devscripts/prepare_manpage.py

   1 #!/usr/bin/env python3
   2
   3 # Allow direct execution
   4 import os
   5 import sys
   6
   7 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
   8
   9
  10 import os.path
  11 import re
  12
  13 from devscripts.utils import (
  14     compose_functions,
  15     get_filename_args,
  16     read_file,
  17     write_file,
  18 )
  19
  20 ROOT_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
  21 README_FILE = os.path.join(ROOT_DIR, 'README.md')
  22
  23 PREFIX = r'''%yt-dlp(1)
  24
  25 # NAME
  26
  27 yt\-dlp \- A feature\-rich command\-line audio/video downloader
  28
  29 # SYNOPSIS
  30
  31 **yt-dlp** \[OPTIONS\] URL [URL...]
  32
  33 # DESCRIPTION
  34
  35 '''
  36
  37
  38 def filter_excluded_sections(readme):
  39     EXCLUDED_SECTION_BEGIN_STRING = re.escape('<!-- MANPAGE: BEGIN EXCLUDED SECTION -->')
  40     EXCLUDED_SECTION_END_STRING = re.escape('<!-- MANPAGE: END EXCLUDED SECTION -->')
  41     return re.sub(
  42         rf'(?s){EXCLUDED_SECTION_BEGIN_STRING}.+?{EXCLUDED_SECTION_END_STRING}\n',
  43         '', readme)
  44
  45
  46 def _convert_code_blocks(readme):
  47     current_code_block = None
  48
  49     for line in readme.splitlines(True):
  50         if current_code_block:
  51             if line == current_code_block:
  52                 current_code_block = None
  53                 yield '\n'
  54             else:
  55                 yield f'    {line}'
  56         elif line.startswith('```'):
  57             current_code_block = line.count('`') * '`' + '\n'
  58             yield '\n'
  59         else:
  60             yield line
  61
  62
  63 def convert_code_blocks(readme):
  64     return ''.join(_convert_code_blocks(readme))
  65
  66
  67 def move_sections(readme):
  68     MOVE_TAG_TEMPLATE = '<!-- MANPAGE: MOVE "%s" SECTION HERE -->'
  69     sections = re.findall(r'(?m)^%s$' % (
  70         re.escape(MOVE_TAG_TEMPLATE).replace(r'\%', '%') % '(.+)'), readme)
  71
  72     for section_name in sections:
  73         move_tag = MOVE_TAG_TEMPLATE % section_name
  74         if readme.count(move_tag) > 1:
  75             raise Exception(f'There is more than one occurrence of "{move_tag}". This is unexpected')
  76
  77         sections = re.findall(rf'(?sm)(^# {re.escape(section_name)}.+?)(?=^# )', readme)
  78         if len(sections) < 1:
  79             raise Exception(f'The section {section_name} does not exist')
  80         elif len(sections) > 1:
  81             raise Exception(f'There are multiple occurrences of section {section_name}, this is unhandled')
  82
  83         readme = readme.replace(sections[0], '', 1).replace(move_tag, sections[0], 1)
  84     return readme
  85
  86
  87 def filter_options(readme):
  88     section = re.search(r'(?sm)^# USAGE AND OPTIONS\n.+?(?=^# )', readme).group(0)
  89     section_new = section.replace('*', R'\*')
  90
  91     options = '# OPTIONS\n'
  92     for line in section_new.split('\n')[1:]:
  93         mobj = re.fullmatch(r'''(?x)
  94                 \s{4}(?P<opt>-(?:,\s|[^\s])+)
  95                 (?:\s(?P<meta>(?:[^\s]|\s(?!\s))+))?
  96                 (\s{2,}(?P<desc>.+))?
  97             ''', line)
  98         if not mobj:
  99             options += f'{line.lstrip()}\n'
 100             continue
 101         option, metavar, description = mobj.group('opt', 'meta', 'desc')
 102
 103         # Pandoc's definition_lists. See http://pandoc.org/README.html
 104         option = f'{option} *{metavar}*' if metavar else option
 105         description = f'{description}\n' if description else ''
 106         options += f'\n{option}\n:   {description}'
 107         continue
 108
 109     return readme.replace(section, options, 1)
 110
 111
 112 TRANSFORM = compose_functions(filter_excluded_sections, convert_code_blocks, move_sections, filter_options)
 113
 114
 115 def main():
 116     write_file(get_filename_args(), PREFIX + TRANSFORM(read_file(README_FILE)))
 117
 118
 119 if __name__ == '__main__':
 120     main()