X-Git-Url: https://jfr.im/git/yt-dlp.git/blobdiff_plain/6cef08ad00b774425a22e4dddaa0586698e4ab1e..93240fc1848de4a94f25844c96e0dcd282ef1d3b:/devscripts/prepare_manpage.py

diff --git a/devscripts/prepare_manpage.py b/devscripts/prepare_manpage.py
index 843ade482..9b12e71e5 100644
--- a/devscripts/prepare_manpage.py
+++ b/devscripts/prepare_manpage.py
@@ -1,78 +1,96 @@
-from __future__ import unicode_literals
+#!/usr/bin/env python3
+
+# Allow direct execution
+import os
+import sys
+
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
 
-import io
-import optparse
 import os.path
 import re
 
+from devscripts.utils import (
+    compose_functions,
+    get_filename_args,
+    read_file,
+    write_file,
+)
+
 ROOT_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 README_FILE = os.path.join(ROOT_DIR, 'README.md')
 
-PREFIX = r'''%youtube-dlc(1)
+PREFIX = r'''%yt-dlp(1)
 
 # NAME
 
-youtube\-dl \- download videos from youtube.com or other video platforms
+yt\-dlp \- A youtube-dl fork with additional features and patches
 
 # SYNOPSIS
 
-**youtube-dlc** \[OPTIONS\] URL [URL...]
+**yt-dlp** \[OPTIONS\] URL [URL...]
+
+# DESCRIPTION
 
 '''
 
 
-def main():
-    parser = optparse.OptionParser(usage='%prog OUTFILE.md')
-    options, args = parser.parse_args()
-    if len(args) != 1:
-        parser.error('Expected an output filename')
+def filter_excluded_sections(readme):
+    EXCLUDED_SECTION_BEGIN_STRING = re.escape('<!-- MANPAGE: BEGIN EXCLUDED SECTION -->')
+    EXCLUDED_SECTION_END_STRING = re.escape('<!-- MANPAGE: END EXCLUDED SECTION -->')
+    return re.sub(
+        rf'(?s){EXCLUDED_SECTION_BEGIN_STRING}.+?{EXCLUDED_SECTION_END_STRING}\n',
+        '', readme)
 
-    outfile, = args
 
-    with io.open(README_FILE, encoding='utf-8') as f:
-        readme = f.read()
+def move_sections(readme):
+    MOVE_TAG_TEMPLATE = '<!-- MANPAGE: MOVE "%s" SECTION HERE -->'
+    sections = re.findall(r'(?m)^%s$' % (
+        re.escape(MOVE_TAG_TEMPLATE).replace(r'\%', '%') % '(.+)'), readme)
 
-    readme = re.sub(r'(?s)^.*?(?=# DESCRIPTION)', '', readme)
-    readme = re.sub(r'\s+youtube-dlc \[OPTIONS\] URL \[URL\.\.\.\]', '', readme)
-    readme = PREFIX + readme
+    for section_name in sections:
+        move_tag = MOVE_TAG_TEMPLATE % section_name
+        if readme.count(move_tag) > 1:
+            raise Exception(f'There is more than one occurrence of "{move_tag}". This is unexpected')
 
-    readme = filter_options(readme)
+        sections = re.findall(rf'(?sm)(^# {re.escape(section_name)}.+?)(?=^# )', readme)
+        if len(sections) < 1:
+            raise Exception(f'The section {section_name} does not exist')
+        elif len(sections) > 1:
+            raise Exception(f'There are multiple occurrences of section {section_name}, this is unhandled')
 
-    with io.open(outfile, 'w', encoding='utf-8') as outf:
-        outf.write(readme)
+        readme = readme.replace(sections[0], '', 1).replace(move_tag, sections[0], 1)
+    return readme
 
 
 def filter_options(readme):
-    ret = ''
-    in_options = False
-    for line in readme.split('\n'):
-        if line.startswith('# '):
-            if line[2:].startswith('OPTIONS'):
-                in_options = True
-            else:
-                in_options = False
-
-        if in_options:
-            if line.lstrip().startswith('-'):
-                split = re.split(r'\s{2,}', line.lstrip())
-                # Description string may start with `-` as well. If there is
-                # only one piece then it's a description bit not an option.
-                if len(split) > 1:
-                    option, description = split
-                    split_option = option.split(' ')
-
-                    if not split_option[-1].startswith('-'):  # metavar
-                        option = ' '.join(split_option[:-1] + ['*%s*' % split_option[-1]])
-
-                    # Pandoc's definition_lists. See http://pandoc.org/README.html
-                    # for more information.
-                    ret += '\n%s\n:   %s\n' % (option, description)
-                    continue
-            ret += line.lstrip() + '\n'
-        else:
-            ret += line + '\n'
-
-    return ret
+    section = re.search(r'(?sm)^# USAGE AND OPTIONS\n.+?(?=^# )', readme).group(0)
+    options = '# OPTIONS\n'
+    for line in section.split('\n')[1:]:
+        mobj = re.fullmatch(r'''(?x)
+                \s{4}(?P<opt>-(?:,\s|[^\s])+)
+                (?:\s(?P<meta>(?:[^\s]|\s(?!\s))+))?
+                (\s{2,}(?P<desc>.+))?
+            ''', line)
+        if not mobj:
+            options += f'{line.lstrip()}\n'
+            continue
+        option, metavar, description = mobj.group('opt', 'meta', 'desc')
+
+        # Pandoc's definition_lists. See http://pandoc.org/README.html
+        option = f'{option} *{metavar}*' if metavar else option
+        description = f'{description}\n' if description else ''
+        options += f'\n{option}\n:   {description}'
+        continue
+
+    return readme.replace(section, options, 1)
+
+
+TRANSFORM = compose_functions(filter_excluded_sections, move_sections, filter_options)
+
+
+def main():
+    write_file(get_filename_args(), PREFIX + TRANSFORM(read_file(README_FILE)))
 
 
 if __name__ == '__main__':