cpython/Doc/tools/toc2bkm.py

161 lines
4.4 KiB
Python
Raw Normal View History

#! /usr/bin/env python
"""Convert a LaTeX .toc file to some PDFTeX magic to create that neat outline.
The output file has an extension of '.bkm' instead of '.out', since hyperref
already uses that extension.
"""
import getopt
import os
import re
import string
import sys
# Ench item in an entry is a tuple of:
#
# Section #, Title String, Page #, List of Sub-entries
#
# The return value of parse_toc() is such a tuple.
cline_re = r"""^
\\contentsline\ \{([a-z]*)} # type of section in $1
\{(?:\\numberline\ \{([0-9.A-Z]+)})? # section number
(.*)} # title string
2000-10-07 09:50:05 -03:00
\{(\d+)}$""" # page number
cline_rx = re.compile(cline_re, re.VERBOSE)
OUTER_TO_INNER = -1
_transition_map = {
('chapter', 'section'): OUTER_TO_INNER,
('section', 'subsection'): OUTER_TO_INNER,
('subsection', 'subsubsection'): OUTER_TO_INNER,
('subsubsection', 'subsection'): 1,
('subsection', 'section'): 1,
('section', 'chapter'): 1,
('subsection', 'chapter'): 2,
('subsubsection', 'section'): 2,
('subsubsection', 'chapter'): 3,
}
INCLUDED_LEVELS = ("chapter", "section", "subsection", "subsubsection")
class BadSectionNesting(Exception):
"""Raised for unsupported section level transitions."""
def __init__(self, level, newsection, path, lineno):
self.level = level
self.newsection = newsection
self.path = path
self.lineno = lineno
def __str__(self):
return ("illegal transition from %s to %s at %s (line %s)"
% (self.level, self.newsection, self.path, self.lineno))
def parse_toc(fp, bigpart=None):
toc = top = []
stack = [toc]
level = bigpart or 'chapter'
lineno = 0
while 1:
2000-10-07 09:50:05 -03:00
line = fp.readline()
if not line:
break
lineno = lineno + 1
m = cline_rx.match(line)
if m:
stype, snum, title, pageno = m.group(1, 2, 3, 4)
title = clean_title(title)
2002-10-16 12:30:17 -03:00
entry = (stype, snum, title, int(pageno), [])
2000-10-07 09:50:05 -03:00
if stype == level:
toc.append(entry)
else:
if stype not in INCLUDED_LEVELS:
# we don't want paragraphs & subparagraphs
continue
try:
direction = _transition_map[(level, stype)]
except KeyError:
raise BadSectionNesting(level, stype, fp.name, lineno)
2000-10-07 09:50:05 -03:00
if direction == OUTER_TO_INNER:
toc = toc[-1][-1]
stack.insert(0, toc)
toc.append(entry)
else:
for i in range(direction):
del stack[0]
toc = stack[0]
toc.append(entry)
level = stype
else:
sys.stderr.write("l.%s: " + line)
return top
hackscore_rx = re.compile(r"\\hackscore\s*{[^}]*}")
raisebox_rx = re.compile(r"\\raisebox\s*{[^}]*}")
title_rx = re.compile(r"\\([a-zA-Z])+\s+")
title_trans = string.maketrans("", "")
def clean_title(title):
title = raisebox_rx.sub("", title)
title = hackscore_rx.sub(r"\\_", title)
pos = 0
while 1:
2000-10-07 09:50:05 -03:00
m = title_rx.search(title, pos)
if m:
start = m.start()
if title[start:start+15] != "\\textunderscore":
title = title[:start] + title[m.end():]
pos = start + 1
else:
break
2002-10-16 12:30:17 -03:00
title = title.translate(title_trans, "{}")
return title
def write_toc(toc, fp):
for entry in toc:
2000-10-07 09:50:05 -03:00
write_toc_entry(entry, fp, 0)
def write_toc_entry(entry, fp, layer):
stype, snum, title, pageno, toc = entry
s = "\\pdfoutline goto name{page%03d}" % pageno
if toc:
2000-10-07 09:50:05 -03:00
s = "%s count -%d" % (s, len(toc))
if snum:
2000-10-07 09:50:05 -03:00
title = "%s %s" % (snum, title)
s = "%s {%s}\n" % (s, title)
fp.write(s)
for entry in toc:
2000-10-07 09:50:05 -03:00
write_toc_entry(entry, fp, layer + 1)
def process(ifn, ofn, bigpart=None):
toc = parse_toc(open(ifn), bigpart)
write_toc(toc, open(ofn, "w"))
def main():
bigpart = None
opts, args = getopt.getopt(sys.argv[1:], "c:")
if opts:
2000-10-07 09:50:05 -03:00
bigpart = opts[0][1]
if not args:
2000-10-07 09:50:05 -03:00
usage()
sys.exit(2)
for filename in args:
2000-10-07 09:50:05 -03:00
base, ext = os.path.splitext(filename)
ext = ext or ".toc"
process(base + ext, base + ".bkm", bigpart)
if __name__ == "__main__":
main()