[viddler] Use API

This commit is contained in:
Philipp Hagemeister 2014-10-25 00:10:11 +02:00
parent cdc5cb7c2b
commit c64ed2a310
3 changed files with 71 additions and 36 deletions

View file

@ -145,7 +145,7 @@ def _repr(v):
info_dict_str = ''.join(
' %s: %s,\n' % (_repr(k), _repr(v))
for k, v in test_info_dict.items())
write_string('\n"info_dict": {' + info_dict_str + '}\n', out=sys.stderr)
write_string('\n"info_dict": {\n' + info_dict_str + '}\n', out=sys.stderr)
self.assertFalse(
missing_keys,
'Missing keys in test definition: %s' % (

View file

@ -89,6 +89,10 @@ class InfoExtractor(object):
format, irrespective of the file format.
-1 for default (order by other properties),
-2 or smaller for less than default.
* source_preference Order number for this video source
(quality takes higher priority)
-1 for default (order by other properties),
-2 or smaller for less than default.
* http_referer HTTP Referer header value to set.
* http_method HTTP method to use for the download.
* http_headers A dictionary of additional HTTP headers
@ -613,6 +617,7 @@ def _formats_key(f):
audio_ext_preference,
f.get('filesize') if f.get('filesize') is not None else -1,
f.get('filesize_approx') if f.get('filesize_approx') is not None else -1,
f.get('source_preference') if f.get('source_preference') is not None else -1,
f.get('format_id'),
)
formats.sort(key=_formats_key)

View file

@ -1,55 +1,85 @@
import json
import re
from __future__ import unicode_literals
from .common import InfoExtractor
from ..utils import (
float_or_none,
int_or_none,
)
class ViddlerIE(InfoExtractor):
_VALID_URL = r'(?P<domain>https?://(?:www\.)?viddler\.com)/(?:v|embed|player)/(?P<id>[a-z0-9]+)'
_VALID_URL = r'https?://(?:www\.)?viddler\.com/(?:v|embed|player)/(?P<id>[a-z0-9]+)'
_TEST = {
u"url": u"http://www.viddler.com/v/43903784",
u'file': u'43903784.mp4',
u'md5': u'fbbaedf7813e514eb7ca30410f439ac9',
u'info_dict': {
u"title": u"Video Made Easy",
u"uploader": u"viddler",
u"duration": 100.89,
"url": "http://www.viddler.com/v/43903784",
'md5': 'ae43ad7cb59431ce043f0ff7fa13cbf4',
'info_dict': {
'id': '43903784',
'ext': 'mp4',
"title": "Video Made Easy",
'description': 'You don\'t need to be a professional to make high-quality video content. Viddler provides some quick and easy tips on how to produce great video content with limited resources. ',
"uploader": "viddler",
'timestamp': 1335371429,
'upload_date': '20120425',
"duration": 100.89,
'thumbnail': 're:^https?://.*\.jpg$',
'view_count': int,
'categories': ['video content', 'high quality video', 'video made easy', 'how to produce video with limited resources', 'viddler'],
}
}
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
video_id = self._match_id(url)
embed_url = mobj.group('domain') + u'/embed/' + video_id
webpage = self._download_webpage(embed_url, video_id)
json_url = (
'http://api.viddler.com/api/v2/viddler.videos.getPlaybackDetails.json?video_id=%s&key=v0vhrt7bg2xq1vyxhkct' %
video_id)
data = self._download_json(json_url, video_id)['video']
video_sources_code = self._search_regex(
r"(?ms)sources\s*:\s*(\{.*?\})", webpage, u'video URLs')
video_sources = json.loads(video_sources_code.replace("'", '"'))
formats = []
for filed in data['files']:
if filed.get('status', 'ready') != 'ready':
continue
f = {
'format_id': filed['profile_id'],
'format_note': filed['profile_name'],
'url': self._proto_relative_url(filed['url']),
'width': int_or_none(filed.get('width')),
'height': int_or_none(filed.get('height')),
'filesize': int_or_none(filed.get('size')),
'ext': filed.get('ext'),
'source_preference': -1,
}
formats.append(f)
formats = [{
'url': video_url,
'format': format_id,
} for video_url, format_id in video_sources.items()]
if filed.get('cdn_url'):
f = f.copy()
f['url'] = self._proto_relative_url(filed['cdn_url'])
f['format_id'] = filed['profile_id'] + '-cdn'
f['source_preference'] = 1
formats.append(f)
title = self._html_search_regex(
r"title\s*:\s*'([^']*)'", webpage, u'title')
uploader = self._html_search_regex(
r"authorName\s*:\s*'([^']*)'", webpage, u'uploader', fatal=False)
duration_s = self._html_search_regex(
r"duration\s*:\s*([0-9.]*)", webpage, u'duration', fatal=False)
duration = float(duration_s) if duration_s else None
thumbnail = self._html_search_regex(
r"thumbnail\s*:\s*'([^']*)'",
webpage, u'thumbnail', fatal=False)
if filed.get('html5_video_source'):
f = f.copy()
f['url'] = self._proto_relative_url(
filed['html5_video_source'])
f['format_id'] = filed['profile_id'] + '-html5'
f['source_preference'] = 0
formats.append(f)
self._sort_formats(formats)
categories = [
t.get('text') for t in data.get('tags', []) if 'text' in t]
return {
'_type': 'video',
'id': video_id,
'title': title,
'thumbnail': thumbnail,
'uploader': uploader,
'duration': duration,
'title': data['title'],
'formats': formats,
'description': data.get('description'),
'timestamp': int_or_none(data.get('upload_time')),
'thumbnail': self._proto_relative_url(data.get('thumbnail_url')),
'uploader': data.get('author'),
'duration': float_or_none(data.get('length')),
'view_count': int_or_none(data.get('view_count')),
'categories': categories,
}