yt-dlp/youtube_dl/extractor/imgur.py
Abhishek Kedia dbee18b552 Improve extraction (Closes #7918)
remove outer parentheses in if

Conflicts:
	youtube_dl/extractor/imgur.py

checked code with flake8

not returning list in case of single images.

using the fact that id with length 5 are albums and more are single videos.
Also for single videos ie ImgurIE both urls - http://imgur.com/gallery/oWeAMW2 and http://imgur.com/oWeAMW2 are equally fine. Change regex to allow thuis.
For albums urls - http://imgur.com/gallery/Q95ko and http://imgur.com/Q95ko are ok. Change regex to allow this also.

update description in ImgurIE Tests.
Also move single video test 'https://imgur.com/gallery/YcAQlkx' from ImgurAlbumIE to ImgurIE.
2015-12-22 21:43:49 +06:00

137 lines
4.5 KiB
Python

from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..compat import compat_urlparse
from ..utils import (
int_or_none,
js_to_json,
mimetype2ext,
ExtractorError,
)
class ImgurIE(InfoExtractor):
_VALID_URL = r'https?://(?:i\.)?imgur\.com/(gallery/)?(?P<id>[a-zA-Z0-9]{6,})'
_TESTS = [{
'url': 'https://i.imgur.com/A61SaA1.gifv',
'info_dict': {
'id': 'A61SaA1',
'ext': 'mp4',
'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
'description': 'Imgur: The most awesome images on the Internet.',
},
}, {
'url': 'https://imgur.com/A61SaA1',
'info_dict': {
'id': 'A61SaA1',
'ext': 'mp4',
'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
'description': 'Imgur: The most awesome images on the Internet.',
},
}, {
'url': 'https://imgur.com/gallery/YcAQlkx',
'info_dict': {
'id': 'YcAQlkx',
'ext': 'mp4',
'title': 'Classic Steve Carell gif...cracks me up everytime....damn the repost downvotes....',
'description': 'Imgur: The most awesome images on the Internet.'
}
}]
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(
compat_urlparse.urljoin(url, video_id), video_id)
width = int_or_none(self._search_regex(
r'<param name="width" value="([0-9]+)"',
webpage, 'width', fatal=False))
height = int_or_none(self._search_regex(
r'<param name="height" value="([0-9]+)"',
webpage, 'height', fatal=False))
video_elements = self._search_regex(
r'(?s)<div class="video-elements">(.*?)</div>',
webpage, 'video elements', default=None)
if not video_elements:
raise ExtractorError(
'No sources found for video %s. Maybe an image?' % video_id,
expected=True)
formats = []
for m in re.finditer(r'<source\s+src="(?P<src>[^"]+)"\s+type="(?P<type>[^"]+)"', video_elements):
formats.append({
'format_id': m.group('type').partition('/')[2],
'url': self._proto_relative_url(m.group('src')),
'ext': mimetype2ext(m.group('type')),
'acodec': 'none',
'width': width,
'height': height,
'http_headers': {
'User-Agent': 'youtube-dl (like wget)',
},
})
gif_json = self._search_regex(
r'(?s)var\s+videoItem\s*=\s*(\{.*?\})',
webpage, 'GIF code', fatal=False)
if gif_json:
gifd = self._parse_json(
gif_json, video_id, transform_source=js_to_json)
formats.append({
'format_id': 'gif',
'preference': -10,
'width': width,
'height': height,
'ext': 'gif',
'acodec': 'none',
'vcodec': 'gif',
'container': 'gif',
'url': self._proto_relative_url(gifd['gifUrl']),
'filesize': gifd.get('size'),
'http_headers': {
'User-Agent': 'youtube-dl (like wget)',
},
})
self._sort_formats(formats)
return {
'id': video_id,
'formats': formats,
'description': self._og_search_description(webpage),
'title': self._og_search_title(webpage),
}
class ImgurAlbumIE(InfoExtractor):
_VALID_URL = r'https?://(?:i\.)?imgur\.com/(gallery/)?(?P<id>[a-zA-Z0-9]{5})(?![a-zA-Z0-9])'
_TEST = {
'url': 'http://imgur.com/gallery/Q95ko',
'info_dict': {
'id': 'Q95ko',
},
'playlist_count': 25,
}
def _real_extract(self, url):
album_id = self._match_id(url)
album_img_data = self._download_json(
'http://imgur.com/gallery/%s/album_images/hit.json?all=true' % album_id, album_id)['data']
if len(album_img_data) == 0:
return self.url_result('http://imgur.com/%s' % album_id)
else:
album_images = album_img_data['images']
entries = [
self.url_result('http://imgur.com/%s' % image['hash'])
for image in album_images if image.get('hash')]
return self.playlist_result(entries, album_id)