1: #!/usr/bin/perl
2:
3: # The LearningOnline Network with CAPA
4: # Converts a texxml file into a single tex file
5: #
6: # Copyright Michigan State University Board of Trustees
7: #
8: # This file is part of the LearningOnline Network with CAPA (LON-CAPA).
9: #
10: # LON-CAPA is free software; you can redistribute it and/or modify
11: # it under the terms of the GNU General Public License as published by
12: # the Free Software Foundation; either version 2 of the License, or
13: # (at your option) any later version.
14: #
15: # LON-CAPA is distributed in the hope that it will be useful,
16: # but WITHOUT ANY WARRANTY; without even the implied warranty of
17: # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18: # GNU General Public License for more details.
19: #
20: # You should have received a copy of the GNU General Public License
21: # along with LON-CAPA; if not, write to the Free Software
22: # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
23: #
24: # /home/httpd/html/adm/gpl.txt
25: #
26: # http://www.lon-capa.org/
27: #
28: # 7-16-2002 Jeremy Bowers
29:
30: use strict;
31: use HTML::TokeParser;
32: use GDBM_File;
33: use File::Temp;
34:
35: # accept texxml document on standard in
36: my $p = HTML::TokeParser->new( $ARGV[0] );
37: my $dirprefix = "../../loncom/html/adm/help/tex/";
38:
39: # Make myself a temp dir for processing POD
40: my $tmpdir = File::Temp::tempdir('loncapahelpgenXXXXXXX', TMPDIR => 1);
41:
42: # Print the header
43: open (LATEX_FILE, $dirprefix . "Latex_Header.tex");
44: print <LATEX_FILE>;
45:
46: while (my $token = $p->get_token())
47: {
48: my $type = $token->[0];
49: if ($type eq 'S') {
50: my $tag = $token->[1];
51: my $attr = $token->[2];
52: if ($tag eq 'section') {
53: my $title = $attr->{'name'};
54: print "\\section{$title}\n\n";
55: }
56:
57: if ($tag eq 'subsection') {
58: my $title = $attr->{'name'};
59: print "\\subsection{$title}\n\n";
60: }
61:
62: if ($tag eq 'subsubsection') {
63: my $title = $attr->{'name'};
64: print "\\subsubsection{$title}\n\n";
65: }
66:
67: if ($tag eq 'file') {
68: my $file = $attr->{'name'};
69: open (LATEX_FILE, $dirprefix . $file);
70: print <LATEX_FILE>;
71: print "\n\n";
72: }
73:
74: if ($tag eq 'tex') {
75: print "\n\n";
76: print $attr->{'content'};
77: print "\n\n";
78: }
79:
80: if ($tag eq 'pod') {
81: my $file = $attr->{'file'};
82: my $section = $attr->{'section'};
83: if (!defined($section)) { $section = ''; }
84: else {
85: $section = "-section '$section'";
86: }
87: my $h1level = $attr->{'h1level'};
88: if (!defined($h1level)) { $h1level = '2'; }
89: $file = '../../loncom/' . $file;
90: my $filename = substr($file, rindex($file, '/') + 1);
91: system ("cp $file $tmpdir\n");
92: system ("cd $tmpdir; pod2latex -h1level $h1level $section $filename\n");
93: my $latexFile = substr($filename, 0, rindex($filename, '.')) . '.tex';
94: open LATEX_FILE, $tmpdir . '/' . $latexFile;
95: # pod2latex inserts \labels and \indexs for every section,
96: # which is horrible because the section names tend to get
97: # reused a lot. This filters those out, so we need to do
98: # create our own indexes.
99: for (<LATEX_FILE>) {
100: $_ =~ s/\\([^{]*)(section|paragraph)(\*?)\{([^\\]+)\\label\{[^\\]+\}\\index\{([^\\]+)\}\}/\\\1\2\3\{\4\}/g;
101: print $_;
102: }
103: print "\n\n";
104: }
105: }
106: }
107:
108: # Print out the footer.
109: open (LATEX_FILE, $dirprefix . "Latex_Footer.tex");
110: print <LATEX_FILE>;
111:
112: # Remove the temp directory
113: system ("rm -rf $tmpdir");
114:
115: __END__
116:
117: =pod
118:
119: =head1 NAME
120:
121: texxml2latex.pl - core script that drives the help file assembly
122: applications
123:
124: =head1 SYNOPSIS
125:
126: LON-CAPA's help system is based on assembling various pieces into
127: LaTeX files for conversion into printed documents. The various pieces
128: can also be used as online help.
129:
130: =head1 OVERVIEW
131:
132: X<help system, overview>LON-CAPA's help system is based on the idea of
133: assembling various pieces as needed to create documents for printing,
134: and using these various pieces for online help. LaTeX is the primary
135: language of the help system, because we can easily convert it to HTML,
136: and it makes the nicest printed documents.
137:
138: The scripts for the help system are stored in /docs/help in the CVS
139: repository.
140:
141: =head2 Data Sources
142:
143: The help system can draw from the following sources to create help
144: documents:
145:
146: =over 4
147:
148: =item * B<LaTeX fragments>: LaTeX fragments stored in
149: C</loncom/html/adm/help/tex> in the CVS repository (which end up in
150: C</home/httpd/html/adm/help/tex>). A "LaTeX fragment" is a file that
151: contains LaTeX-style markup, but is not a complete LaTeX file with
152: header and footer.
153:
154: =item * B<perl POD documentation>: POD documentation may be extracted
155: from perl modules used in LON-CAPA, using the syntax described in
156: podselect's man page.
157:
158: =back
159:
160: =head2 Online Help
161:
162: The online aspect of the help system is covered in the documentation
163: for loncommon.pm; see L<Apache::loncommon>, look for
164: C<help_open_topic>.
165:
166: Online help can only come from LaTeX fragments.
167:
168: Access to the printed documents is partially provided online by
169: rendering the help files structure in a way that allows the user to
170: click through to the underlying help files; see
171: L<http://msu.loncapa.org/adm/help/author.manual.access.hlp> for an
172: example. It's not very good, but it's marginally better then nothing.
173:
174: =head2 Offline Documents
175:
176: Offline documents are generated from XML documents which tell a
177: rendering script how to assemble the various LaTeX fragments into a
178: single LaTeX file, which is then rendered into PostScript and PDF
179: files, suitable for download and printing.
180:
181: =head1 texxml And Rendering texxml
182:
183: =head2 texxml
184:
185: X<texxml>
186: texxml is a little XML file format used to specify to the texxml2*.pl
187: scripts how to assemble the input sources into LaTeX documents. texxml
188: files end in the .texxml extension, and there is one texxml file per
189: final rendered document.
190:
191: The texxml format is as follows: There is a root <texxml> element,
192: with no attributes and the following children:
193:
194: =over 4
195:
196: =item * B<title>: The B<name> attribute of this tag is used as the
197: title of the document in texxml2index.pl; it is ignored in
198: texxml2latex.pl. If you don't intend to offer online-access
199: to the rendered documents this may be skipped.
200:
201: =item * B<section>, B<subsection>, and B<subsubsection>: These create
202: the corresponding environments in the output file. The B<name>
203: attribute is used to determine the name of the section.
204:
205: =item * B<file>: The C<name> attribute specifies a LaTeX fragment by
206: filename. The file is assumed to be located in the
207: C<loncom/html/adm/help/tex/> directory in the CVS repository. The
208: C<.tex> is required.
209:
210: =item * B<tex>: The contents of the B<content> attribute are directly
211: inserted into the rendered LaTeX file, followed by a paragraph
212: break. This is generally used for little connective paragraphs in
213: the documentation that don't make sense in the online help. See
214: C<author.manual.texxml> for several example usages.
215:
216: =item * B<pod>: The B<file> attribute specified a file to draw the POD
217: documentation out of. The B<section> attribute is a section
218: specification matching the format specified in the man page of
219: podselect. By default, all POD will be included. The file is
220: assumed to be relative to the C<loncom> directory in the CVS
221: repository; you are allowed to escape from that with .. if
222: necessary. The B<h1level> attribute can be used to change
223: the default depth of the headings; by default, this is set to 2,
224: which makes =head1 a "subsection". Setting this higher can allow
225: you to bundle several related pod files together; see
226: developer.manual.texxml for examples.
227:
228: =back
229:
230: texxml2latex.pl will automatically include C<Latex_Header.tex> at the
231: beginning and C<Latex_Footer.tex> at the end, to make a complete
232: document LaTeX document.
233:
234: =head2 Rendering texxml X<texxml, rendering>
235:
236: =head3 render.texxml.pl X<render.texxml.pl>
237:
238: The C<render.texxml.pl> script takes a .texxml file, and produces
239: PostScript and PDF files. The LaTeX files will be given access to .eps
240: files in the C</loncom/html/adm/help/eps/> directory while
241: rendering. Call it as follows, from the C<doc/help> directory:
242:
243: perl render.texxml.pl -- author.manual.texxml
244:
245: substituting the appropriate texxml file.
246:
247: =head3 texxml2latex.pl X<texxml2latex.pl>
248:
249: texxml2latex.pl is a perl script that takes texxml in and assembles
250: the final LaTeX file, outputting it on stout. Invoke it as follows:
251:
252: perl texxml2latex.pl author.manual.texx
253:
254: Note that there is no error handling; if the script can not find a
255: .tex file, it is simply ignored. Generally, if a file is not in the
256: final render, it either could not be found, or you do not have
257: sufficient permissions with the current user to read it.
258:
259: =head3 texxml2index.pl X<texxml2index.pl>
260:
261: texxml2index.pl is a perl script that takes texxml in and assembles a
262: file that can be used online to access all the .tex files that are
263: specified in the .texxml file. For an example of how this looks
264: online, see
265: C<http://msu.loncapa.org/adm/help/author.manual.access.hlp>.
266:
267: =head2 texxml support
268:
269: There are a couple of scripts that you may find useful for creating
270: texxml-based help:
271:
272: =head3 latexSplitter.py X<latexSplitter.py>
273:
274: latexSplitter.py is a Python script that helps you seperate a
275: monolithic .tex file into the small pieces LON-CAPA's help system
276: expects. Invoke it like this:
277:
278: python latexSplitter.py monolithic.tex
279:
280: where C<monolithic.tex> is the .tex file you want to split into
281: pieces. This requires Python 2.1 or greater (2.0 may work); on many
282: modern RedHat installs this is installed by default under the
283: executable name C<python2>.
284:
285: Use the program by highlighting the desired section, give it a file
286: name in the textbox near the bottom, and hit the bottom button. The
287: program will remove that text from the textbox, and create a file in
288: the C<loncom/html/adm/help/tex/> directory containing that LaTeX. For
289: consistency, you should use underscores rather then spaces in the
290: filename, and note there are a few naming conventions for the .tex
291: files, which you can see just by listing the
292: C<loncom/html/adm/help/tex/> directory.
293:
294: The idea behind this program is that if you are writing a big document
295: from scratch, you can use a "real" program like LyX to create the .tex
296: file, then easily split it with this program.
297:
298: =head3 simpleEdit.py X<simpleEdit.py>
299:
300: simpleEdit.py is a python script that takes a .texxml file and shows
301: all the tex files that went into in sequence, allowing you to "edit"
302: the entire document as one entity. Note this is intended for simple
303: typo corrections and such in context, not major modification of the
304: document. Invoke it with
305:
306: python simpleEdit.py author.manual.texxml
307:
308: Make your changes, and hit the "Save" button to save them.
309:
310: =head2 texxml LaTeX Feature Support
311:
312: =head3 Cross-referencing
313:
314: LaTeX has a cross-referencing system build around labeling points in
315: the document with \label, and referencing those labels with \ref. In a
316: complete LaTeX document, there's no problem because all \refs and
317: \labels are present. However, for the online help, \ref'ing something
318: that is not in the current LaTeX fragment causes a TTH error when it
319: can't find the crossreference.
320:
321: The solution is to do the cross-references for TTH. When LON-CAPA is
322: installed, the C<rebuildLabelHahs.pl>X<rebuildLabelHash.pl> script
323: is executed, which extracts all the labels from the LaTeX fragments
324: and stores them in the C<fragmentLabels.gdbm>X<fragmentLabels.gdbm> hash.
325: The C<lonhelp.pm> handler then replaces \refs with appropriate
326: HTML to provide a link to the referenced help file while online. Thus,
327: you can freely use references, even in online help.
328:
329: =head3 Indexing
330:
331: LaTeX has a popular index making package called MakeIndex. LON-CAPA's
332: help system supports this, so you can create indices using the \index
333: LaTeX command. In perl POD files, use the X command. Note that in both
334: cases the index text is not included in the render, so you need to
335: specify the exact index.
336:
337: =head1 Writing POD: Style
338:
339: Adopting a little bit from everybody who has included POD in their
340: documents to date, the help system is going to expect the following
341: format for POD documentation.
342:
343: The POD should start with a C<=head1> with the title C<NAME> (in caps
344: as shown). The following paragraph should extremely briefly describe
345: what the module does and contains. Example:
346:
347: =head1 NAME
348:
349: Apache::lonflunkstudent - provides interface to set all
350: student assessments point score to 0
351:
352: Next should be a C<head1> titled C<SYNOPSIS> which contains a
353: paragraph or two description of the module.
354:
355: =head1 SYNOPSIS
356:
357: lonflunkstudent provides a handler to select a student and set all
358: assignment values to zero, thereby flunking the student.
359:
360: Routines for setting all assessments to some value are provided by
361: this module, as well as some useful student taunting routines.
362:
363: Optionally, an C<OVERVIEW> section can be included. This can then be
364: extracted by the help system for the LON-CAPA subsystems overview
365: chapter. The overview should be a relatively high-level, but still
366: technical, overview of the module, sufficient to give the reader
367: enough context to understand what the module does, what it might be
368: useful for in other contexts, and what is going on in the code when it
369: is read.
370:
371: The remainder should be formatted as appropriate for the file, such
372: that discarding the NAME, SYNOPSIS, and OVERVIEW sections provides a
373: useful API overview of the module.
374:
375: Routines that are private to the module should B<not> be documented;
376: document them in perl comments, or, as is the style of the time, not
377: at all, as is appropriate.
378:
379: Method and function names should be bolded when being
380: documented. Indexing should be done as appropriate, using the X
381: perldoc command. Literal string such as filename should be enclosed in
382: the C command, like this: C</home/httpd/lonTabs/>.
383:
384: =cut
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>