Git-Mirrors/brozzler
1
0
Fork 0
mirror of https://github.com/internetarchive/brozzler.git synced 2025-07-29 17:48:52 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

add missing imports and fix mimetype issue

Browse source
This commit is contained in:
Noah Levitt 2018-08-15 17:41:35 -07:00
parent 33520da8f9
commit c2ad8427e1
1 changed files with 18 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

14
brozzler/ydl.py
View file

@ -26,6 +26,8 @@ import tempfile
import urlcanon import urlcanon
import os import os
import json import json
import doublethink
import datetime
_orig_webpage_read_content = youtube_dl.extractor.generic.GenericIE._webpage_read_content _orig_webpage_read_content = youtube_dl.extractor.generic.GenericIE._webpage_read_content
def _webpage_read_content(self, *args, **kwargs): def _webpage_read_content(self, *args, **kwargs):
@ -133,12 +135,22 @@ def _build_youtube_dl(worker, destdir, site):
'extractor %r found a video in %s', ie.IE_NAME, url) 'extractor %r found a video in %s', ie.IE_NAME, url)
def _push_stitched_up_vid_to_warcprox(self, site, info_dict, ctx): def _push_stitched_up_vid_to_warcprox(self, site, info_dict, ctx):
# XXX Don't know how to get the right content-type. Youtube-dl
# doesn't supply it. Sometimes (with --hls-prefer-native)
# youtube-dl produces a stitched-up video that /usr/bin/file fails
# to identify (says "application/octet-stream"). `ffprobe` doesn't
# give us a mimetype.
if info_dict.get('ext') == 'mp4':
mimetype = 'video/mp4'
else:
try: try:
import magic import magic
mimetype = magic.from_file(ctx['filename'], mime=True) mimetype = magic.from_file(ctx['filename'], mime=True)
except ImportError as e: except ImportError as e:
mimetype = 'video/%s' % info_dict['ext'] mimetype = 'video/%s' % info_dict['ext']
self.logger.warn('guessing mimetype %s because %r', mimetype, e) self.logger.warn(
'guessing mimetype %s because %r', mimetype, e)
url = 'youtube-dl:%05d:%s' % ( url = 'youtube-dl:%05d:%s' % (
info_dict.get('playlist_index') or 1, info_dict.get('playlist_index') or 1,
info_dict['webpage_url']) info_dict['webpage_url'])