0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-18 01:35:12 +00:00
yt-dlp/youtube_dlc/postprocessor/metadatafromtitle.py

49 lines
1.6 KiB
Python
Raw Normal View History

from __future__ import unicode_literals
import re
from .common import PostProcessor
class MetadataFromTitlePP(PostProcessor):
def __init__(self, downloader, titleformat):
super(MetadataFromTitlePP, self).__init__(downloader)
self._titleformat = titleformat
self._titleregex = (self.format_to_regex(titleformat)
if re.search(r'%\(\w+\)s', titleformat)
else titleformat)
def format_to_regex(self, fmt):
r"""
Converts a string like
'%(title)s - %(artist)s'
to a regex like
'(?P<title>.+)\ \-\ (?P<artist>.+)'
"""
lastpos = 0
2016-02-14 09:37:17 +00:00
regex = ''
# replace %(..)s with regex group and escape other string parts
for match in re.finditer(r'%\((\w+)\)s', fmt):
regex += re.escape(fmt[lastpos:match.start()])
regex += r'(?P<' + match.group(1) + '>.+)'
lastpos = match.end()
if lastpos < len(fmt):
2017-04-12 19:38:43 +00:00
regex += re.escape(fmt[lastpos:])
return regex
def run(self, info):
title = info['title']
match = re.match(self._titleregex, title)
if match is None:
self._downloader.to_screen(
'[fromtitle] Could not interpret title of video as "%s"'
% self._titleformat)
return [], info
for attribute, value in match.groupdict().items():
info[attribute] = value
self._downloader.to_screen(
'[fromtitle] parsed %s: %s'
% (attribute, value if value is not None else 'NA'))
return [], info