3 """usage: %(program)s [options...] file ...
5 Options specifying formats to build:
6 --html HyperText Markup Language (default)
7 --pdf Portable Document Format
9 --dvi 'DeVice Indepentent' format from TeX
10 --text ASCII text (requires lynx)
12 More than one output format may be specified, or --all.
15 --address, -a Specify an address for page footers.
16 --link Specify the number of levels to include on each page.
17 --split, -s Specify a section level for page splitting, default: %(max_split_depth)s.
18 --iconserver, -i Specify location of icons (default: ../).
19 --image-type Specify the image type to use in HTML output;
20 values: gif (default), png.
21 --numeric Don't rename the HTML files; just keep node#.html for
23 --style Specify the CSS file to use for the output (filename,
25 --up-link URL to a parent document.
26 --up-title Title of a parent document.
29 --a4 Format for A4 paper.
30 --letter Format for US letter paper (the default).
31 --help, -H Show this text.
32 --logging, -l Log stdout and stderr to a file (*.how).
33 --debugging, -D Echo commands as they are executed.
34 --keep, -k Keep temporary files around.
35 --quiet, -q Do not print command output to stdout.
36 (stderr is also lost, sorry; see *.how for errors)
49 MYDIR = os.path.abspath(sys.path[0])
51 ISTFILE = os.path.join(MYDIR, "texinputs", "python.ist")
52 NODE2LABEL_SCRIPT = os.path.join(MYDIR, "node2label.pl")
53 L2H_INIT_FILE = os.path.join(MYDIR, "perl", "l2hinit.perl")
55 BIBTEX_BINARY = "bibtex"
56 DVIPS_BINARY = "dvips"
57 LATEX_BINARY = "latex"
58 LATEX2HTML_BINARY = "latex2html"
60 MAKEINDEX_BINARY = "makeindex"
61 PDFLATEX_BINARY = "pdflatex"
63 PYTHON_BINARY = "python"
67 print __doc__ % options
69 def error(options, message, err=2):
70 sys.stdout = sys.stderr
78 program = os.path.basename(sys.argv[0])
94 global_module_index = None
95 style_file = os.path.join(MYDIR, "html", "style.css")
96 about_file = os.path.join(MYDIR, "html", "about.dat")
100 DEFAULT_FORMATS = ("html",)
101 ALL_FORMATS = ("dvi", "html", "pdf", "ps", "text")
105 self.l2h_init_files = []
107 def __getitem__(self, key):
108 # This is used when formatting the usage message.
110 return getattr(self, key)
111 except AttributeError:
114 def parse(self, args):
115 opts, args = getopt.getopt(args, "Hi:a:s:lDkqr:",
116 ["all", "postscript", "help", "iconserver=",
117 "address=", "a4", "letter", "l2h-init=",
118 "link=", "split=", "logging", "debugging",
119 "keep", "quiet", "runs=", "image-type=",
120 "about=", "numeric", "style=", "paper=",
121 "up-link=", "up-title=", "dir=",
122 "global-module-index="]
123 + list(self.ALL_FORMATS))
124 for opt, arg in opts:
126 self.formats = list(self.ALL_FORMATS)
127 elif opt in ("-H", "--help"):
130 elif opt == "--iconserver":
131 self.icon_server = arg
132 elif opt in ("-a", "--address"):
136 elif opt == "--letter":
137 self.paper = "letter"
138 elif opt == "--link":
139 self.max_link_depth = int(arg)
140 elif opt in ("-s", "--split"):
141 self.max_split_depth = int(arg)
142 elif opt in ("-l", "--logging"):
143 self.logging = self.logging + 1
144 elif opt in ("-D", "--debugging"):
145 self.debugging = self.debugging + 1
146 elif opt in ("-k", "--keep"):
147 self.discard_temps = 0
148 elif opt in ("-q", "--quiet"):
150 elif opt in ("-r", "--runs"):
152 elif opt == "--image-type":
153 self.image_type = arg
154 elif opt == "--about":
155 # always make this absolute:
156 self.about_file = os.path.normpath(
157 os.path.abspath(arg))
158 elif opt == "--numeric":
160 elif opt == "--style":
161 self.style_file = os.path.abspath(arg)
162 elif opt == "--l2h-init":
163 self.l2h_init_files.append(os.path.abspath(arg))
164 elif opt == "--up-link":
166 elif opt == "--up-title":
168 elif opt == "--global-module-index":
169 self.global_module_index = arg
172 elif opt == "--paper":
177 elif opt[2:] in self.ALL_FORMATS:
178 self.add_format(opt[2:])
179 elif opt == "--postscript":
181 self.add_format("ps")
184 # return the args to allow the caller access:
188 def add_format(self, format):
189 """Add a format to the formats list if not present."""
190 if not format in self.formats:
191 self.formats.append(format)
193 def initialize(self):
194 """Complete initialization. This is needed if parse() isn't used."""
195 # add the default format if no formats were specified:
197 self.formats = self.DEFAULT_FORMATS
198 # determine the base set of texinputs directories:
199 texinputs = string.split(os.environ.get("TEXINPUTS", ""), os.pathsep)
202 self.base_texinputs = [
203 os.path.join(MYDIR, "paper-" + self.paper),
204 os.path.join(MYDIR, "texinputs"),
211 def __init__(self, options, path):
212 self.options = options
213 self.doctype = get_doctype(path)
214 self.filedir, self.doc = split_pathname(path)
215 self.log_filename = self.doc + ".how"
216 if os.path.exists(self.log_filename):
217 os.unlink(self.log_filename)
218 if os.path.exists(self.doc + ".l2h"):
219 self.l2h_aux_init_file = tempfile.mktemp()
221 self.l2h_aux_init_file = self.doc + ".l2h"
222 self.write_l2h_aux_init_file()
225 self.setup_texinputs()
226 formats = self.options.formats
227 if "dvi" in formats or "ps" in formats:
233 if "html" in formats:
235 self.build_html(self.options.builddir or self.doc)
236 if self.options.icon_server == ".":
237 pattern = os.path.join(MYDIR, "html", "icons",
238 "*." + self.options.image_type)
239 imgs = glob.glob(pattern)
242 "Could not locate support images of type %s."
243 % `self.options.image_type`)
245 new_fn = os.path.join(self.doc, os.path.basename(fn))
246 shutil.copyfile(fn, new_fn)
247 if "text" in formats:
250 need_html = "html" not in formats
251 if self.options.max_split_depth != 1:
252 fp = open(self.l2h_aux_init_file, "a")
253 fp.write("# re-hack this file for --text:\n")
254 l2hoption(fp, "MAX_SPLIT_DEPTH", "1")
257 tempdir = self.doc + "-temp-html"
260 self.build_html(tempdir, max_split_depth=1)
261 self.build_text(tempdir)
262 if self.options.discard_temps:
265 def setup_texinputs(self):
266 texinputs = [self.filedir] + list(self.options.base_texinputs)
267 os.environ["TEXINPUTS"] = string.join(['.']+texinputs, os.pathsep)
268 self.message("TEXINPUTS=" + os.environ["TEXINPUTS"])
270 def build_aux(self, binary=None):
272 binary = LATEX_BINARY
273 new_index( "%s.ind" % self.doc, "genindex")
274 new_index("mod%s.ind" % self.doc, "modindex")
275 self.run("%s %s" % (binary, self.doc))
276 self.use_bibtex = check_for_bibtex(self.doc + ".aux")
280 self.use_latex(LATEX_BINARY)
283 self.use_latex(PDFLATEX_BINARY)
285 def use_latex(self, binary):
286 self.require_temps(binary=binary)
287 if self.latex_runs < 2:
288 if os.path.isfile("mod%s.idx" % self.doc):
289 self.run("%s mod%s.idx" % (MAKEINDEX_BINARY, self.doc))
291 if os.path.isfile(self.doc + ".idx"):
293 # call to Doc/tools/fix_hack omitted; doesn't appear necessary
294 self.run("%s %s.idx" % (MAKEINDEX_BINARY, self.doc))
296 indfix.process(self.doc + ".ind")
298 self.run("%s %s" % (BIBTEX_BINARY, self.doc))
299 self.process_synopsis_files()
301 # let the doctype-specific handler do some intermediate work:
303 self.run("%s %s" % (binary, self.doc))
304 self.latex_runs = self.latex_runs + 1
305 if os.path.isfile("mod%s.idx" % self.doc):
306 self.run("%s -s %s mod%s.idx"
307 % (MAKEINDEX_BINARY, ISTFILE, self.doc))
309 self.run("%s -s %s %s.idx"
310 % (MAKEINDEX_BINARY, ISTFILE, self.doc))
311 indfix.process(self.doc + ".ind")
312 self.process_synopsis_files()
314 # and now finish it off:
316 if os.path.isfile(self.doc + ".toc") and binary == PDFLATEX_BINARY:
318 if self.doctype == "manual":
322 toc2bkm.process(self.doc + ".toc", self.doc + ".bkm", bigpart)
324 self.run("%s %s" % (BIBTEX_BINARY, self.doc))
325 self.run("%s %s" % (binary, self.doc))
326 self.latex_runs = self.latex_runs + 1
328 def process_synopsis_files(self):
329 synopsis_files = glob.glob(self.doc + "*.syn")
330 for path in synopsis_files:
331 uniqify_module_table(path)
334 self.run("%s -N0 -o %s.ps %s" % (DVIPS_BINARY, self.doc, self.doc))
336 def build_html(self, builddir=None, max_split_depth=None):
339 if max_split_depth is None:
340 max_split_depth = self.options.max_split_depth
342 for p in string.split(os.environ["TEXINPUTS"], os.pathsep):
343 fn = os.path.join(p, self.doc + ".tex")
344 if os.path.isfile(fn):
348 self.warning("Could not locate %s.tex; aborting." % self.doc)
350 # remove leading ./ (or equiv.); might avoid problems w/ dvips
351 if texfile[:2] == os.curdir + os.sep:
352 texfile = texfile[2:]
353 # build the command line and run LaTeX2HTML:
354 if not os.path.isdir(builddir):
357 for fname in glob.glob(os.path.join(builddir, "*.html")):
359 args = [LATEX2HTML_BINARY,
360 "-init_file", self.l2h_aux_init_file,
364 self.run(string.join(args)) # XXX need quoting!
366 shutil.copyfile(self.options.style_file,
367 os.path.join(builddir, self.doc + ".css"))
368 shutil.copyfile(os.path.join(builddir, self.doc + ".html"),
369 os.path.join(builddir, "index.html"))
370 if max_split_depth != 1:
371 if self.options.numeric:
372 label_file = os.path.join(builddir, "labels.pl")
373 fp = open(label_file)
375 target = " = q/about/;\n"
381 if line[-x:] == target:
383 m = re.search(r"\|(node\d+\.[a-z]+)\|", line)
384 about_node = m.group(1)
385 shutil.copyfile(os.path.join(builddir, about_node),
386 os.path.join(builddir, "about.html"))
392 self.run("%s %s *.html" % (PERL_BINARY, NODE2LABEL_SCRIPT))
396 def build_text(self, tempdir=None):
399 indexfile = os.path.join(tempdir, "index.html")
400 self.run("%s -nolist -dump %s >%s.txt"
401 % (LYNX_BINARY, indexfile, self.doc))
403 def require_temps(self, binary=None):
404 if not self.latex_runs:
405 self.build_aux(binary=binary)
407 def write_l2h_aux_init_file(self):
408 options = self.options
409 fp = open(self.l2h_aux_init_file, "w")
410 d = string_to_perl(os.path.dirname(L2H_INIT_FILE))
411 fp.write("package main;\n"
412 "push (@INC, '%s');\n"
415 fp.write(open(L2H_INIT_FILE).read())
416 for filename in options.l2h_init_files:
417 fp.write("\n# initialization code incorporated from:\n# ")
420 fp.write(open(filename).read())
422 "# auxillary init file for latex2html\n"
423 "# generated by mkhowto\n"
424 "$NO_AUTO_LINK = 1;\n"
426 l2hoption(fp, "ABOUT_FILE", options.about_file)
427 l2hoption(fp, "ICONSERVER", options.icon_server)
428 l2hoption(fp, "IMAGE_TYPE", options.image_type)
429 l2hoption(fp, "ADDRESS", options.address)
430 l2hoption(fp, "MAX_LINK_DEPTH", options.max_link_depth)
431 l2hoption(fp, "MAX_SPLIT_DEPTH", options.max_split_depth)
432 l2hoption(fp, "EXTERNAL_UP_LINK", options.up_link)
433 l2hoption(fp, "EXTERNAL_UP_TITLE", options.up_title)
434 l2hoption(fp, "GLOBAL_MODULE_INDEX", options.global_module_index)
439 self.__have_temps = 0
440 for pattern in ("%s.aux", "%s.log", "%s.out", "%s.toc", "%s.bkm",
441 "%s.idx", "%s.ilg", "%s.ind", "%s.pla",
443 "mod%s.idx", "mod%s.ind", "mod%s.ilg",
445 safe_unlink(pattern % self.doc)
446 map(safe_unlink, glob.glob(self.doc + "*.syn"))
447 for spec in ("IMG*", "*.pl", "WARNINGS", "index.dat", "modindex.dat"):
448 pattern = os.path.join(self.doc, spec)
449 map(safe_unlink, glob.glob(pattern))
450 if "dvi" not in self.options.formats:
451 safe_unlink(self.doc + ".dvi")
452 if os.path.isdir(self.doc + "-temp-html"):
453 shutil.rmtree(self.doc + "-temp-html", ignore_errors=1)
454 if not self.options.logging:
455 os.unlink(self.log_filename)
456 if not self.options.debugging:
457 os.unlink(self.l2h_aux_init_file)
459 def run(self, command):
460 self.message(command)
461 rc = os.system("(%s) </dev/null >>%s 2>&1"
462 % (command, self.log_filename))
465 "Session transcript and error messages are in %s."
467 sys.stderr.write("The relevant lines from the transcript are:\n")
468 sys.stderr.write("-" * 72 + "\n")
469 sys.stderr.writelines(get_run_transcript(self.log_filename))
472 def message(self, msg):
474 if not self.options.quiet:
478 def warning(self, msg):
479 msg = "*** %s\n" % msg
480 sys.stderr.write(msg)
484 fp = open(self.log_filename, "a")
489 def get_run_transcript(filename):
490 """Return lines from the transcript file for the most recent run() call."""
492 lines = fp.readlines()
498 if line[:4] == "+++ ":
504 def safe_unlink(path):
505 """Unlink a file without raising an error if it doesn't exist."""
512 def split_pathname(path):
513 path = os.path.normpath(os.path.join(os.getcwd(), path))
514 dirname, basename = os.path.split(path)
515 if basename[-4:] == ".tex":
516 basename = basename[:-4]
517 return dirname, basename
520 _doctype_rx = re.compile(r"\\documentclass(?:\[[^]]*\])?{([a-zA-Z]*)}")
521 def get_doctype(path):
528 m = _doctype_rx.match(line)
539 args = options.parse(sys.argv[1:])
540 except getopt.error, msg:
543 # attempt to locate single .tex file in current directory:
544 args = glob.glob("*.tex")
546 error(options, "No file to process.")
548 error(options, "Could not deduce which files should be processed.")
550 # parameters are processed, let's go!
553 Job(options, path).build()
556 def l2hoption(fp, option, value):
558 fp.write('$%s = "%s";\n' % (option, string_to_perl(str(value))))
562 for c in map(chr, range(1, 256)):
564 _to_perl["@"] = "\\@"
565 _to_perl["$"] = "\\$"
566 _to_perl['"'] = '\\"'
568 def string_to_perl(s):
569 return string.join(map(_to_perl.get, s), '')
572 def check_for_bibtex(filename):
574 pos = string.find(fp.read(), r"\bibdata{")
578 def uniqify_module_table(filename):
579 lines = open(filename).readlines()
581 if lines[-1] == lines[-2]:
583 open(filename, "w").writelines(lines)
586 def new_index(filename, label="genindex"):
587 fp = open(filename, "w")
596 if __name__ == "__main__":