP) {  /* We had an odd number of hexits */
      extrach = H;
      haveextrach = 1;
    } else haveextrach = 0;
#ifdef DEBUG_DECRYPTION
    printf("T1_Decrypt(): Decrypted %d bytes\n",n);
#endif
    return n;
  } else {
    for (n = len; n>0; n--) {
      H = *inp++;
      *p++ = H ^ (r >> 8);
      r = (H + r) * c1 + c2;
    }
    return len;
  }
} /* end Decrypt */
 
/* -------------------------------------------------------------- */
/* This function has been adapted to support pfb-files with multiple
   data segments */
STATIC int T1Fill(f) /* Refill stream buffer */
  F_FILE *f;         /* Stream descriptor */
{
  int rc,i;
  static unsigned char hdr_buf[6];

  if (starthex80){ /* we have a pfb-file -> be aware of pfb-blocks */
    if ( pfbblocklen-accu >= F_BUFSIZ){
      /* fill the buffer */
      rc = read(f->fd, f->b_base, F_BUFSIZ);
      bytecnt+=rc;
      accu +=rc;
    }
    else{
      if (pfbblocklen-accu>0){
	/* read the remaining of the pfb-block ... */
	rc = read(f->fd, f->b_base, pfbblocklen-accu);
	bytecnt +=rc;
	accu +=rc;
	/* ... and examine the next header */
	i=read(f->fd, hdr_buf, 6);
	bytecnt +=i;
	pfbblocklen=0;
	pfbblocklen += hdr_buf[2]&0xFF  ;
	pfbblocklen += (hdr_buf[3] & 0xFF)  <<8;
	pfbblocklen += (hdr_buf[4] & 0xFF)  <<16;
	pfbblocklen += (hdr_buf[5] & 0xFF)  <<24;
#ifdef DEBUG_PFB_BLOCKS	
	printf("t1io: New segment, length=%d, type=%d\n",
	       pfbblocklen, hdr_buf[1]);
#endif	
	accu=0;
      }
      else{
	/* We are at the beginning of a new block ->
	   examine header */
	i=read(f->fd, hdr_buf, 6);
	pfbblocklen=0;
	pfbblocklen += hdr_buf[2]&0xFF  ;
	pfbblocklen += (hdr_buf[3] & 0xFF)  <<8;
	pfbblocklen += (hdr_buf[4] & 0xFF)  <<16;
	pfbblocklen += (hdr_buf[5] & 0xFF)  <<24;
#ifdef DEBUG_PFB_BLOCKS	
	printf("t1io: New segment, length=%d, type=%d\n",
	       pfbblocklen, hdr_buf[1]);
#endif
	accu=0;
	/* header read, now fill the buffer */
	if (pfbblocklen-accu >= F_BUFSIZ){
	  rc = read(f->fd, f->b_base, F_BUFSIZ);
	  accu +=rc;
	}
	else{
	  /* we have the unusual case that the pfb-block size is
	     shorter than F_BUFSIZ -> Read this block only */
	  rc = read(f->fd, f->b_base, pfbblocklen);
	  accu +=rc;
	}
      }
    }
  }
  else{
    /* We have a pfa-file -> read straight ahead and fill buffer */
    rc = read(f->fd, f->b_base, F_BUFSIZ);
  }
  
  /* propagate any error or eof to current file */
  if (rc <= 0) {
    if (rc == 0)    /* means EOF */
      f->flags |= FIOEOF;
    else {
      f->error = (short)-rc;
      f->flags |= FIOERROR;
      rc = 0;
    }
  }

  f->b_ptr = f->b_base;
#ifdef DEBUG_DECRYPTION
  printf("T1_Fill(): read %d bytes\n", rc);
#endif
  
  if (Decrypt){
    rc = T1Decrypt(f->b_base, rc);
#ifdef DEBUG_DECRYPTION
    printf("T1_Fill(): decrypted %d bytes\n", rc);
#endif
  }
  
  return rc;
} /* end Fill */


void T1io_reset(void)
{
  pfbblocklen=0;
  accu=0;
  starthex80=0;
  eexec_startOK=0;
  eexec_endOK=0;
  in_eexec=0;
}