From 4c07546e7a5e5882abdda896009b744e947df1c4 Mon Sep 17 00:00:00 2001 From: James Taylor Date: Thu, 17 Oct 2019 19:58:13 -0700 Subject: Extraction: Replace youtube-dl with custom-built watch page extraction --- youtube_dl/postprocessor/metadatafromtitle.py | 48 --------------------------- 1 file changed, 48 deletions(-) delete mode 100644 youtube_dl/postprocessor/metadatafromtitle.py (limited to 'youtube_dl/postprocessor/metadatafromtitle.py') diff --git a/youtube_dl/postprocessor/metadatafromtitle.py b/youtube_dl/postprocessor/metadatafromtitle.py deleted file mode 100644 index f5c14d9..0000000 --- a/youtube_dl/postprocessor/metadatafromtitle.py +++ /dev/null @@ -1,48 +0,0 @@ -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.+)\ \-\ (?P<artist>.+)' - """ - lastpos = 0 - 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): - 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 -- cgit v1.2.3