Mirrors/yt-dlp
2
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-14 23:32:33 +00:00
Code Issues Projects Releases Packages Wiki Activity

[flipagram] Improve extraction (Closes #9898)

Browse source
This commit is contained in:
Sergey M․ 2016-07-09 03:31:17 +07:00
parent 0de168f7ed
commit 0af985069b
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D
1 changed files with 78 additions and 66 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

144
youtube_dl/extractor/flipagram.py
View file

@ -2,102 +2,114 @@
from __future__ import unicode_literals
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
int_or_none,
parse_iso8601,
unified_strdate,
float_or_none,
try_get,
unified_timestamp,
)
class FlipagramIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?flipagram\.com/f/(?P<id>[^/?_]+)'
_TESTS = [{
'url': 'https://flipagram.com/f/myrWjW9RJw',
'md5': '541988fb6c4c7c375215ea22a4a21841',
'info_dict': {
'id': 'myrWjW9RJw',
'title': 'Flipagram by crystaldolce featuring King and Lionheart by Of Monsters and Men',
'description': 'Herbie\'s first bannana🍌🐢🍌. #animals #pets #reptile #tortoise #sulcata #tort #justatreat #snacktime #bannanas #rescuepets #ofmonstersandmen @animals',
'ext': 'mp4',
'uploader': 'Crystal Dolce',
'creator': 'Crystal Dolce',
'uploader_id': 'crystaldolce',
}
}, {
_VALID_URL = r'https?://(?:www\.)?flipagram\.com/f/(?P<id>[^/?#&]+)'
_TEST = {
'url': 'https://flipagram.com/f/nyvTSJMKId',
'only_matching': True,
}]
'md5': '888dcf08b7ea671381f00fab74692755',
'info_dict': {
'id': 'nyvTSJMKId',
'ext': 'mp4',
'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction',
'description': 'md5:d55e32edc55261cae96a41fa85ff630e',
'duration': 35.571,
'timestamp': 1461244995,
'upload_date': '20160421',
'uploader': 'kitty juria',
'uploader_id': 'sjuria101',
'creator': 'kitty juria',
'view_count': int,
'like_count': int,
'repost_count': int,
'comment_count': int,
'comments': list,
'formats': 'mincount:2',
},
}
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
self.report_extraction(video_id)
user_data = self._parse_json(self._search_regex(r'window.reactH2O\s*=\s*({.+});', webpage, 'user data'), video_id)
content_data = self._search_json_ld(webpage, video_id)
video_data = self._parse_json(
self._search_regex(
r'window\.reactH2O\s*=\s*({.+});', webpage, 'video data'),
video_id)
flipagram = user_data.get('flipagram', {})
counts = flipagram.get('counts', {})
user = flipagram.get('user', {})
video = flipagram.get('video', {})
flipagram = video_data['flipagram']
video = flipagram['video']
thumbnails = []
for cover in flipagram.get('covers', []):
if not cover.get('url'):
continue
thumbnails.append({
'url': self._proto_relative_url(cover.get('url')),
'width': int_or_none(cover.get('width')),
'height': int_or_none(cover.get('height')),
})
json_ld = self._search_json_ld(webpage, video_id, default=False)
title = json_ld.get('title') or flipagram['captionText']
description = json_ld.get('description') or flipagram.get('captionText')
# Note that this only retrieves comments that are initally loaded.
# For videos with large amounts of comments, most won't be retrieved.
comments = []
for comment in user_data.get('comments', {}).get(video_id, {}).get('items', []):
text = comment.get('comment', [])
comments.append({
'author': comment.get('user', {}).get('name'),
'author_id': comment.get('user', {}).get('username'),
'id': comment.get('id'),
'text': text[0] if text else '',
'timestamp': unified_timestamp(comment.get('created', '')),
})
formats = [{
'url': video['url'],
'width': int_or_none(video.get('width')),
'height': int_or_none(video.get('height')),
'filesize': int_or_none(video_data.get('size')),
}]
tags = [tag for item in flipagram['story'][1:] for tag in item]
formats = []
if flipagram.get('music', {}).get('track', {}).get('previewUrl', {}):
preview_url = try_get(
flipagram, lambda x: x['music']['track']['previewUrl'], compat_str)
if preview_url:
formats.append({
'url': flipagram.get('music').get('track').get('previewUrl'),
'url': preview_url,
'ext': 'm4a',
'vcodec': 'none',
})
formats.append({
'url': video.get('url'),
'ext': 'mp4',
'width': int_or_none(video.get('width')),
'height': int_or_none(video.get('height')),
'filesize': int_or_none(video.get('size')),
})
self._sort_formats(formats)
counts = flipagram.get('counts', {})
user = flipagram.get('user', {})
video_data = flipagram.get('video', {})
thumbnails = [{
'url': self._proto_relative_url(cover['url']),
'width': int_or_none(cover.get('width')),
'height': int_or_none(cover.get('height')),
'filesize': int_or_none(cover.get('size')),
} for cover in flipagram.get('covers', []) if cover.get('url')]
# Note that this only retrieves comments that are initally loaded.
# For videos with large amounts of comments, most won't be retrieved.
comments = []
for comment in video_data.get('comments', {}).get(video_id, {}).get('items', []):
text = comment.get('comment')
if not text or not isinstance(text, list):
continue
comments.append({
'author': comment.get('user', {}).get('name'),
'author_id': comment.get('user', {}).get('username'),
'id': comment.get('id'),
'text': text[0],
'timestamp': unified_timestamp(comment.get('created')),
})
return {
'id': video_id,
'title': content_data['title'],
'formats': formats,
'title': title,
'description': description,
'duration': float_or_none(flipagram.get('duration'), 1000),
'thumbnails': thumbnails,
'description': content_data.get('description'),
'timestamp': unified_timestamp(flipagram.get('iso8601Created')),
'uploader': user.get('name'),
'creator': user.get('name'),
'timestamp': parse_iso8601(flipagram.get('iso801Created')),
'upload_date': unified_strdate(flipagram.get('created')),
'uploader_id': user.get('username'),
'creator': user.get('name'),
'view_count': int_or_none(counts.get('plays')),
'like_count': int_or_none(counts.get('likes')),
'repost_count': int_or_none(counts.get('reflips')),
'comment_count': int_or_none(counts.get('comments')),
'comments': comments,
'tags': tags,
'formats': formats,
}