PublicMirrors/you-get
1
0
Fork 0
mirror of https://github.com/soimort/you-get.git synced 2025-01-24 14:05:01 +03:00
Code Issues Actions Packages Projects Releases 1 Wiki Activity

[lizhi] overhaul

Browse Source 
Lizhi extractor has stopped working. In particular, there are two major
changes:

- URL format change: no more #/ in URL paths;
- The /api/audio/{radio_id}/{audio_id} API now returns 404.

This is a rewrite based on the /api/radio_audios API.
This commit is contained in:
Zhiming Wang 2016-12-04 19:36:17 -05:00
parent 61d9bf124e
commit 606e0a786e
No known key found for this signature in database
GPG Key ID: BBD31D4D110044B8
1 changed files with 46 additions and 28 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

74
src/you_get/extractors/lizhi.py
View File

@ -4,37 +4,55 @@ __all__ = ['lizhi_download']
import json
from ..common import *
def lizhi_download_playlist(url, output_dir = '.', merge = True, info_only = False, **kwargs):
# like this http://www.lizhi.fm/#/31365/
#api desc: s->start l->length band->some radio
#http://www.lizhi.fm/api/radio_audios?s=0&l=100&band=31365
band_id = match1(url,r'#/(\d+)')
#try to get a considerable large l to reduce html parsing task.
api_url = 'http://www.lizhi.fm/api/radio_audios?s=0&l=65535&band='+band_id
content_json = json.loads(get_content(api_url))
for sound in content_json:
title = sound["name"]
res_url = sound["url"]
songtype, ext, size = url_info(res_url,faker=True)
print_info(site_info, title, songtype, size)
if not info_only:
#no referer no speed!
download_urls([res_url], title, ext, size, output_dir, merge=merge ,refer = 'http://www.lizhi.fm',faker=True)
pass
# radio_id: e.g. 549759 from http://www.lizhi.fm/549759/
#
# Returns a list of tuples (audio_id, title, url) for each episode
# (audio) in the radio playlist. url is the direct link to the audio
# file.
def lizhi_extract_playlist_info(radio_id):
# /api/radio_audios API parameters:
#
# - s: starting episode
# - l: count (per page)
# - band: radio_id
#
# We use l=65535 for poor man's pagination (that is, no pagination
# at all -- hope all fits on a single page).
#
# TODO: Use /api/radio?band={radio_id} to get number of episodes
# (au_cnt), then handle pagination properly.
api_url = 'http://www.lizhi.fm/api/radio_audios?s=0&l=65535&band=%s' % radio_id
api_response = json.loads(get_content(api_url))
return [(ep['id'], ep['name'], ep['url']) for ep in api_response]
def lizhi_download(url, output_dir = '.', merge = True, info_only = False, **kwargs):
# url like http://www.lizhi.fm/#/549759/18864883431656710
api_id = match1(url,r'#/(\d+/\d+)')
api_url = 'http://www.lizhi.fm/api/audio/'+api_id
content_json = json.loads(get_content(api_url))
title = content_json["audio"]["name"]
res_url = content_json["audio"]["url"]
songtype, ext, size = url_info(res_url,faker=True)
print_info(site_info, title, songtype, size)
def lizhi_download_audio(audio_id, title, url, output_dir='.', info_only=False):
filetype, ext, size = url_info(url)
print_info(site_info, title, filetype, size)
if not info_only:
#no referer no speed!
download_urls([res_url], title, ext, size, output_dir, merge=merge ,refer = 'http://www.lizhi.fm',faker=True)
download_urls([url], title, ext, size, output_dir=output_dir)
def lizhi_download_playlist(url, output_dir='.', info_only=False, **kwargs):
# Sample URL: http://www.lizhi.fm/549759/
radio_id = match1(url,r'/(\d+)')
if not radio_id:
raise NotImplementedError('%s not supported' % url)
for audio_id, title, url in lizhi_extract_playlist_info(radio_id):
lizhi_download_audio(audio_id, title, url, output_dir=output_dir, info_only=info_only)
def lizhi_download(url, output_dir='.', info_only=False, **kwargs):
# Sample URL: http://www.lizhi.fm/549759/18864883431656710/
m = re.search(r'/(?P<radio_id>\d+)/(?P<audio_id>\d+)', url)
if not m:
raise NotImplementedError('%s not supported' % url)
radio_id = m.group('radio_id')
audio_id = m.group('audio_id')
# Look for the audio_id among the full list of episodes
for aid, title, url in lizhi_extract_playlist_info(radio_id):
if aid == audio_id:
lizhi_download_audio(audio_id, title, url, output_dir=output_dir, info_only=info_only)
break
else:
raise NotImplementedError('Audio #%s not found in playlist #%s' % (audio_id, radio_id))
site_info = "lizhi.fm"
download = lizhi_download