summaryrefslogtreecommitdiff
path: root/support/try/try.nw
blob: 0f173bd654978de8291426bfe4e1252d21b16e49 (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

%
% filename: try.nw
% version: 4.0
% date: 2013-2014
% author: Ajabu Tex <ajabutex@gmail.com>
% description: Automation tool for TeX documents.
%
% This file is part of TrY.
% TrY is free software: you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published by
% the Free Software Foundation, either version 3 of the License, or
% (at your option) any later version.
%
% TrY is distributed in the hope that it will be useful,
% but WITHOUT ANY WARRANTY; without even the implied warranty of
% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
% GNU General Public License for more details.
%
% You should have received a copy of the GNU General Public License
% along with TrY.  If not, see <http://www.gnu.org/licenses/>.
%


\let\PART\part
\def\part{\par\vfill\eject\PART}
\def\TrY{{\sl TrY}}


\title{TrY}
\author{Ajabu Tex\\ajabutex@gmail.com}
\date{January 2nd, 2014}
\maketitle
\tableofcontents

@
\part{Preliminaries}
\section{Introduction} This document describes the algorithms needed
to implement a system for the automation of the compilation of (La)\TeX\
documents.

The characteristics of the implementation allow to use the program for
the automation of other procedures, such as the compilation of files
of {\it Literate Programming}, and in all cases in which a sequence of
[[bash]] commands is needed to process a text file.

The language used is {\sl Python}, chosen for its expressive power,
stability, large endowment of standard libraries and the fact to be
natively installed on most Linux systems.


\section{Scope of use}
There are already some great programs aimed to automate the
compilation of \TeX\ and \LaTeX\ documents: {\sl latexmk} by John
Collins (http://www.phys.psu.e\-du/\textasciitilde
collins/software/latexmk-jcc/), {\sl Rubber} by Emmanuel Beffara
(http://la\-unchpad.net/rubber), {\sl arara} by Paulo Cereda
(http://cereda.github.io/arara/) and many others; you can find some of
them on CTAN (http://www.ctan.org/to\/-pic/compilation).

{\sl arara} is the program from which I took inspiration for the
creation of \TrY. I really like the idea behind the implementation of
{\sl arara}, namely that of making explicit in the document the list
of commands needed to compile it.

{\sl arara} is a nice piece of software but has some limitations that
make it useless for my work, subtle limits relating to the handling of
the Linux filters, pipes and subshells.

On the other hand \TrY\ works only on Linux (and perhaps Mac OS), but
on Linux practically has no limitation. \TrY\ can do anything that can
be done in a Linux terminal. In practice the \TrY\ commands are
nothing but [[bash]] commands inserted in the comment lines of a \TeX\
document.

Last but not least, \TrY\ could not be simpler than it is.


\section{How to use}
\subsection{Installation}
Be sure that the file [[try]] is executable and then copy
it in the [[/usr/bin]] directory with the following commands:
\begin{quote}
[[chmod +x try
sudo cp try /usr/bin/try]]
\end{quote}

\subsection{Mechanics}
\TrY\ scans the document looking for \TrY\ statements into the
comment lines and executes them one after the other.

The typical form of the \TrY\ commands is the following:

\begin{quote}
[[%$ <command> <parameter> <parameter> ...]]
\end{quote}

\begin{enumerate}
\item The first element of the line must be the comment symbol
([[%]]);
\item immediately followed by the dollar sign ([[$]]);
\item then follows a space;
\item finally, a Linux command with its parameters.
\end{enumerate}

So let's say you are writing a \LaTeX\ document named [[nicedoc.tex]];
if you put at the beginning of the file the following line of comment:

\begin{quote}
[[%
%$ lualatex --enable-write18 nicedoc.tex
%]]
\end{quote}

you can simply compile the document with the following command line:

\begin{quote}
[[try nicedoc.tex]]
\end{quote}

As an added benefit, anyone who will read your manuscript
will know that the required engine for the compilation is Lua\LaTeX\
and that is required the activation of the [[write18]] option.

In a \TrY\ statement you can give instructions to process any file,
but if the file name passed to the command as a parameter is the same
of the name of the document itself, the file name can be replaced by
the placeholder `[[$0]]'.

So, in the example above, you can use into the manuscript the
following statement:

\begin{quote}
[[%
%$ lualatex --enable-write18 $0
%]]
\end{quote}

that is a little bit more compact.

Another example: you are writing [[bigdoc.tex]], a complex \LaTeX\
document, with management of many different fonts, extensive indexes
and some parts computed on-the-fly by python. The typical cicle of
compilation consists of several bash commands. But you can type
those commands at the beginning of the file once and forever as \TrY\
statements:

\begin{quote}
[[%
%$ xelatex -shell-escape $0
%$ bibtex -min-crossref=3 bigdoc.aux
%$ xelatex -shell-escape $0
%$ xelatex -shell-escape $0
%]]
\end{quote}

and let \TrY\ do the hard work for you. The nice thing is that this
way the document show explicitly the engine needed, the number of
compilations required and so on, making easier the collaborative
editing of the document.

With \TrY\ you can also process other files than \TeX\ or \LaTeX\
documents.  In this case you have to tell the program
what is the comment symbol used, using the statement
[[$trycommentchar=<a_comment_char>$]]. For example:

\begin{quote}
[[#
# Sample python listing
#
# $trycommentchar=#$
#
#$ python -B $0 -W message
#]]
\end{quote}

As a last example, you can put at the beginning of a [[noweb]] file of
\textit{Literate Programming} the following statements. Let's say that
the name of the file is [[niceprog.nw]]:

\begin{quote}
\begin{verbatim}
%
% extraction and compilation of the documentation
%$ noweave -index -latex $0 > niceprog-doc.tex
%$ pdflatex niceprog-doc.tex
%$ pdflatex niceprog-doc.tex
%$ pdflatex niceprog-doc.tex
%
% extraction and compilation of the program
%$ notangle -Rniceprog $0 > niceprog.pas
%$ fpc -Fudirectory -gh niceprog.pas
%
\end{verbatim}
\end{quote}

Now you have only to launch \TrY\ in a terminal with the following
command line

\begin{quote}
[[try niceprog.nw]]
\end{quote}

to obtain the documentation, the listing and the executable file in
just a single step.

\subsection{The command line}
To process a document with \TrY\ open a terminal and use the following
command line:

\begin{quote}
\begin{verbatim}
$ try [file [--log] [--safe] [--verbose] | --help | --version]
$ try [file [-l] [-s] [-v] | -h | -V]
\end{verbatim}
\end{quote}

\begin{description}
\item[[[file]]] is the name of the document, including the extension,
that you want to process;
\item[[[-h | --help]]] print a help message and exits;
\item[[[-s | --safe]]] print the list of commands before to execute
them;
\item[[[-l | --log]]] produce a log file with the operations carried
out by the program;
\item[[[-v | --verbose]]] print the output of the program;
\item[[[-V | --version]]] output version information and exit.
\end{description}

If you use a programmable text editor, it can be set for use with
\TrY.  For example, to create a command that launches \TrY\ from
GEdit you can do this way:

\begin{enumerate}
\item menu [[Tools -> Manage externals tools]]
\item list [[All Languages]]. Click on [[Add a new tool]]. Name
it [[Compile TrY]]
\item In the [[Edit]] pane write:
\begin{quote}
[[#!/bin/bash
try $GEDIT_CURRENT_DOCUMENT_NAME --log --verbose
]]
\end{quote}
\item [[Shortcut Key -> Shift+Ctrl+%]] (or whatever you want)
\item [[Save -> Current document]]
\item [[Input -> Nothing]]
\item [[Output -> Display in bottom pane]]
\item [[Applicability -> All documents -> All Languages]]
\end{enumerate}

Now you can start building with \TrY\ the document you are working on
with GEdit simply by pressing the key combination [[Shift+Ctrl+%]].

@
\part{The code}
Here is an outline of the program:
<<try>>=
<<shabang>>
<<license statements>>
<<standard libraries>>
<<global constants>>
<<procedure to notify messages on the screen and in the logfile>>
<<procedure to get the list of command-line arguments>>
<<procedures to collect the TrY statements>>
<<procedure for the request for a confirmation>>
<<procedure for the execution of the TrY statements>>
<<main module>>
@
The main module collects the arguments in the variable [[args]] from
the command line; if a filename is passed it scans the document to
find lines of comment containing \TrY\ statements that are then stored
in the variable [[try_statements]]. At last executes the \TrY\
statements itself, but only if has a confirmation from the user. Every
step of this process is notified on the screen and in the logfile if
the proper parameters are passed to the program via the command line.
<<standard libraries>>=
import sys
<<main module>>=
if __name__ == "__main__":
    args = get_args()
    if not args.filename =='':
        notify('This is ' + prog_name + ' ver. ' + prog_version + \
            ' revision ' + prog_revisiondate + '\n', args)
        try_statements = get_trystatements_from(args)
        if len(try_statements) == 0:
            notify('No commands found in ' + args.filename + '.', args)
            sys.exit('Nothing to execute.')
        else:
            if args.safe:
                if execution_confirmed_of(try_statements):
                    execute_statements(try_statements, args)
                else:
                    notify('Execution stopped by the user.', args)
                    sys.exit('Execution aborted.')
            else:
                execute_statements(try_statements, args)

@
Now we have to build every single command called by the main module.
First of all we have to define some global constants, namely the
strings used in the output of the program.
<<global constants>>=
prog_name = 'TrY'
prog_cmd = 'try'
prog_version = '4.0'
prog_author = 'Ajabu Tex'
prog_revisiondate = 'Jan 2, 2014'
prog_releasedate = '2014'
prog_license = prog_name + ' ' + prog_version + '\n' + \
    'Copyright 2013-' + prog_releasedate + ' ' + prog_author + '\n' + \
    'There is NO warranty.  Redistribution of this software is\n' + \
    'covered by the terms of both the ' + prog_name + ' copyright and\n' + \
    'the GNU General Public License.\n' + \
    'For more information about these matters, see the file\n' + \
    'named COPYING and the ' + prog_name + ' source.\n' + \
    'Author of ' + prog_name + ': ' + prog_author + '.'

@ %def prog_name prog_cmd prog_version prog_author
@ %def prog_revisiondate prog_releasedate prog_license
@
The function [[append_to_logfile()]] takes as input a file name and a
string. If the string is that opening the file then the file is
overwritten; otherwise is opened in `append' mode. Then the function
calculates the day and time, join them to the string passed as
argument and writhe this new string into the log file. The name of the
log file is the same name of the file to process --included the
extension-- plus the extension [[.trylog]]. Example: if I call the
program with the following command line

\begin{quote}
[[try -l niceprog.nw]]
\end{quote}

the name of the log file will be [[niceprog.nw.trylog]].
<<standard libraries>>=
from datetime import datetime
<<procedure to notify messages on the screen and in the logfile>>=
def append_to_logfile(a_file_name, a_string_line):
    if a_string_line.startswith('This is ' + prog_name + \
    ' ver. ' + prog_version + ' revision ' + prog_revisiondate):
        openmode = 'w'
    else:
        openmode = 'a'
    now = datetime.now()
    time_string = now.strftime("%d %b %Y %H:%M:%S.%f")
    log_string = time_string + ' ' + a_string_line
    logfile_name = a_file_name + '.trylog'
    logfile = open(logfile_name, openmode)
    logfile.write(log_string + '\n')
    logfile.close

@ %def append_to_logfile
@
Every time the program has to send something to the screen or to the
logfile, it calls the function [[notify()]] passing to it the text to
show and the arguments passed to the program by the command line.
<<procedure to notify messages on the screen and in the logfile>>=
def notify(text, args):
    if args.log:
        append_to_logfile(args.filename, text)
    if args.verbose:
        print (text)

@ %def notify
@
The first operation the program has to do is to collect the parameters
from the command line. This issue is exploited with the standard
library [[argparse]]. The program can be called with the following
parameters:

\begin{quote}
\begin{description}
  \item[log] send an output in a log file;
  \item[safe] safe mode: before to execute the list of statements
  found in the processed file prints them out on the screen and asks
  the user for a confirmation;
  \item[verbose] outputs on the screen the stages of processing;
  \item[version] print on the screen the version number and copyright
  information;
  \item[file] the name of the file to process;
\end{description}
\end{quote}

where [[file]] is a positional parameter and the others are optional
parameters. So we have the following collection of parameters,
returned by the function [[get_args]] as a list. The function
[[get_args]] collect the command line parameters and manage all the
stage pertaining to the command line itself. If there are no
parameters prints a help screen; if the program is called with the
[[--version]] parameter prints a screen with the version informations.

To get useful documentation from the python help system type in a
terminal:

\begin{quote}
\begin{verbatim}
$ pydoc argparse
$ pydoc argparse.ArgumentParser
\end{verbatim}
\end{quote}
<<standard libraries>>=
import argparse
<<procedure to get the list of command-line arguments>>=
def get_args():
    parser = argparse.ArgumentParser(
        prog=prog_cmd,
        usage='%(prog)s [options] filename',
        formatter_class=argparse.RawDescriptionHelpFormatter,
        description='''\
            ''' + prog_name + '''
            TeX automation tool''',
        epilog='(C) 2013-' + prog_releasedate + ' ' + prog_author)
    parser.add_argument('-l', '--log',
        action='store_true',
        help='generate a log output')
    parser.add_argument('-s' ,'--safe',
        action='store_true',
        help='print the list of commands prior to execute them')
    parser.add_argument('-v', '--verbose',
        action='store_true',
        help='print the command output')
    parser.add_argument('-V', '--version',
        action='version',
    version=prog_license)
    parser.add_argument('filename', nargs='?', default='')
    args = parser.parse_args()
    if args.filename == '':
        parser.print_help()
    return args

@ %def get_args
@
Now the variable [[args]] contain a list whith the name of the
file to process and the parameters to apply.

The extraction and the collection of the \TrY\ statements from a file
goes through three stages: first we have to know what is the comment
[[string.rpartition()]]
flag used in the file, then we have to extract from the file all the
lines that are TrY statements, at last we must to return that
statements as a list of string ready to be passed to the procedure of
execution.
<<procedures to collect the TrY statements>>=
<<procedure for the recognition of the char used as comment flag>>
<<procedure to extract TrY statements from a list of strings>>
<<procedure to return the collected statements>>
@
[[get_commentchar_from()]] takes as input a list of lines of text
and returns the default comment character (`[[%]]') or the one
defined by the statement [[$trycommentchar=...$]] that may be present.

The splitting of the statement is a bit tricky: the function
[[string.rpartition(sep)]] search for the separator sep in a string,
starting at the end of the string, and return a list containing the
part before it, the separator itself, and the part after it.

In our case the ipothetyc statement [["% $trycommentchar=x$"]] is
splitted in the list [[["% ", "$trycommentchar=", "x$"]]]. We take the
last element and split it again with the function
[[string.rsplit([sep]), that return a list of the words in the string,
using sep as the delimiter string, starting at the end of the string
and working to the front. In this way we obtain the list [[["x",
"$"]]], where the first element is the one we were looking for.

For the python documentation:

\begin{quote}
\begin{verbatim}
$ pydoc str.rpartition
$ pydoc str.rsplit
\end{verbatim}
\end{quote}
<<procedure for the recognition of the char used as comment flag>>=
def get_commentchar_from(str_list):
    result = '%'
    for str_line in str_list:
        if '$trycommentchar=' in str_line:
            str_line = str_line.rpartition('$trycommentchar=')[2]
            str_line = str_line.rsplit('$')[0]
            result = str_line
    return result

@ %def get_commentchar_from
@
The function [[get_trystatement_list_from()]] takes as input a list
of lines of text, the list of arguments and the comment flag.
Returns a list of \TrY\ stateents, which is a list of bash commands
that are present on the lines that begin with the string [[flag]],
in which has been replaced any occurrence of the string `[[$0]]'
with the name of the file to process.

A useful python documentation is in:

\begin{quote}
[[$ pydoc str.partition]]
\end{quote}
<<procedure to extract TrY statements from a list of strings>>=
def get_trystatement_list_from(str_list, args, flag):
    file_name = args.filename
    result = []
    for str_line in str_list:
        if str_line.startswith(flag):
            if '$0' in str_line:
                str_line = str_line.replace('$0', file_name)
            str_line = str_line.strip()
            str_line = str_line.partition(flag)[2]
            notify('Found command ' + str_line, args)
            result.append(str_line)
    return result

@ %def get_trystatement_list_from
@
[[get_trystatements_from()]] takes in input the list of arguments
passed to the program via the command line. Open the file and extract
the whole its content as a list of strings. Those strings are filtered
through the procedure [[get_commentchar_from()]] and
[[get_trystatement_list_from()]] to obtain a list of bash commands to
pass to the procedure of execution.
<<procedure to return the collected statements>>=
def get_trystatements_from(args):
    file_name = args.filename
    notify('Looking for ' + prog_name + ' commands in ' + file_name, args)
    f = open(file_name, 'r')
    str_lines = f.readlines()
    f.close()
    comment_char = get_commentchar_from(str_lines)
    notify('Comment char used: ' + comment_char, args)
    try_statement_flag = comment_char + '$ '
    result = get_trystatement_list_from(str_lines,
        args, try_statement_flag)
    return result

@ %def get_trystatements_from
@
[[execution_confirmed_of]] takes the list of statements found into the
processed file, print them out on the screen and ask the user a
confirmation about the execution. The function returns a boolean
value.
<<procedure for the request for a confirmation>>=
def execution_confirmed_of(str_lst):
    prompt = 'These are the commands that will be processed:\n'
    for str_line in str_lst:
        prompt += str_line + '\n'
    prompt += '\n\nDo you want to continue? [Yes/No]\n'
    answer = raw_input(prompt)
    result = answer in ['Y', 'y', 'Yes', 'yes']
    return result

@ %def execution_confirmed_of
@
[[execute_statements()]] takes as input a list of strings representing
Unix commands and passes them to the operating system for the
execution.  If there's an error the execution is aborted.

Useful python documentation:

\begin{quote}
\begin{verbatim}
$ pydoc subprocess
$ pydoc subprocess.call
\end{verbatim}
\end{quote}
<<standard libraries>>=
import os
import subprocess

<<procedure for the execution of the TrY statements>>=
def execute_statements(try_statements, args):
    if len(try_statements) == 0:
        notify('No commands found.', args)
        sys.exit()
    else:
        for statement in try_statements:
            notify('Executing instruction: ' + statement, args)
            exit_status = subprocess.call([statement],
                stdout=subprocess.PIPE, shell=True)
            if exit_status != 0:
                notify(prog_name + ': error at system level. \n' + \
                    'Execution aborted with exit status ' + \
                    str(exit_status) + '\n', args)
                sys.exit(exit_status)
            else:
                notify('SUCCESS', args)

@ %def execute_statements
@
That's all. At the end here are listed the shabang and the usual
declarations about license and copyright, written as stated by the GNU
GPL web page: http://www.gnu.org/licenses/gpl-howto.en.html
<<shabang>>=
#!/usr/bin/env python

<<license statements>>=
# Copyright 2013-2014 Ajabu Tex
#
# This file is part of TrY.
#
# TrY is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# TrY is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with TrY.  If not, see <http://www.gnu.org/licenses/>.

@
\part{What's new in version 4}
The code was almost all rewritten to implement two main new features:

\textit{the safe mode:}\ before of executing the instructions found in
the processed file \TrY\ prints them out on the screen and asks for a
confirmation from the user;

\textit{the command line:}\ now is implemented with the [[argparse]]
library. The code is more solid and robust, easily readable and
maintenable;

I've also rewritten almost all the documentation, now much more in a
\textit{Literate Programming} style.

At last the terms of license are no more those of LPPL but GNU GPL --
General Public License.

@
\part{Credits and licence}
\begingroup\parindent0pt\obeylines
Program: TrY
Version: 4.0
Release: 2013-2014
Author: Ajabu Tex \texttt{$<$ajabutex@gmail.com$>$}
Description: Automation tool for TeX documents
Terms of use: GNU General Public License.
\endgroup\par\medskip

The project repository is hosted on Bitbucket; feel free to contribute
in any way. If you want to fork the project or want to send a pull
request, or simply want to download the latest version, go to
\begin{quote}
[[https://bitbucket.org/ajabutex/try]]
\end{quote}

The program is also kindly hosted as a package on CTAN:
\begin{quote}
[[http://www.ctan.org/pkg/try]]
\end{quote}

To obtain the program code and the documentation the source file
[[try.nw]] was compiled with [[noweb]], a simple tool for {\it
Literate Programming} written by Norman Ramsey. You can find
informations about it at:
\begin{quote}
[[
http://www.cs.tufts.edu/~nr/noweb/
http://www.ctan.org/tex-archive/web/noweb
]]
\end{quote}

@
\part{Indexes}
\section{Chunks}
\nowebchunks
\section{Identifiers}
\nowebindex