get_tex_sections.py

#!/usr/bin/env python
from __future__ import print_function
import re
import argparse


def _find_ending_brace(s, open_brace=1):
    escaping = False
    output = ''
    remain = ''
    for i, c in enumerate(s):
        if escaping:
            escaping = False
        else:
            if c == '%':
                break
            elif c == '{':
                open_brace += 1
            elif c == '}':
                open_brace -= 1
                if open_brace == 0:
                    remain = s[i+1:].strip()
                    break
            escaping = (c == '\\')
        output += c
    return open_brace, output, remain


def _print_counter(counter):
    _counter = list(counter)
    output = ''
    while sum(_counter):
        output += '{}.'.format(_counter.pop(0))
    return output.rstrip('.')


def _iter_section(fp, re_sec):
    remain = ''
    open_brace = 0
    while True:
        try:
            if remain:
                l = remain
                remain = ''
            else:
                l = next(fp)
            m = re_sec.search(l)
            if m is None:
                continue
            level, skip, title = m.groups()
            open_brace, title, remain = _find_ending_brace(title.strip() + ' ', 1)
            while open_brace:
                open_brace, title_this, remain = _find_ending_brace(next(fp).strip() + ' ', open_brace)
                title += title_this
            yield level, skip, title
        except StopIteration:
            if open_brace:
                raise ValueError('misformed tex file')
            break


def iter_tex_sections(files, max_sublevel=2, no_number='*'):
    counter = [0]*(max_sublevel+1)
    re_sec = re.compile(r'\\((?:sub){{0,{}}})section(\*?){{(.*)'.format(max_sublevel))
    if isinstance(files, basestring):
        files = [files]
    for this_file in files:
        with open(this_file) as f:
            for level, skip, title in _iter_section(f, re_sec):
                if not skip:
                    level = len(level) // 3
                    counter[level] += 1
                    for i in range(level+1, len(counter)):
                        counter[i] = 0
                yield '{} {}'.format(no_number if skip else _print_counter(counter), title).strip()


def main():
    parser = argparse.ArgumentParser(description='Extract section titles from a TeX file')
    parser.add_argument('files', metavar='TEX_FILE', nargs='+')
    parser.add_argument('--max-sublevel', default=2, type=int, help='maximal subsection level (default: 2)')
    parser.add_argument('--no-number', default='*', help='Bullet point for sections without numbering')
    args = parser.parse_args()
    for l in iter_tex_sections(**args.__dict__):
        print(l)


if __name__ == '__main__':
    main()