1#! /usr/bin/env python
2
3# Convert GNU texinfo files into HTML, one file per node.
4# Based on Texinfo 2.14.
5# Usage: texi2html [-d] [-d] [-c] inputfile outputdirectory
6# The input file must be a complete texinfo file, e.g. emacs.texi.
7# This creates many files (one per info node) in the output directory,
8# overwriting existing files of the same name.  All files created have
9# ".html" as their extension.
10
11
12# XXX To do:
13# - handle @comment*** correctly
14# - handle @xref {some words} correctly
15# - handle @ftable correctly (items aren't indexed?)
16# - handle @itemx properly
17# - handle @exdent properly
18# - add links directly to the proper line from indices
19# - check against the definitive list of @-cmds; we still miss (among others):
20# - @defindex (hard)
21# - @c(omment) in the middle of a line (rarely used)
22# - @this* (not really needed, only used in headers anyway)
23# - @today{} (ever used outside title page?)
24
25# More consistent handling of chapters/sections/etc.
26# Lots of documentation
27# Many more options:
28#       -top    designate top node
29#       -links  customize which types of links are included
30#       -split  split at chapters or sections instead of nodes
31#       -name   Allow different types of filename handling. Non unix systems
32#               will have problems with long node names
33#       ...
34# Support the most recent texinfo version and take a good look at HTML 3.0
35# More debugging output (customizable) and more flexible error handling
36# How about icons ?
37
38# rpyron 2002-05-07
39# Robert Pyron <rpyron@alum.mit.edu>
40# 1. BUGFIX: In function makefile(), strip blanks from the nodename.
41#    This is necessary to match the behavior of parser.makeref() and
42#    parser.do_node().
43# 2. BUGFIX fixed KeyError in end_ifset (well, I may have just made
44#    it go away, rather than fix it)
45# 3. BUGFIX allow @menu and menu items inside @ifset or @ifclear
46# 4. Support added for:
47#       @uref        URL reference
48#       @image       image file reference (see note below)
49#       @multitable  output an HTML table
50#       @vtable
51# 5. Partial support for accents, to match MAKEINFO output
52# 6. I added a new command-line option, '-H basename', to specify
53#    HTML Help output. This will cause three files to be created
54#    in the current directory:
55#       `basename`.hhp  HTML Help Workshop project file
56#       `basename`.hhc  Contents file for the project
57#       `basename`.hhk  Index file for the project
58#    When fed into HTML Help Workshop, the resulting file will be
59#    named `basename`.chm.
60# 7. A new class, HTMLHelp, to accomplish item 6.
61# 8. Various calls to HTMLHelp functions.
62# A NOTE ON IMAGES: Just as 'outputdirectory' must exist before
63# running this program, all referenced images must already exist
64# in outputdirectory.
65
66import os
67import sys
68import string
69import re
70
71MAGIC = '\\input texinfo'
72
73cmprog = re.compile('^@([a-z]+)([ \t]|$)')        # Command (line-oriented)
74blprog = re.compile('^[ \t]*$')                   # Blank line
75kwprog = re.compile('@[a-z]+')                    # Keyword (embedded, usually
76                                                  # with {} args)
77spprog = re.compile('[\n@{}&<>]')                 # Special characters in
78                                                  # running text
79                                                  #
80                                                  # menu item (Yuck!)
81miprog = re.compile('^\* ([^:]*):(:|[ \t]*([^\t,\n.]+)([^ \t\n]*))[ \t\n]*')
82#                   0    1     1 2        3          34         42        0
83#                         -----            ----------  ---------
84#                                 -|-----------------------------
85#                    -----------------------------------------------------
86
87
88
89
90class HTMLNode:
91    """Some of the parser's functionality is separated into this class.
92
93    A Node accumulates its contents, takes care of links to other Nodes
94    and saves itself when it is finished and all links are resolved.
95    """
96
97    DOCTYPE = '<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN">'
98
99    type = 0
100    cont = ''
101    epilogue = '</BODY></HTML>\n'
102
103    def __init__(self, dir, name, topname, title, next, prev, up):
104        self.dirname = dir
105        self.name = name
106        if topname:
107            self.topname = topname
108        else:
109            self.topname = name
110        self.title = title
111        self.next = next
112        self.prev = prev
113        self.up = up
114        self.lines = []
115
116    def write(self, *lines):
117        map(self.lines.append, lines)
118
119    def flush(self):
120        fp = open(self.dirname + '/' + makefile(self.name), 'w')
121        fp.write(self.prologue)
122        fp.write(self.text)
123        fp.write(self.epilogue)
124        fp.close()
125
126    def link(self, label, nodename, rel=None, rev=None):
127        if nodename:
128            if nodename.lower() == '(dir)':
129                addr = '../dir.html'
130                title = ''
131            else:
132                addr = makefile(nodename)
133                title = ' TITLE="%s"' % nodename
134            self.write(label, ': <A HREF="', addr, '"', \
135                       rel and (' REL=' + rel) or "", \
136                       rev and (' REV=' + rev) or "", \
137                       title, '>', nodename, '</A>  \n')
138
139    def finalize(self):
140        length = len(self.lines)
141        self.text = ''.join(self.lines)
142        self.lines = []
143        self.open_links()
144        self.output_links()
145        self.close_links()
146        links = ''.join(self.lines)
147        self.lines = []
148        self.prologue = (
149            self.DOCTYPE +
150            '\n<HTML><HEAD>\n'
151            '  <!-- Converted with texi2html and Python -->\n'
152            '  <TITLE>' + self.title + '</TITLE>\n'
153            '  <LINK REL=Next HREF="'
154                + makefile(self.next) + '" TITLE="' + self.next + '">\n'
155            '  <LINK REL=Previous HREF="'
156                + makefile(self.prev) + '" TITLE="' + self.prev  + '">\n'
157            '  <LINK REL=Up HREF="'
158                + makefile(self.up) + '" TITLE="' + self.up  + '">\n'
159            '</HEAD><BODY>\n' +
160            links)
161        if length > 20:
162            self.epilogue = '<P>\n%s</BODY></HTML>\n' % links
163
164    def open_links(self):
165        self.write('<HR>\n')
166
167    def close_links(self):
168        self.write('<HR>\n')
169
170    def output_links(self):
171        if self.cont != self.next:
172            self.link('  Cont', self.cont)
173        self.link('  Next', self.next, rel='Next')
174        self.link('  Prev', self.prev, rel='Previous')
175        self.link('  Up', self.up, rel='Up')
176        if self.name <> self.topname:
177            self.link('  Top', self.topname)
178
179
180class HTML3Node(HTMLNode):
181
182    DOCTYPE = '<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML Level 3//EN//3.0">'
183
184    def open_links(self):
185        self.write('<DIV CLASS=Navigation>\n <HR>\n')
186
187    def close_links(self):
188        self.write(' <HR>\n</DIV>\n')
189
190
191class TexinfoParser:
192
193    COPYRIGHT_SYMBOL = "&copy;"
194    FN_ID_PATTERN = "(%(id)s)"
195    FN_SOURCE_PATTERN = '<A NAME=footnoteref%(id)s' \
196                        ' HREF="#footnotetext%(id)s">' \
197                        + FN_ID_PATTERN + '</A>'
198    FN_TARGET_PATTERN = '<A NAME=footnotetext%(id)s' \
199                        ' HREF="#footnoteref%(id)s">' \
200                        + FN_ID_PATTERN + '</A>\n%(text)s<P>\n'
201    FN_HEADER = '\n<P>\n<HR NOSHADE SIZE=1 WIDTH=200>\n' \
202                '<STRONG><EM>Footnotes</EM></STRONG>\n<P>'
203
204
205    Node = HTMLNode
206
207    # Initialize an instance
208    def __init__(self):
209        self.unknown = {}       # statistics about unknown @-commands
210        self.filenames = {}     # Check for identical filenames
211        self.debugging = 0      # larger values produce more output
212        self.print_headers = 0  # always print headers?
213        self.nodefp = None      # open file we're writing to
214        self.nodelineno = 0     # Linenumber relative to node
215        self.links = None       # Links from current node
216        self.savetext = None    # If not None, save text head instead
217        self.savestack = []     # If not None, save text head instead
218        self.htmlhelp = None    # html help data
219        self.dirname = 'tmp'    # directory where files are created
220        self.includedir = '.'   # directory to search @include files
221        self.nodename = ''      # name of current node
222        self.topname = ''       # name of top node (first node seen)
223        self.title = ''         # title of this whole Texinfo tree
224        self.resetindex()       # Reset all indices
225        self.contents = []      # Reset table of contents
226        self.numbering = []     # Reset section numbering counters
227        self.nofill = 0         # Normal operation: fill paragraphs
228        self.values={'html': 1} # Names that should be parsed in ifset
229        self.stackinfo={}       # Keep track of state in the stack
230        # XXX The following should be reset per node?!
231        self.footnotes = []     # Reset list of footnotes
232        self.itemarg = None     # Reset command used by @item
233        self.itemnumber = None  # Reset number for @item in @enumerate
234        self.itemindex = None   # Reset item index name
235        self.node = None
236        self.nodestack = []
237        self.cont = 0
238        self.includedepth = 0
239
240    # Set htmlhelp helper class
241    def sethtmlhelp(self, htmlhelp):
242        self.htmlhelp = htmlhelp
243
244    # Set (output) directory name
245    def setdirname(self, dirname):
246        self.dirname = dirname
247
248    # Set include directory name
249    def setincludedir(self, includedir):
250        self.includedir = includedir
251
252    # Parse the contents of an entire file
253    def parse(self, fp):
254        line = fp.readline()
255        lineno = 1
256        while line and (line[0] == '%' or blprog.match(line)):
257            line = fp.readline()
258            lineno = lineno + 1
259        if line[:len(MAGIC)] <> MAGIC:
260            raise SyntaxError, 'file does not begin with %r' % (MAGIC,)
261        self.parserest(fp, lineno)
262
263    # Parse the contents of a file, not expecting a MAGIC header
264    def parserest(self, fp, initial_lineno):
265        lineno = initial_lineno
266        self.done = 0
267        self.skip = 0
268        self.stack = []
269        accu = []
270        while not self.done:
271            line = fp.readline()
272            self.nodelineno = self.nodelineno + 1
273            if not line:
274                if accu:
275                    if not self.skip: self.process(accu)
276                    accu = []
277                if initial_lineno > 0:
278                    print '*** EOF before @bye'
279                break
280            lineno = lineno + 1
281            mo = cmprog.match(line)
282            if mo:
283                a, b = mo.span(1)
284                cmd = line[a:b]
285                if cmd in ('noindent', 'refill'):
286                    accu.append(line)
287                else:
288                    if accu:
289                        if not self.skip:
290                            self.process(accu)
291                        accu = []
292                    self.command(line, mo)
293            elif blprog.match(line) and \
294                 'format' not in self.stack and \
295                 'example' not in self.stack:
296                if accu:
297                    if not self.skip:
298                        self.process(accu)
299                        if self.nofill:
300                            self.write('\n')
301                        else:
302                            self.write('<P>\n')
303                        accu = []
304            else:
305                # Append the line including trailing \n!
306                accu.append(line)
307        #
308        if self.skip:
309            print '*** Still skipping at the end'
310        if self.stack:
311            print '*** Stack not empty at the end'
312            print '***', self.stack
313        if self.includedepth == 0:
314            while self.nodestack:
315                self.nodestack[-1].finalize()
316                self.nodestack[-1].flush()
317                del self.nodestack[-1]
318
319    # Start saving text in a buffer instead of writing it to a file
320    def startsaving(self):
321        if self.savetext <> None:
322            self.savestack.append(self.savetext)
323            # print '*** Recursively saving text, expect trouble'
324        self.savetext = ''
325
326    # Return the text saved so far and start writing to file again
327    def collectsavings(self):
328        savetext = self.savetext
329        if len(self.savestack) > 0:
330            self.savetext = self.savestack[-1]
331            del self.savestack[-1]
332        else:
333            self.savetext = None
334        return savetext or ''
335
336    # Write text to file, or save it in a buffer, or ignore it
337    def write(self, *args):
338        try:
339            text = ''.join(args)
340        except:
341            print args
342            raise TypeError
343        if self.savetext <> None:
344            self.savetext = self.savetext + text
345        elif self.nodefp:
346            self.nodefp.write(text)
347        elif self.node:
348            self.node.write(text)
349
350    # Complete the current node -- write footnotes and close file
351    def endnode(self):
352        if self.savetext <> None:
353            print '*** Still saving text at end of node'
354            dummy = self.collectsavings()
355        if self.footnotes:
356            self.writefootnotes()
357        if self.nodefp:
358            if self.nodelineno > 20:
359                self.write('<HR>\n')
360                [name, next, prev, up] = self.nodelinks[:4]
361                self.link('Next', next)
362                self.link('Prev', prev)
363                self.link('Up', up)
364                if self.nodename <> self.topname:
365                    self.link('Top', self.topname)
366                self.write('<HR>\n')
367            self.write('</BODY>\n')
368            self.nodefp.close()
369            self.nodefp = None
370        elif self.node:
371            if not self.cont and \
372               (not self.node.type or \
373                (self.node.next and self.node.prev and self.node.up)):
374                self.node.finalize()
375                self.node.flush()
376            else:
377                self.nodestack.append(self.node)
378            self.node = None
379        self.nodename = ''
380
381    # Process a list of lines, expanding embedded @-commands
382    # This mostly distinguishes between menus and normal text
383    def process(self, accu):
384        if self.debugging > 1:
385            print '!'*self.debugging, 'process:', self.skip, self.stack,
386            if accu: print accu[0][:30],
387            if accu[0][30:] or accu[1:]: print '...',
388            print
389        if self.inmenu():
390            # XXX should be done differently
391            for line in accu:
392                mo = miprog.match(line)
393                if not mo:
394                    line = line.strip() + '\n'
395                    self.expand(line)
396                    continue
397                bgn, end = mo.span(0)
398                a, b = mo.span(1)
399                c, d = mo.span(2)
400                e, f = mo.span(3)
401                g, h = mo.span(4)
402                label = line[a:b]
403                nodename = line[c:d]
404                if nodename[0] == ':': nodename = label
405                else: nodename = line[e:f]
406                punct = line[g:h]
407                self.write('  <LI><A HREF="',
408                           makefile(nodename),
409                           '">', nodename,
410                           '</A>', punct, '\n')
411                self.htmlhelp.menuitem(nodename)
412                self.expand(line[end:])
413        else:
414            text = ''.join(accu)
415            self.expand(text)
416
417    # find 'menu' (we might be inside 'ifset' or 'ifclear')
418    def inmenu(self):
419        #if 'menu' in self.stack:
420        #    print 'inmenu   :', self.skip, self.stack, self.stackinfo
421        stack = self.stack
422        while stack and stack[-1] in ('ifset','ifclear'):
423            try:
424                if self.stackinfo[len(stack)]:
425                    return 0
426            except KeyError:
427                pass
428            stack = stack[:-1]
429        return (stack and stack[-1] == 'menu')
430
431    # Write a string, expanding embedded @-commands
432    def expand(self, text):
433        stack = []
434        i = 0
435        n = len(text)
436        while i < n:
437            start = i
438            mo = spprog.search(text, i)
439            if mo:
440                i = mo.start()
441            else:
442                self.write(text[start:])
443                break
444            self.write(text[start:i])
445            c = text[i]
446            i = i+1
447            if c == '\n':
448                self.write('\n')
449                continue
450            if c == '<':
451                self.write('&lt;')
452                continue
453            if c == '>':
454                self.write('&gt;')
455                continue
456            if c == '&':
457                self.write('&amp;')
458                continue
459            if c == '{':
460                stack.append('')
461                continue
462            if c == '}':
463                if not stack:
464                    print '*** Unmatched }'
465                    self.write('}')
466                    continue
467                cmd = stack[-1]
468                del stack[-1]
469                try:
470                    method = getattr(self, 'close_' + cmd)
471                except AttributeError:
472                    self.unknown_close(cmd)
473                    continue
474                method()
475                continue
476            if c <> '@':
477                # Cannot happen unless spprog is changed
478                raise RuntimeError, 'unexpected funny %r' % c
479            start = i
480            while i < n and text[i] in string.ascii_letters: i = i+1
481            if i == start:
482                # @ plus non-letter: literal next character
483                i = i+1
484                c = text[start:i]
485                if c == ':':
486                    # `@:' means no extra space after
487                    # preceding `.', `?', `!' or `:'
488                    pass
489                else:
490                    # `@.' means a sentence-ending period;
491                    # `@@', `@{', `@}' quote `@', `{', `}'
492                    self.write(c)
493                continue
494            cmd = text[start:i]
495            if i < n and text[i] == '{':
496                i = i+1
497                stack.append(cmd)
498                try:
499                    method = getattr(self, 'open_' + cmd)
500                except AttributeError:
501                    self.unknown_open(cmd)
502                    continue
503                method()
504                continue
505            try:
506                method = getattr(self, 'handle_' + cmd)
507            except AttributeError:
508                self.unknown_handle(cmd)
509                continue
510            method()
511        if stack:
512            print '*** Stack not empty at para:', stack
513
514    # --- Handle unknown embedded @-commands ---
515
516    def unknown_open(self, cmd):
517        print '*** No open func for @' + cmd + '{...}'
518        cmd = cmd + '{'
519        self.write('@', cmd)
520        if not self.unknown.has_key(cmd):
521            self.unknown[cmd] = 1
522        else:
523            self.unknown[cmd] = self.unknown[cmd] + 1
524
525    def unknown_close(self, cmd):
526        print '*** No close func for @' + cmd + '{...}'
527        cmd = '}' + cmd
528        self.write('}')
529        if not self.unknown.has_key(cmd):
530            self.unknown[cmd] = 1
531        else:
532            self.unknown[cmd] = self.unknown[cmd] + 1
533
534    def unknown_handle(self, cmd):
535        print '*** No handler for @' + cmd
536        self.write('@', cmd)
537        if not self.unknown.has_key(cmd):
538            self.unknown[cmd] = 1
539        else:
540            self.unknown[cmd] = self.unknown[cmd] + 1
541
542    # XXX The following sections should be ordered as the texinfo docs
543
544    # --- Embedded @-commands without {} argument list --
545
546    def handle_noindent(self): pass
547
548    def handle_refill(self): pass
549
550    # --- Include file handling ---
551
552    def do_include(self, args):
553        file = args
554        file = os.path.join(self.includedir, file)
555        try:
556            fp = open(file, 'r')
557        except IOError, msg:
558            print '*** Can\'t open include file', repr(file)
559            return
560        print '!'*self.debugging, '--> file', repr(file)
561        save_done = self.done
562        save_skip = self.skip
563        save_stack = self.stack
564        self.includedepth = self.includedepth + 1
565        self.parserest(fp, 0)
566        self.includedepth = self.includedepth - 1
567        fp.close()
568        self.done = save_done
569        self.skip = save_skip
570        self.stack = save_stack
571        print '!'*self.debugging, '<-- file', repr(file)
572
573    # --- Special Insertions ---
574
575    def open_dmn(self): pass
576    def close_dmn(self): pass
577
578    def open_dots(self): self.write('...')
579    def close_dots(self): pass
580
581    def open_bullet(self): pass
582    def close_bullet(self): pass
583
584    def open_TeX(self): self.write('TeX')
585    def close_TeX(self): pass
586
587    def handle_copyright(self): self.write(self.COPYRIGHT_SYMBOL)
588    def open_copyright(self): self.write(self.COPYRIGHT_SYMBOL)
589    def close_copyright(self): pass
590
591    def open_minus(self): self.write('-')
592    def close_minus(self): pass
593
594    # --- Accents ---
595
596    # rpyron 2002-05-07
597    # I would like to do at least as well as makeinfo when
598    # it is producing HTML output:
599    #
600    #   input               output
601    #     @"o                 @"o                umlaut accent
602    #     @'o                 'o                 acute accent
603    #     @,{c}               @,{c}              cedilla accent
604    #     @=o                 @=o                macron/overbar accent
605    #     @^o                 @^o                circumflex accent
606    #     @`o                 `o                 grave accent
607    #     @~o                 @~o                tilde accent
608    #     @dotaccent{o}       @dotaccent{o}      overdot accent
609    #     @H{o}               @H{o}              long Hungarian umlaut
610    #     @ringaccent{o}      @ringaccent{o}     ring accent
611    #     @tieaccent{oo}      @tieaccent{oo}     tie-after accent
612    #     @u{o}               @u{o}              breve accent
613    #     @ubaraccent{o}      @ubaraccent{o}     underbar accent
614    #     @udotaccent{o}      @udotaccent{o}     underdot accent
615    #     @v{o}               @v{o}              hacek or check accent
616    #     @exclamdown{}       &#161;             upside-down !
617    #     @questiondown{}     &#191;             upside-down ?
618    #     @aa{},@AA{}         &#229;,&#197;      a,A with circle
619    #     @ae{},@AE{}         &#230;,&#198;      ae,AE ligatures
620    #     @dotless{i}         @dotless{i}        dotless i
621    #     @dotless{j}         @dotless{j}        dotless j
622    #     @l{},@L{}           l/,L/              suppressed-L,l
623    #     @o{},@O{}           &#248;,&#216;      O,o with slash
624    #     @oe{},@OE{}         oe,OE              oe,OE ligatures
625    #     @ss{}               &#223;             es-zet or sharp S
626    #
627    # The following character codes and approximations have been
628    # copied from makeinfo's HTML output.
629
630    def open_exclamdown(self): self.write('&#161;')   # upside-down !
631    def close_exclamdown(self): pass
632    def open_questiondown(self): self.write('&#191;') # upside-down ?
633    def close_questiondown(self): pass
634    def open_aa(self): self.write('&#229;') # a with circle
635    def close_aa(self): pass
636    def open_AA(self): self.write('&#197;') # A with circle
637    def close_AA(self): pass
638    def open_ae(self): self.write('&#230;') # ae ligatures
639    def close_ae(self): pass
640    def open_AE(self): self.write('&#198;') # AE ligatures
641    def close_AE(self): pass
642    def open_o(self): self.write('&#248;')  # o with slash
643    def close_o(self): pass
644    def open_O(self): self.write('&#216;')  # O with slash
645    def close_O(self): pass
646    def open_ss(self): self.write('&#223;') # es-zet or sharp S
647    def close_ss(self): pass
648    def open_oe(self): self.write('oe')     # oe ligatures
649    def close_oe(self): pass
650    def open_OE(self): self.write('OE')     # OE ligatures
651    def close_OE(self): pass
652    def open_l(self): self.write('l/')      # suppressed-l
653    def close_l(self): pass
654    def open_L(self): self.write('L/')      # suppressed-L
655    def close_L(self): pass
656
657    # --- Special Glyphs for Examples ---
658
659    def open_result(self): self.write('=&gt;')
660    def close_result(self): pass
661
662    def open_expansion(self): self.write('==&gt;')
663    def close_expansion(self): pass
664
665    def open_print(self): self.write('-|')
666    def close_print(self): pass
667
668    def open_error(self): self.write('error--&gt;')
669    def close_error(self): pass
670
671    def open_equiv(self): self.write('==')
672    def close_equiv(self): pass
673
674    def open_point(self): self.write('-!-')
675    def close_point(self): pass
676
677    # --- Cross References ---
678
679    def open_pxref(self):
680        self.write('see ')
681        self.startsaving()
682    def close_pxref(self):
683        self.makeref()
684
685    def open_xref(self):
686        self.write('See ')
687        self.startsaving()
688    def close_xref(self):
689        self.makeref()
690
691    def open_ref(self):
692        self.startsaving()
693    def close_ref(self):
694        self.makeref()
695
696    def open_inforef(self):
697        self.write('See info file ')
698        self.startsaving()
699    def close_inforef(self):
700        text = self.collectsavings()
701        args = [s.strip() for s in text.split(',')]
702        while len(args) < 3: args.append('')
703        node = args[0]
704        file = args[2]
705        self.write('`', file, '\', node `', node, '\'')
706
707    def makeref(self):
708        text = self.collectsavings()
709        args = [s.strip() for s in text.split(',')]
710        while len(args) < 5: args.append('')
711        nodename = label = args[0]
712        if args[2]: label = args[2]
713        file = args[3]
714        title = args[4]
715        href = makefile(nodename)
716        if file:
717            href = '../' + file + '/' + href
718        self.write('<A HREF="', href, '">', label, '</A>')
719
720    # rpyron 2002-05-07  uref support
721    def open_uref(self):
722        self.startsaving()
723    def close_uref(self):
724        text = self.collectsavings()
725        args = [s.strip() for s in text.split(',')]
726        while len(args) < 2: args.append('')
727        href = args[0]
728        label = args[1]
729        if not label: label = href
730        self.write('<A HREF="', href, '">', label, '</A>')
731
732    # rpyron 2002-05-07  image support
733    # GNU makeinfo producing HTML output tries `filename.png'; if
734    # that does not exist, it tries `filename.jpg'. If that does
735    # not exist either, it complains. GNU makeinfo does not handle
736    # GIF files; however, I include GIF support here because
737    # MySQL documentation uses GIF files.
738
739    def open_image(self):
740        self.startsaving()
741    def close_image(self):
742        self.makeimage()
743    def makeimage(self):
744        text = self.collectsavings()
745        args = [s.strip() for s in text.split(',')]
746        while len(args) < 5: args.append('')
747        filename = args[0]
748        width    = args[1]
749        height   = args[2]
750        alt      = args[3]
751        ext      = args[4]
752
753        # The HTML output will have a reference to the image
754        # that is relative to the HTML output directory,
755        # which is what 'filename' gives us. However, we need
756        # to find it relative to our own current directory,
757        # so we construct 'imagename'.
758        imagelocation = self.dirname + '/' + filename
759
760        if   os.path.exists(imagelocation+'.png'):
761            filename += '.png'
762        elif os.path.exists(imagelocation+'.jpg'):
763            filename += '.jpg'
764        elif os.path.exists(imagelocation+'.gif'):   # MySQL uses GIF files
765            filename += '.gif'
766        else:
767            print "*** Cannot find image " + imagelocation
768        #TODO: what is 'ext'?
769        self.write('<IMG SRC="', filename, '"',                     \
770                    width  and (' WIDTH="'  + width  + '"') or "",  \
771                    height and (' HEIGHT="' + height + '"') or "",  \
772                    alt    and (' ALT="'    + alt    + '"') or "",  \
773                    '/>' )
774        self.htmlhelp.addimage(imagelocation)
775
776
777    # --- Marking Words and Phrases ---
778
779    # --- Other @xxx{...} commands ---
780
781    def open_(self): pass # Used by {text enclosed in braces}
782    def close_(self): pass
783
784    open_asis = open_
785    close_asis = close_
786
787    def open_cite(self): self.write('<CITE>')
788    def close_cite(self): self.write('</CITE>')
789
790    def open_code(self): self.write('<CODE>')
791    def close_code(self): self.write('</CODE>')
792
793    def open_t(self): self.write('<TT>')
794    def close_t(self): self.write('</TT>')
795
796    def open_dfn(self): self.write('<DFN>')
797    def close_dfn(self): self.write('</DFN>')
798
799    def open_emph(self): self.write('<EM>')
800    def close_emph(self): self.write('</EM>')
801
802    def open_i(self): self.write('<I>')
803    def close_i(self): self.write('</I>')
804
805    def open_footnote(self):
806        # if self.savetext <> None:
807        #       print '*** Recursive footnote -- expect weirdness'
808        id = len(self.footnotes) + 1
809        self.write(self.FN_SOURCE_PATTERN % {'id': repr(id)})
810        self.startsaving()
811
812    def close_footnote(self):
813        id = len(self.footnotes) + 1
814        self.footnotes.append((id, self.collectsavings()))
815
816    def writefootnotes(self):
817        self.write(self.FN_HEADER)
818        for id, text in self.footnotes:
819            self.write(self.FN_TARGET_PATTERN
820                       % {'id': repr(id), 'text': text})
821        self.footnotes = []
822
823    def open_file(self): self.write('<CODE>')
824    def close_file(self): self.write('</CODE>')
825
826    def open_kbd(self): self.write('<KBD>')
827    def close_kbd(self): self.write('</KBD>')
828
829    def open_key(self): self.write('<KEY>')
830    def close_key(self): self.write('</KEY>')
831
832    def open_r(self): self.write('<R>')
833    def close_r(self): self.write('</R>')
834
835    def open_samp(self): self.write('`<SAMP>')
836    def close_samp(self): self.write('</SAMP>\'')
837
838    def open_sc(self): self.write('<SMALLCAPS>')
839    def close_sc(self): self.write('</SMALLCAPS>')
840
841    def open_strong(self): self.write('<STRONG>')
842    def close_strong(self): self.write('</STRONG>')
843
844    def open_b(self): self.write('<B>')
845    def close_b(self): self.write('</B>')
846
847    def open_var(self): self.write('<VAR>')
848    def close_var(self): self.write('</VAR>')
849
850    def open_w(self): self.write('<NOBREAK>')
851    def close_w(self): self.write('</NOBREAK>')
852
853    def open_url(self): self.startsaving()
854    def close_url(self):
855        text = self.collectsavings()
856        self.write('<A HREF="', text, '">', text, '</A>')
857
858    def open_email(self): self.startsaving()
859    def close_email(self):
860        text = self.collectsavings()
861        self.write('<A HREF="mailto:', text, '">', text, '</A>')
862
863    open_titlefont = open_
864    close_titlefont = close_
865
866    def open_small(self): pass
867    def close_small(self): pass
868
869    def command(self, line, mo):
870        a, b = mo.span(1)
871        cmd = line[a:b]
872        args = line[b:].strip()
873        if self.debugging > 1:
874            print '!'*self.debugging, 'command:', self.skip, self.stack, \
875                  '@' + cmd, args
876        try:
877            func = getattr(self, 'do_' + cmd)
878        except AttributeError:
879            try:
880                func = getattr(self, 'bgn_' + cmd)
881            except AttributeError:
882                # don't complain if we are skipping anyway
883                if not self.skip:
884                    self.unknown_cmd(cmd, args)
885                return
886            self.stack.append(cmd)
887            func(args)
888            return
889        if not self.skip or cmd == 'end':
890            func(args)
891
892    def unknown_cmd(self, cmd, args):
893        print '*** unknown', '@' + cmd, args
894        if not self.unknown.has_key(cmd):
895            self.unknown[cmd] = 1
896        else:
897            self.unknown[cmd] = self.unknown[cmd] + 1
898
899    def do_end(self, args):
900        words = args.split()
901        if not words:
902            print '*** @end w/o args'
903        else:
904            cmd = words[0]
905            if not self.stack or self.stack[-1] <> cmd:
906                print '*** @end', cmd, 'unexpected'
907            else:
908                del self.stack[-1]
909            try:
910                func = getattr(self, 'end_' + cmd)
911            except AttributeError:
912                self.unknown_end(cmd)
913                return
914            func()
915
916    def unknown_end(self, cmd):
917        cmd = 'end ' + cmd
918        print '*** unknown', '@' + cmd
919        if not self.unknown.has_key(cmd):
920            self.unknown[cmd] = 1
921        else:
922            self.unknown[cmd] = self.unknown[cmd] + 1
923
924    # --- Comments ---
925
926    def do_comment(self, args): pass
927    do_c = do_comment
928
929    # --- Conditional processing ---
930
931    def bgn_ifinfo(self, args): pass
932    def end_ifinfo(self): pass
933
934    def bgn_iftex(self, args): self.skip = self.skip + 1
935    def end_iftex(self): self.skip = self.skip - 1
936
937    def bgn_ignore(self, args): self.skip = self.skip + 1
938    def end_ignore(self): self.skip = self.skip - 1
939
940    def bgn_tex(self, args): self.skip = self.skip + 1
941    def end_tex(self): self.skip = self.skip - 1
942
943    def do_set(self, args):
944        fields = args.split(' ')
945        key = fields[0]
946        if len(fields) == 1:
947            value = 1
948        else:
949            value = ' '.join(fields[1:])
950        self.values[key] = value
951
952    def do_clear(self, args):
953        self.values[args] = None
954
955    def bgn_ifset(self, args):
956        if args not in self.values.keys() \
957           or self.values[args] is None:
958            self.skip = self.skip + 1
959            self.stackinfo[len(self.stack)] = 1
960        else:
961            self.stackinfo[len(self.stack)] = 0
962    def end_ifset(self):
963        try:
964            if self.stackinfo[len(self.stack) + 1]:
965                self.skip = self.skip - 1
966            del self.stackinfo[len(self.stack) + 1]
967        except KeyError:
968            print '*** end_ifset: KeyError :', len(self.stack) + 1
969
970    def bgn_ifclear(self, args):
971        if args in self.values.keys() \
972           and self.values[args] is not None:
973            self.skip = self.skip + 1
974            self.stackinfo[len(self.stack)] = 1
975        else:
976            self.stackinfo[len(self.stack)] = 0
977    def end_ifclear(self):
978        try:
979            if self.stackinfo[len(self.stack) + 1]:
980                self.skip = self.skip - 1
981            del self.stackinfo[len(self.stack) + 1]
982        except KeyError:
983            print '*** end_ifclear: KeyError :', len(self.stack) + 1
984
985    def open_value(self):
986        self.startsaving()
987
988    def close_value(self):
989        key = self.collectsavings()
990        if key in self.values.keys():
991            self.write(self.values[key])
992        else:
993            print '*** Undefined value: ', key
994
995    # --- Beginning a file ---
996
997    do_finalout = do_comment
998    do_setchapternewpage = do_comment
999    do_setfilename = do_comment
1000
1001    def do_settitle(self, args):
1002        self.startsaving()
1003        self.expand(args)
1004        self.title = self.collectsavings()
1005    def do_parskip(self, args): pass
1006
1007    # --- Ending a file ---
1008
1009    def do_bye(self, args):
1010        self.endnode()
1011        self.done = 1
1012
1013    # --- Title page ---
1014
1015    def bgn_titlepage(self, args): self.skip = self.skip + 1
1016    def end_titlepage(self): self.skip = self.skip - 1
1017    def do_shorttitlepage(self, args): pass
1018
1019    def do_center(self, args):
1020        # Actually not used outside title page...
1021        self.write('<H1>')
1022        self.expand(args)
1023        self.write('</H1>\n')
1024    do_title = do_center
1025    do_subtitle = do_center
1026    do_author = do_center
1027
1028    do_vskip = do_comment
1029    do_vfill = do_comment
1030    do_smallbook = do_comment
1031
1032    do_paragraphindent = do_comment
1033    do_setchapternewpage = do_comment
1034    do_headings = do_comment
1035    do_footnotestyle = do_comment
1036
1037    do_evenheading = do_comment
1038    do_evenfooting = do_comment
1039    do_oddheading = do_comment
1040    do_oddfooting = do_comment
1041    do_everyheading = do_comment
1042    do_everyfooting = do_comment
1043
1044    # --- Nodes ---
1045
1046    def do_node(self, args):
1047        self.endnode()
1048        self.nodelineno = 0
1049        parts = [s.strip() for s in args.split(',')]
1050        while len(parts) < 4: parts.append('')
1051        self.nodelinks = parts
1052        [name, next, prev, up] = parts[:4]
1053        file = self.dirname + '/' + makefile(name)
1054        if self.filenames.has_key(file):
1055            print '*** Filename already in use: ', file
1056        else:
1057            if self.debugging: print '!'*self.debugging, '--- writing', file
1058        self.filenames[file] = 1
1059        # self.nodefp = open(file, 'w')
1060        self.nodename = name
1061        if self.cont and self.nodestack:
1062            self.nodestack[-1].cont = self.nodename
1063        if not self.topname: self.topname = name
1064        title = name
1065        if self.title: title = title + ' -- ' + self.title
1066        self.node = self.Node(self.dirname, self.nodename, self.topname,
1067                              title, next, prev, up)
1068        self.htmlhelp.addnode(self.nodename,next,prev,up,file)
1069
1070    def link(self, label, nodename):
1071        if nodename:
1072            if nodename.lower() == '(dir)':
1073                addr = '../dir.html'
1074            else:
1075                addr = makefile(nodename)
1076            self.write(label, ': <A HREF="', addr, '" TYPE="',
1077                       label, '">', nodename, '</A>  \n')
1078
1079    # --- Sectioning commands ---
1080
1081    def popstack(self, type):
1082        if (self.node):
1083            self.node.type = type
1084            while self.nodestack:
1085                if self.nodestack[-1].type > type:
1086                    self.nodestack[-1].finalize()
1087                    self.nodestack[-1].flush()
1088                    del self.nodestack[-1]
1089                elif self.nodestack[-1].type == type:
1090                    if not self.nodestack[-1].next:
1091                        self.nodestack[-1].next = self.node.name
1092                    if not self.node.prev:
1093                        self.node.prev = self.nodestack[-1].name
1094                    self.nodestack[-1].finalize()
1095                    self.nodestack[-1].flush()
1096                    del self.nodestack[-1]
1097                else:
1098                    if type > 1 and not self.node.up:
1099                        self.node.up = self.nodestack[-1].name
1100                    break
1101
1102    def do_chapter(self, args):
1103        self.heading('H1', args, 0)
1104        self.popstack(1)
1105
1106    def do_unnumbered(self, args):
1107        self.heading('H1', args, -1)
1108        self.popstack(1)
1109    def do_appendix(self, args):
1110        self.heading('H1', args, -1)
1111        self.popstack(1)
1112    def do_top(self, args):
1113        self.heading('H1', args, -1)
1114    def do_chapheading(self, args):
1115        self.heading('H1', args, -1)
1116    def do_majorheading(self, args):
1117        self.heading('H1', args, -1)
1118
1119    def do_section(self, args):
1120        self.heading('H1', args, 1)
1121        self.popstack(2)
1122
1123    def do_unnumberedsec(self, args):
1124        self.heading('H1', args, -1)
1125        self.popstack(2)
1126    def do_appendixsec(self, args):
1127        self.heading('H1', args, -1)
1128        self.popstack(2)
1129    do_appendixsection = do_appendixsec
1130    def do_heading(self, args):
1131        self.heading('H1', args, -1)
1132
1133    def do_subsection(self, args):
1134        self.heading('H2', args, 2)
1135        self.popstack(3)
1136    def do_unnumberedsubsec(self, args):
1137        self.heading('H2', args, -1)
1138        self.popstack(3)
1139    def do_appendixsubsec(self, args):
1140        self.heading('H2', args, -1)
1141        self.popstack(3)
1142    def do_subheading(self, args):
1143        self.heading('H2', args, -1)
1144
1145    def do_subsubsection(self, args):
1146        self.heading('H3', args, 3)
1147        self.popstack(4)
1148    def do_unnumberedsubsubsec(self, args):
1149        self.heading('H3', args, -1)
1150        self.popstack(4)
1151    def do_appendixsubsubsec(self, args):
1152        self.heading('H3', args, -1)
1153        self.popstack(4)
1154    def do_subsubheading(self, args):
1155        self.heading('H3', args, -1)
1156
1157    def heading(self, type, args, level):
1158        if level >= 0:
1159            while len(self.numbering) <= level:
1160                self.numbering.append(0)
1161            del self.numbering[level+1:]
1162            self.numbering[level] = self.numbering[level] + 1
1163            x = ''
1164            for i in self.numbering:
1165                x = x + repr(i) + '.'
1166            args = x + ' ' + args
1167            self.contents.append((level, args, self.nodename))
1168        self.write('<', type, '>')
1169        self.expand(args)
1170        self.write('</', type, '>\n')
1171        if self.debugging or self.print_headers:
1172            print '---', args
1173
1174    def do_contents(self, args):
1175        # pass
1176        self.listcontents('Table of Contents', 999)
1177
1178    def do_shortcontents(self, args):
1179        pass
1180        # self.listcontents('Short Contents', 0)
1181    do_summarycontents = do_shortcontents
1182
1183    def listcontents(self, title, maxlevel):
1184        self.write('<H1>', title, '</H1>\n<UL COMPACT PLAIN>\n')
1185        prevlevels = [0]
1186        for level, title, node in self.contents:
1187            if level > maxlevel:
1188                continue
1189            if level > prevlevels[-1]:
1190                # can only advance one level at a time
1191                self.write('  '*prevlevels[-1], '<UL PLAIN>\n')
1192                prevlevels.append(level)
1193            elif level < prevlevels[-1]:
1194                # might drop back multiple levels
1195                while level < prevlevels[-1]:
1196                    del prevlevels[-1]
1197                    self.write('  '*prevlevels[-1],
1198                               '</UL>\n')
1199            self.write('  '*level, '<LI> <A HREF="',
1200                       makefile(node), '">')
1201            self.expand(title)
1202            self.write('</A>\n')
1203        self.write('</UL>\n' * len(prevlevels))
1204
1205    # --- Page lay-out ---
1206
1207    # These commands are only meaningful in printed text
1208
1209    def do_page(self, args): pass
1210
1211    def do_need(self, args): pass
1212
1213    def bgn_group(self, args): pass
1214    def end_group(self): pass
1215
1216    # --- Line lay-out ---
1217
1218    def do_sp(self, args):
1219        if self.nofill:
1220            self.write('\n')
1221        else:
1222            self.write('<P>\n')
1223
1224    def do_hline(self, args):
1225        self.write('<HR>')
1226
1227    # --- Function and variable definitions ---
1228
1229    def bgn_deffn(self, args):
1230        self.write('<DL>')
1231        self.do_deffnx(args)
1232
1233    def end_deffn(self):
1234        self.write('</DL>\n')
1235
1236    def do_deffnx(self, args):
1237        self.write('<DT>')
1238        words = splitwords(args, 2)
1239        [category, name], rest = words[:2], words[2:]
1240        self.expand('@b{%s}' % name)
1241        for word in rest: self.expand(' ' + makevar(word))
1242        #self.expand(' -- ' + category)
1243        self.write('\n<DD>')
1244        self.index('fn', name)
1245
1246    def bgn_defun(self, args): self.bgn_deffn('Function ' + args)
1247    end_defun = end_deffn
1248    def do_defunx(self, args): self.do_deffnx('Function ' + args)
1249
1250    def bgn_defmac(self, args): self.bgn_deffn('Macro ' + args)
1251    end_defmac = end_deffn
1252    def do_defmacx(self, args): self.do_deffnx('Macro ' + args)
1253
1254    def bgn_defspec(self, args): self.bgn_deffn('{Special Form} ' + args)
1255    end_defspec = end_deffn
1256    def do_defspecx(self, args): self.do_deffnx('{Special Form} ' + args)
1257
1258    def bgn_defvr(self, args):
1259        self.write('<DL>')
1260        self.do_defvrx(args)
1261
1262    end_defvr = end_deffn
1263
1264    def do_defvrx(self, args):
1265        self.write('<DT>')
1266        words = splitwords(args, 2)
1267        [category, name], rest = words[:2], words[2:]
1268        self.expand('@code{%s}' % name)
1269        # If there are too many arguments, show them
1270        for word in rest: self.expand(' ' + word)
1271        #self.expand(' -- ' + category)
1272        self.write('\n<DD>')
1273        self.index('vr', name)
1274
1275    def bgn_defvar(self, args): self.bgn_defvr('Variable ' + args)
1276    end_defvar = end_defvr
1277    def do_defvarx(self, args): self.do_defvrx('Variable ' + args)
1278
1279    def bgn_defopt(self, args): self.bgn_defvr('{User Option} ' + args)
1280    end_defopt = end_defvr
1281    def do_defoptx(self, args): self.do_defvrx('{User Option} ' + args)
1282
1283    # --- Ditto for typed languages ---
1284
1285    def bgn_deftypefn(self, args):
1286        self.write('<DL>')
1287        self.do_deftypefnx(args)
1288
1289    end_deftypefn = end_deffn
1290
1291    def do_deftypefnx(self, args):
1292        self.write('<DT>')
1293        words = splitwords(args, 3)
1294        [category, datatype, name], rest = words[:3], words[3:]
1295        self.expand('@code{%s} @b{%s}' % (datatype, name))
1296        for word in rest: self.expand(' ' + makevar(word))
1297        #self.expand(' -- ' + category)
1298        self.write('\n<DD>')
1299        self.index('fn', name)
1300
1301
1302    def bgn_deftypefun(self, args): self.bgn_deftypefn('Function ' + args)
1303    end_deftypefun = end_deftypefn
1304    def do_deftypefunx(self, args): self.do_deftypefnx('Function ' + args)
1305
1306    def bgn_deftypevr(self, args):
1307        self.write('<DL>')
1308        self.do_deftypevrx(args)
1309
1310    end_deftypevr = end_deftypefn
1311
1312    def do_deftypevrx(self, args):
1313        self.write('<DT>')
1314        words = splitwords(args, 3)
1315        [category, datatype, name], rest = words[:3], words[3:]
1316        self.expand('@code{%s} @b{%s}' % (datatype, name))
1317        # If there are too many arguments, show them
1318        for word in rest: self.expand(' ' + word)
1319        #self.expand(' -- ' + category)
1320        self.write('\n<DD>')
1321        self.index('fn', name)
1322
1323    def bgn_deftypevar(self, args):
1324        self.bgn_deftypevr('Variable ' + args)
1325    end_deftypevar = end_deftypevr
1326    def do_deftypevarx(self, args):
1327        self.do_deftypevrx('Variable ' + args)
1328
1329    # --- Ditto for object-oriented languages ---
1330
1331    def bgn_defcv(self, args):
1332        self.write('<DL>')
1333        self.do_defcvx(args)
1334
1335    end_defcv = end_deftypevr
1336
1337    def do_defcvx(self, args):
1338        self.write('<DT>')
1339        words = splitwords(args, 3)
1340        [category, classname, name], rest = words[:3], words[3:]
1341        self.expand('@b{%s}' % name)
1342        # If there are too many arguments, show them
1343        for word in rest: self.expand(' ' + word)
1344        #self.expand(' -- %s of @code{%s}' % (category, classname))
1345        self.write('\n<DD>')
1346        self.index('vr', '%s @r{on %s}' % (name, classname))
1347
1348    def bgn_defivar(self, args):
1349        self.bgn_defcv('{Instance Variable} ' + args)
1350    end_defivar = end_defcv
1351    def do_defivarx(self, args):
1352        self.do_defcvx('{Instance Variable} ' + args)
1353
1354    def bgn_defop(self, args):
1355        self.write('<DL>')
1356        self.do_defopx(args)
1357
1358    end_defop = end_defcv
1359
1360    def do_defopx(self, args):
1361        self.write('<DT>')
1362        words = splitwords(args, 3)
1363        [category, classname, name], rest = words[:3], words[3:]
1364        self.expand('@b{%s}' % name)
1365        for word in rest: self.expand(' ' + makevar(word))
1366        #self.expand(' -- %s of @code{%s}' % (category, classname))
1367        self.write('\n<DD>')
1368        self.index('fn', '%s @r{on %s}' % (name, classname))
1369
1370    def bgn_defmethod(self, args):
1371        self.bgn_defop('Method ' + args)
1372    end_defmethod = end_defop
1373    def do_defmethodx(self, args):
1374        self.do_defopx('Method ' + args)
1375
1376    # --- Ditto for data types ---
1377
1378    def bgn_deftp(self, args):
1379        self.write('<DL>')
1380        self.do_deftpx(args)
1381
1382    end_deftp = end_defcv
1383
1384    def do_deftpx(self, args):
1385        self.write('<DT>')
1386        words = splitwords(args, 2)
1387        [category, name], rest = words[:2], words[2:]
1388        self.expand('@b{%s}' % name)
1389        for word in rest: self.expand(' ' + word)
1390        #self.expand(' -- ' + category)
1391        self.write('\n<DD>')
1392        self.index('tp', name)
1393
1394    # --- Making Lists and Tables
1395
1396    def bgn_enumerate(self, args):
1397        if not args:
1398            self.write('<OL>\n')
1399            self.stackinfo[len(self.stack)] = '</OL>\n'
1400        else:
1401            self.itemnumber = args
1402            self.write('<UL>\n')
1403            self.stackinfo[len(self.stack)] = '</UL>\n'
1404    def end_enumerate(self):
1405        self.itemnumber = None
1406        self.write(self.stackinfo[len(self.stack) + 1])
1407        del self.stackinfo[len(self.stack) + 1]
1408
1409    def bgn_itemize(self, args):
1410        self.itemarg = args
1411        self.write('<UL>\n')
1412    def end_itemize(self):
1413        self.itemarg = None
1414        self.write('</UL>\n')
1415
1416    def bgn_table(self, args):
1417        self.itemarg = args
1418        self.write('<DL>\n')
1419    def end_table(self):
1420        self.itemarg = None
1421        self.write('</DL>\n')
1422
1423    def bgn_ftable(self, args):
1424        self.itemindex = 'fn'
1425        self.bgn_table(args)
1426    def end_ftable(self):
1427        self.itemindex = None
1428        self.end_table()
1429
1430    def bgn_vtable(self, args):
1431        self.itemindex = 'vr'
1432        self.bgn_table(args)
1433    def end_vtable(self):
1434        self.itemindex = None
1435        self.end_table()
1436
1437    def do_item(self, args):
1438        if self.itemindex: self.index(self.itemindex, args)
1439        if self.itemarg:
1440            if self.itemarg[0] == '@' and self.itemarg[1] and \
1441                            self.itemarg[1] in string.ascii_letters:
1442                args = self.itemarg + '{' + args + '}'
1443            else:
1444                # some other character, e.g. '-'
1445                args = self.itemarg + ' ' + args
1446        if self.itemnumber <> None:
1447            args = self.itemnumber + '. ' + args
1448            self.itemnumber = increment(self.itemnumber)
1449        if self.stack and self.stack[-1] == 'table':
1450            self.write('<DT>')
1451            self.expand(args)
1452            self.write('\n<DD>')
1453        elif self.stack and self.stack[-1] == 'multitable':
1454            self.write('<TR><TD>')
1455            self.expand(args)
1456            self.write('</TD>\n</TR>\n')
1457        else:
1458            self.write('<LI>')
1459            self.expand(args)
1460            self.write('  ')
1461    do_itemx = do_item # XXX Should suppress leading blank line
1462
1463    # rpyron 2002-05-07  multitable support
1464    def bgn_multitable(self, args):
1465        self.itemarg = None     # should be handled by columnfractions
1466        self.write('<TABLE BORDER="">\n')
1467    def end_multitable(self):
1468        self.itemarg = None
1469        self.write('</TABLE>\n<BR>\n')
1470    def handle_columnfractions(self):
1471        # It would be better to handle this, but for now it's in the way...
1472        self.itemarg = None
1473    def handle_tab(self):
1474        self.write('</TD>\n    <TD>')
1475
1476    # --- Enumerations, displays, quotations ---
1477    # XXX Most of these should increase the indentation somehow
1478
1479    def bgn_quotation(self, args): self.write('<BLOCKQUOTE>')
1480    def end_quotation(self): self.write('</BLOCKQUOTE>\n')
1481
1482    def bgn_example(self, args):
1483        self.nofill = self.nofill + 1
1484        self.write('<PRE>')
1485    def end_example(self):
1486        self.write('</PRE>\n')
1487        self.nofill = self.nofill - 1
1488
1489    bgn_lisp = bgn_example # Synonym when contents are executable lisp code
1490    end_lisp = end_example
1491
1492    bgn_smallexample = bgn_example # XXX Should use smaller font
1493    end_smallexample = end_example
1494
1495    bgn_smalllisp = bgn_lisp # Ditto
1496    end_smalllisp = end_lisp
1497
1498    bgn_display = bgn_example
1499    end_display = end_example
1500
1501    bgn_format = bgn_display
1502    end_format = end_display
1503
1504    def do_exdent(self, args): self.expand(args + '\n')
1505    # XXX Should really mess with indentation
1506
1507    def bgn_flushleft(self, args):
1508        self.nofill = self.nofill + 1
1509        self.write('<PRE>\n')
1510    def end_flushleft(self):
1511        self.write('</PRE>\n')
1512        self.nofill = self.nofill - 1
1513
1514    def bgn_flushright(self, args):
1515        self.nofill = self.nofill + 1
1516        self.write('<ADDRESS COMPACT>\n')
1517    def end_flushright(self):
1518        self.write('</ADDRESS>\n')
1519        self.nofill = self.nofill - 1
1520
1521    def bgn_menu(self, args):
1522        self.write('<DIR>\n')
1523        self.write('  <STRONG><EM>Menu</EM></STRONG><P>\n')
1524        self.htmlhelp.beginmenu()
1525    def end_menu(self):
1526        self.write('</DIR>\n')
1527        self.htmlhelp.endmenu()
1528
1529    def bgn_cartouche(self, args): pass
1530    def end_cartouche(self): pass
1531
1532    # --- Indices ---
1533
1534    def resetindex(self):
1535        self.noncodeindices = ['cp']
1536        self.indextitle = {}
1537        self.indextitle['cp'] = 'Concept'
1538        self.indextitle['fn'] = 'Function'
1539        self.indextitle['ky'] = 'Keyword'
1540        self.indextitle['pg'] = 'Program'
1541        self.indextitle['tp'] = 'Type'
1542        self.indextitle['vr'] = 'Variable'
1543        #
1544        self.whichindex = {}
1545        for name in self.indextitle.keys():
1546            self.whichindex[name] = []
1547
1548    def user_index(self, name, args):
1549        if self.whichindex.has_key(name):
1550            self.index(name, args)
1551        else:
1552            print '*** No index named', repr(name)
1553
1554    def do_cindex(self, args): self.index('cp', args)
1555    def do_findex(self, args): self.index('fn', args)
1556    def do_kindex(self, args): self.index('ky', args)
1557    def do_pindex(self, args): self.index('pg', args)
1558    def do_tindex(self, args): self.index('tp', args)
1559    def do_vindex(self, args): self.index('vr', args)
1560
1561    def index(self, name, args):
1562        self.whichindex[name].append((args, self.nodename))
1563        self.htmlhelp.index(args, self.nodename)
1564
1565    def do_synindex(self, args):
1566        words = args.split()
1567        if len(words) <> 2:
1568            print '*** bad @synindex', args
1569            return
1570        [old, new] = words
1571        if not self.whichindex.has_key(old) or \
1572                  not self.whichindex.has_key(new):
1573            print '*** bad key(s) in @synindex', args
1574            return
1575        if old <> new and \
1576                  self.whichindex[old] is not self.whichindex[new]:
1577            inew = self.whichindex[new]
1578            inew[len(inew):] = self.whichindex[old]
1579            self.whichindex[old] = inew
1580    do_syncodeindex = do_synindex # XXX Should use code font
1581
1582    def do_printindex(self, args):
1583        words = args.split()
1584        for name in words:
1585            if self.whichindex.has_key(name):
1586                self.prindex(name)
1587            else:
1588                print '*** No index named', repr(name)
1589
1590    def prindex(self, name):
1591        iscodeindex = (name not in self.noncodeindices)
1592        index = self.whichindex[name]
1593        if not index: return
1594        if self.debugging:
1595            print '!'*self.debugging, '--- Generating', \
1596                  self.indextitle[name], 'index'
1597        #  The node already provides a title
1598        index1 = []
1599        junkprog = re.compile('^(@[a-z]+)?{')
1600        for key, node in index:
1601            sortkey = key.lower()
1602            # Remove leading `@cmd{' from sort key
1603            # -- don't bother about the matching `}'
1604            oldsortkey = sortkey
1605            while 1:
1606                mo = junkprog.match(sortkey)
1607                if not mo:
1608                    break
1609                i = mo.end()
1610                sortkey = sortkey[i:]
1611            index1.append((sortkey, key, node))
1612        del index[:]
1613        index1.sort()
1614        self.write('<DL COMPACT>\n')
1615        prevkey = prevnode = None
1616        for sortkey, key, node in index1:
1617            if (key, node) == (prevkey, prevnode):
1618                continue
1619            if self.debugging > 1: print '!'*self.debugging, key, ':', node
1620            self.write('<DT>')
1621            if iscodeindex: key = '@code{' + key + '}'
1622            if key != prevkey:
1623                self.expand(key)
1624            self.write('\n<DD><A HREF="%s">%s</A>\n' % (makefile(node), node))
1625            prevkey, prevnode = key, node
1626        self.write('</DL>\n')
1627
1628    # --- Final error reports ---
1629
1630    def report(self):
1631        if self.unknown:
1632            print '--- Unrecognized commands ---'
1633            cmds = self.unknown.keys()
1634            cmds.sort()
1635            for cmd in cmds:
1636                print cmd.ljust(20), self.unknown[cmd]
1637
1638
1639class TexinfoParserHTML3(TexinfoParser):
1640
1641    COPYRIGHT_SYMBOL = "&copy;"
1642    FN_ID_PATTERN = "[%(id)s]"
1643    FN_SOURCE_PATTERN = '<A ID=footnoteref%(id)s ' \
1644                        'HREF="#footnotetext%(id)s">' + FN_ID_PATTERN + '</A>'
1645    FN_TARGET_PATTERN = '<FN ID=footnotetext%(id)s>\n' \
1646                        '<P><A HREF="#footnoteref%(id)s">' + FN_ID_PATTERN \
1647                        + '</A>\n%(text)s</P></FN>\n'
1648    FN_HEADER = '<DIV CLASS=footnotes>\n  <HR NOSHADE WIDTH=200>\n' \
1649                '  <STRONG><EM>Footnotes</EM></STRONG>\n  <P>\n'
1650
1651    Node = HTML3Node
1652
1653    def bgn_quotation(self, args): self.write('<BQ>')
1654    def end_quotation(self): self.write('</BQ>\n')
1655
1656    def bgn_example(self, args):
1657        # this use of <CODE> would not be legal in HTML 2.0,
1658        # but is in more recent DTDs.
1659        self.nofill = self.nofill + 1
1660        self.write('<PRE CLASS=example><CODE>')
1661    def end_example(self):
1662        self.write("</CODE></PRE>\n")
1663        self.nofill = self.nofill - 1
1664
1665    def bgn_flushleft(self, args):
1666        self.nofill = self.nofill + 1
1667        self.write('<PRE CLASS=flushleft>\n')
1668
1669    def bgn_flushright(self, args):
1670        self.nofill = self.nofill + 1
1671        self.write('<DIV ALIGN=right CLASS=flushright><ADDRESS COMPACT>\n')
1672    def end_flushright(self):
1673        self.write('</ADDRESS></DIV>\n')
1674        self.nofill = self.nofill - 1
1675
1676    def bgn_menu(self, args):
1677        self.write('<UL PLAIN CLASS=menu>\n')
1678        self.write('  <LH>Menu</LH>\n')
1679    def end_menu(self):
1680        self.write('</UL>\n')
1681
1682
1683# rpyron 2002-05-07
1684class HTMLHelp:
1685    """
1686    This class encapsulates support for HTML Help. Node names,
1687    file names, menu items, index items, and image file names are
1688    accumulated until a call to finalize(). At that time, three
1689    output files are created in the current directory:
1690
1691        `helpbase`.hhp  is a HTML Help Workshop project file.
1692                        It contains various information, some of
1693                        which I do not understand; I just copied
1694                        the default project info from a fresh
1695                        installation.
1696        `helpbase`.hhc  is the Contents file for the project.
1697        `helpbase`.hhk  is the Index file for the project.
1698
1699    When these files are used as input to HTML Help Workshop,
1700    the resulting file will be named:
1701
1702        `helpbase`.chm
1703
1704    If none of the defaults in `helpbase`.hhp are changed,
1705    the .CHM file will have Contents, Index, Search, and
1706    Favorites tabs.
1707    """
1708
1709    codeprog = re.compile('@code{(.*?)}')
1710
1711    def __init__(self,helpbase,dirname):
1712        self.helpbase    = helpbase
1713        self.dirname     = dirname
1714        self.projectfile = None
1715        self.contentfile = None
1716        self.indexfile   = None
1717        self.nodelist    = []
1718        self.nodenames   = {}         # nodename : index
1719        self.nodeindex   = {}
1720        self.filenames   = {}         # filename : filename
1721        self.indexlist   = []         # (args,nodename) == (key,location)
1722        self.current     = ''
1723        self.menudict    = {}
1724        self.dumped      = {}
1725
1726
1727    def addnode(self,name,next,prev,up,filename):
1728        node = (name,next,prev,up,filename)
1729        # add this file to dict
1730        # retrieve list with self.filenames.values()
1731        self.filenames[filename] = filename
1732        # add this node to nodelist
1733        self.nodeindex[name] = len(self.nodelist)
1734        self.nodelist.append(node)
1735        # set 'current' for menu items
1736        self.current = name
1737        self.menudict[self.current] = []
1738
1739    def menuitem(self,nodename):
1740        menu = self.menudict[self.current]
1741        menu.append(nodename)
1742
1743
1744    def addimage(self,imagename):
1745        self.filenames[imagename] = imagename
1746
1747    def index(self, args, nodename):
1748        self.indexlist.append((args,nodename))
1749
1750    def beginmenu(self):
1751        pass
1752
1753    def endmenu(self):
1754        pass
1755
1756    def finalize(self):
1757        if not self.helpbase:
1758            return
1759
1760        # generate interesting filenames
1761        resultfile   = self.helpbase + '.chm'
1762        projectfile  = self.helpbase + '.hhp'
1763        contentfile  = self.helpbase + '.hhc'
1764        indexfile    = self.helpbase + '.hhk'
1765
1766        # generate a reasonable title
1767        title        = self.helpbase
1768
1769        # get the default topic file
1770        (topname,topnext,topprev,topup,topfile) = self.nodelist[0]
1771        defaulttopic = topfile
1772
1773        # PROJECT FILE
1774        try:
1775            fp = open(projectfile,'w')
1776            print>>fp, '[OPTIONS]'
1777            print>>fp, 'Auto Index=Yes'
1778            print>>fp, 'Binary TOC=No'
1779            print>>fp, 'Binary Index=Yes'
1780            print>>fp, 'Compatibility=1.1'
1781            print>>fp, 'Compiled file=' + resultfile + ''
1782            print>>fp, 'Contents file=' + contentfile + ''
1783            print>>fp, 'Default topic=' + defaulttopic + ''
1784            print>>fp, 'Error log file=ErrorLog.log'
1785            print>>fp, 'Index file=' + indexfile + ''
1786            print>>fp, 'Title=' + title + ''
1787            print>>fp, 'Display compile progress=Yes'
1788            print>>fp, 'Full-text search=Yes'
1789            print>>fp, 'Default window=main'
1790            print>>fp, ''
1791            print>>fp, '[WINDOWS]'
1792            print>>fp, ('main=,"' + contentfile + '","' + indexfile
1793                        + '","","",,,,,0x23520,222,0x1046,[10,10,780,560],'
1794                        '0xB0000,,,,,,0')
1795            print>>fp, ''
1796            print>>fp, '[FILES]'
1797            print>>fp, ''
1798            self.dumpfiles(fp)
1799            fp.close()
1800        except IOError, msg:
1801            print projectfile, ':', msg
1802            sys.exit(1)
1803
1804        # CONTENT FILE
1805        try:
1806            fp = open(contentfile,'w')
1807            print>>fp, '<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN">'
1808            print>>fp, '<!-- This file defines the table of contents -->'
1809            print>>fp, '<HTML>'
1810            print>>fp, '<HEAD>'
1811            print>>fp, ('<meta name="GENERATOR" '
1812                        'content="Microsoft&reg; HTML Help Workshop 4.1">')
1813            print>>fp, '<!-- Sitemap 1.0 -->'
1814            print>>fp, '</HEAD>'
1815            print>>fp, '<BODY>'
1816            print>>fp, '   <OBJECT type="text/site properties">'
1817            print>>fp, '     <param name="Window Styles" value="0x800025">'
1818            print>>fp, '     <param name="comment" value="title:">'
1819            print>>fp, '     <param name="comment" value="base:">'
1820            print>>fp, '   </OBJECT>'
1821            self.dumpnodes(fp)
1822            print>>fp, '</BODY>'
1823            print>>fp, '</HTML>'
1824            fp.close()
1825        except IOError, msg:
1826            print contentfile, ':', msg
1827            sys.exit(1)
1828
1829        # INDEX FILE
1830        try:
1831            fp = open(indexfile  ,'w')
1832            print>>fp, '<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN">'
1833            print>>fp, '<!-- This file defines the index -->'
1834            print>>fp, '<HTML>'
1835            print>>fp, '<HEAD>'
1836            print>>fp, ('<meta name="GENERATOR" '
1837                        'content="Microsoft&reg; HTML Help Workshop 4.1">')
1838            print>>fp, '<!-- Sitemap 1.0 -->'
1839            print>>fp, '</HEAD>'
1840            print>>fp, '<BODY>'
1841            print>>fp, '<OBJECT type="text/site properties">'
1842            print>>fp, '</OBJECT>'
1843            self.dumpindex(fp)
1844            print>>fp, '</BODY>'
1845            print>>fp, '</HTML>'
1846            fp.close()
1847        except IOError, msg:
1848            print indexfile  , ':', msg
1849            sys.exit(1)
1850
1851    def dumpfiles(self, outfile=sys.stdout):
1852        filelist = self.filenames.values()
1853        filelist.sort()
1854        for filename in filelist:
1855            print>>outfile, filename
1856
1857    def dumpnodes(self, outfile=sys.stdout):
1858        self.dumped = {}
1859        if self.nodelist:
1860            nodename, dummy, dummy, dummy, dummy = self.nodelist[0]
1861            self.topnode = nodename
1862
1863        print>>outfile,  '<UL>'
1864        for node in self.nodelist:
1865            self.dumpnode(node,0,outfile)
1866        print>>outfile,  '</UL>'
1867
1868    def dumpnode(self, node, indent=0, outfile=sys.stdout):
1869        if node:
1870            # Retrieve info for this node
1871            (nodename,next,prev,up,filename) = node
1872            self.current = nodename
1873
1874            # Have we been dumped already?
1875            if self.dumped.has_key(nodename):
1876                return
1877            self.dumped[nodename] = 1
1878
1879            # Print info for this node
1880            print>>outfile, ' '*indent,
1881            print>>outfile, '<LI><OBJECT type="text/sitemap">',
1882            print>>outfile, '<param name="Name" value="' + nodename +'">',
1883            print>>outfile, '<param name="Local" value="'+ filename +'">',
1884            print>>outfile, '</OBJECT>'
1885
1886            # Does this node have menu items?
1887            try:
1888                menu = self.menudict[nodename]
1889                self.dumpmenu(menu,indent+2,outfile)
1890            except KeyError:
1891                pass
1892
1893    def dumpmenu(self, menu, indent=0, outfile=sys.stdout):
1894        if menu:
1895            currentnode = self.current
1896            if currentnode != self.topnode:    # XXX this is a hack
1897                print>>outfile, ' '*indent + '<UL>'
1898                indent += 2
1899            for item in menu:
1900                menunode = self.getnode(item)
1901                self.dumpnode(menunode,indent,outfile)
1902            if currentnode != self.topnode:    # XXX this is a hack
1903                print>>outfile, ' '*indent + '</UL>'
1904                indent -= 2
1905
1906    def getnode(self, nodename):
1907        try:
1908            index = self.nodeindex[nodename]
1909            return self.nodelist[index]
1910        except KeyError:
1911            return None
1912        except IndexError:
1913            return None
1914
1915    # (args,nodename) == (key,location)
1916    def dumpindex(self, outfile=sys.stdout):
1917        print>>outfile,  '<UL>'
1918        for (key,location) in self.indexlist:
1919            key = self.codeexpand(key)
1920            location = makefile(location)
1921            location = self.dirname + '/' + location
1922            print>>outfile, '<LI><OBJECT type="text/sitemap">',
1923            print>>outfile, '<param name="Name" value="' + key + '">',
1924            print>>outfile, '<param name="Local" value="' + location + '">',
1925            print>>outfile, '</OBJECT>'
1926        print>>outfile,  '</UL>'
1927
1928    def codeexpand(self, line):
1929        co = self.codeprog.match(line)
1930        if not co:
1931            return line
1932        bgn, end = co.span(0)
1933        a, b = co.span(1)
1934        line = line[:bgn] + line[a:b] + line[end:]
1935        return line
1936
1937
1938# Put @var{} around alphabetic substrings
1939def makevar(str):
1940    return '@var{'+str+'}'
1941
1942
1943# Split a string in "words" according to findwordend
1944def splitwords(str, minlength):
1945    words = []
1946    i = 0
1947    n = len(str)
1948    while i < n:
1949        while i < n and str[i] in ' \t\n': i = i+1
1950        if i >= n: break
1951        start = i
1952        i = findwordend(str, i, n)
1953        words.append(str[start:i])
1954    while len(words) < minlength: words.append('')
1955    return words
1956
1957
1958# Find the end of a "word", matching braces and interpreting @@ @{ @}
1959fwprog = re.compile('[@{} ]')
1960def findwordend(str, i, n):
1961    level = 0
1962    while i < n:
1963        mo = fwprog.search(str, i)
1964        if not mo:
1965            break
1966        i = mo.start()
1967        c = str[i]; i = i+1
1968        if c == '@': i = i+1 # Next character is not special
1969        elif c == '{': level = level+1
1970        elif c == '}': level = level-1
1971        elif c == ' ' and level <= 0: return i-1
1972    return n
1973
1974
1975# Convert a node name into a file name
1976def makefile(nodename):
1977    nodename = nodename.strip()
1978    return fixfunnychars(nodename) + '.html'
1979
1980
1981# Characters that are perfectly safe in filenames and hyperlinks
1982goodchars = string.ascii_letters + string.digits + '!@-=+.'
1983
1984# Replace characters that aren't perfectly safe by dashes
1985# Underscores are bad since Cern HTTPD treats them as delimiters for
1986# encoding times, so you get mismatches if you compress your files:
1987# a.html.gz will map to a_b.html.gz
1988def fixfunnychars(addr):
1989    i = 0
1990    while i < len(addr):
1991        c = addr[i]
1992        if c not in goodchars:
1993            c = '-'
1994            addr = addr[:i] + c + addr[i+1:]
1995        i = i + len(c)
1996    return addr
1997
1998
1999# Increment a string used as an enumeration
2000def increment(s):
2001    if not s:
2002        return '1'
2003    for sequence in string.digits, string.lowercase, string.uppercase:
2004        lastc = s[-1]
2005        if lastc in sequence:
2006            i = sequence.index(lastc) + 1
2007            if i >= len(sequence):
2008                if len(s) == 1:
2009                    s = sequence[0]*2
2010                    if s == '00':
2011                        s = '10'
2012                else:
2013                    s = increment(s[:-1]) + sequence[0]
2014            else:
2015                s = s[:-1] + sequence[i]
2016            return s
2017    return s # Don't increment
2018
2019
2020def test():
2021    import sys
2022    debugging = 0
2023    print_headers = 0
2024    cont = 0
2025    html3 = 0
2026    htmlhelp = ''
2027
2028    while sys.argv[1] == ['-d']:
2029        debugging = debugging + 1
2030        del sys.argv[1]
2031    if sys.argv[1] == '-p':
2032        print_headers = 1
2033        del sys.argv[1]
2034    if sys.argv[1] == '-c':
2035        cont = 1
2036        del sys.argv[1]
2037    if sys.argv[1] == '-3':
2038        html3 = 1
2039        del sys.argv[1]
2040    if sys.argv[1] == '-H':
2041        helpbase = sys.argv[2]
2042        del sys.argv[1:3]
2043    if len(sys.argv) <> 3:
2044        print 'usage: texi2hh [-d [-d]] [-p] [-c] [-3] [-H htmlhelp]', \
2045              'inputfile outputdirectory'
2046        sys.exit(2)
2047
2048    if html3:
2049        parser = TexinfoParserHTML3()
2050    else:
2051        parser = TexinfoParser()
2052    parser.cont = cont
2053    parser.debugging = debugging
2054    parser.print_headers = print_headers
2055
2056    file = sys.argv[1]
2057    dirname  = sys.argv[2]
2058    parser.setdirname(dirname)
2059    parser.setincludedir(os.path.dirname(file))
2060
2061    htmlhelp = HTMLHelp(helpbase, dirname)
2062    parser.sethtmlhelp(htmlhelp)
2063
2064    try:
2065        fp = open(file, 'r')
2066    except IOError, msg:
2067        print file, ':', msg
2068        sys.exit(1)
2069
2070    parser.parse(fp)
2071    fp.close()
2072    parser.report()
2073
2074    htmlhelp.finalize()
2075
2076
2077if __name__ == "__main__":
2078    test()
2079