commit: 54856480d7bac670c9d571d4191f5f35aadc5270
parent 1dd12708c2042e5d78887e4c5026d6196cc02bb2
Author: Remita Amine <remitamine@gmail.com>
Date: Tue, 19 Jan 2021 10:23:02 +0100
[ninegag] improve extraction
Diffstat:
1 file changed, 83 insertions(+), 106 deletions(-)
diff --git a/youtube_dl/extractor/ninegag.py b/youtube_dl/extractor/ninegag.py
@@ -1,148 +1,125 @@
from __future__ import unicode_literals
-import re
-
from .common import InfoExtractor
from ..utils import (
determine_ext,
- url_or_none,
+ ExtractorError,
int_or_none,
- float_or_none,
- ExtractorError
+ try_get,
+ url_or_none,
)
class NineGagIE(InfoExtractor):
IE_NAME = '9gag'
- _VALID_URL = r'https?://(?:www\.)?9gag\.com/gag/(?P<id>[a-zA-Z0-9]+)'
+ _VALID_URL = r'https?://(?:www\.)?9gag\.com/gag/(?P<id>[^/?&#]+)'
- _TESTS = [{
- 'url': 'https://9gag.com/gag/an5Qz5b',
- 'info_dict': {
- 'id': 'an5Qz5b',
- 'ext': 'webm',
- 'title': 'Dogs playing tetherball',
- 'upload_date': '20191108',
- 'timestamp': 1573243994,
- 'age_limit': 0,
- 'categories': [
- 'Wholesome'
- ],
- 'tags': [
- 'Dog'
- ]
- }
- }, {
+ _TEST = {
'url': 'https://9gag.com/gag/ae5Ag7B',
'info_dict': {
'id': 'ae5Ag7B',
- 'ext': 'webm',
+ 'ext': 'mp4',
'title': 'Capybara Agility Training',
'upload_date': '20191108',
'timestamp': 1573237208,
- 'age_limit': 0,
- 'categories': [
- 'Awesome'
- ],
- 'tags': [
- 'Weimaraner',
- 'American Pit Bull Terrier'
- ]
+ 'categories': ['Awesome'],
+ 'tags': ['Weimaraner', 'American Pit Bull Terrier'],
+ 'duration': 44,
+ 'like_count': int,
+ 'dislike_count': int,
+ 'comment_count': int,
}
- }]
-
- _EXTERNAL_VIDEO_PROVIDERS = {
- 'Youtube': 'https://youtube.com/watch?v=%s'
}
def _real_extract(self, url):
- video_id = self._match_id(url)
- webpage = self._download_webpage(url, video_id)
- rawJsonData = self._search_regex(
- r'window._config\s*=\s*JSON.parse\(["\']({.+?})["\']\);',
- webpage,
- 'data')
- rawJsonData = rawJsonData.replace('\\"', '"').replace('\\\\/', '/')
- data = self._parse_json(rawJsonData, video_id)['data']['post']
-
- if data['type'] == 'Video':
- vid = data['video']['id']
- ie_key = data['video']['source'].capitalize()
- return {
- '_type': 'url_transparent',
- 'url': self._EXTERNAL_VIDEO_PROVIDERS[ie_key] % vid,
- 'ie_key': ie_key,
- 'id': vid,
- 'duration': data['video'].get('duration'),
- 'start_time': data['video'].get('startTs')
- }
+ post_id = self._match_id(url)
+ post = self._download_json(
+ 'https://9gag.com/v1/post', post_id, query={
+ 'id': post_id
+ })['data']['post']
- if data['type'] == 'EmbedVideo':
- vid = data['video']['id']
- ie_key = data['video']['source'].capitalize()
- return {
- '_type': 'url_transparent',
- 'url': data['video']['embedUrl'],
- #'ie_key': vid,
- 'start_time': data['video'].get('startTs')
- }
-
- if data['type'] != 'Animated':
+ if post.get('type') != 'Animated':
raise ExtractorError(
'The given url does not contain a video',
expected=True)
+ title = post['title']
+
duration = None
formats = []
thumbnails = []
- for key in data['images']:
- image = data['images'][key]
- if 'duration' in image and duration is None:
- duration = int_or_none(image['duration'])
- url = url_or_none(image.get('url'))
- if url == None:
+ for key, image in (post.get('images') or {}).items():
+ image_url = url_or_none(image.get('url'))
+ if not image_url:
continue
- ext = determine_ext(url)
- if ext == 'jpg' or ext == 'png':
- thumbnail = {
- 'url': url,
- 'width': float_or_none(image.get('width')),
- 'height': float_or_none(image.get('height'))
- }
- thumbnails.append(thumbnail)
- elif ext == 'webm' or ext == 'mp4':
- formats.append({
- 'format_id': re.sub(r'.*_([^\.]+).(.*)', r'\1_\2', url),
+ ext = determine_ext(image_url)
+ image_id = key.strip('image')
+ common = {
+ 'url': image_url,
+ 'width': int_or_none(image.get('width')),
+ 'height': int_or_none(image.get('height')),
+ }
+ if ext in ('jpg', 'png'):
+ webp_url = image.get('webpUrl')
+ if webp_url:
+ t = common.copy()
+ t.update({
+ 'id': image_id + '-webp',
+ 'url': webp_url,
+ })
+ thumbnails.append(t)
+ common.update({
+ 'id': image_id,
'ext': ext,
- 'url': url,
- 'width': float_or_none(image.get('width')),
- 'height': float_or_none(image.get('height'))
})
- section = None
- postSection = data.get('postSection')
- if postSection != None and 'name' in postSection:
- section = re.sub(r'\\[^\\]{5}', '', postSection['name'])
- age_limit = int_or_none(data.get('nsfw'))
- if age_limit != None:
- age_limit = age_limit * 18
+ thumbnails.append(common)
+ elif ext in ('webm', 'mp4'):
+ if not duration:
+ duration = int_or_none(image.get('duration'))
+ common['acodec'] = 'none' if image.get('hasAudio') == 0 else None
+ for vcodec in ('vp8', 'vp9', 'h265'):
+ c_url = image.get(vcodec + 'Url')
+ if not c_url:
+ continue
+ c_f = common.copy()
+ c_f.update({
+ 'format_id': image_id + '-' + vcodec,
+ 'url': c_url,
+ 'vcodec': vcodec,
+ })
+ formats.append(c_f)
+ common.update({
+ 'ext': ext,
+ 'format_id': image_id,
+ })
+ formats.append(common)
+ self._sort_formats(formats)
+
+ section = try_get(post, lambda x: x['postSection']['name'])
+
tags = None
- if 'tags' in data:
+ post_tags = post.get('tags')
+ if post_tags:
tags = []
- for tag in data.get('tags') or []:
- tags.append(tag.get('key'))
+ for tag in post_tags:
+ tag_key = tag.get('key')
+ if not tag_key:
+ continue
+ tags.append(tag_key)
+
+ get_count = lambda x: int_or_none(post.get(x + 'Count'))
return {
- 'id': video_id,
- 'title': data['title'],
- 'timestamp': int_or_none(data.get('creationTs')),
+ 'id': post_id,
+ 'title': title,
+ 'timestamp': int_or_none(post.get('creationTs')),
'duration': duration,
'formats': formats,
'thumbnails': thumbnails,
- 'like_count': int_or_none(data.get('upVoteCount')),
- 'dislike_count': int_or_none(data.get('downVoteCount')),
- 'comment_count': int_or_none(data.get('commentsCount')),
- 'age_limit': age_limit,
- 'categories': [section],
+ 'like_count': get_count('upVote'),
+ 'dislike_count': get_count('downVote'),
+ 'comment_count': get_count('comments'),
+ 'age_limit': 18 if post.get('nsfw') == 1 else None,
+ 'categories': [section] if section else None,
'tags': tags,
- 'is_live': False
}
