Ponysearch/searx/engines/soundcloud.py

110 lines
3.2 KiB
Python
Raw Normal View History

"""
Soundcloud (Music)
@website https://soundcloud.com
@provide-api yes (https://developers.soundcloud.com/)
@using-api yes
@results JSON
@stable yes
@parse url, title, content, publishedDate, embedded
"""
2014-09-02 18:12:30 +02:00
import re
2013-10-17 21:21:23 +02:00
from json import loads
2016-11-30 18:43:03 +01:00
from lxml import html
from dateutil import parser
from searx import logger
from searx.poolrequests import get as http_get
2016-11-30 18:43:03 +01:00
from searx.url_utils import quote_plus, urlencode
try:
from cStringIO import StringIO
except:
from io import StringIO
2013-10-17 21:21:23 +02:00
2014-09-02 18:12:30 +02:00
# engine dependent config
2013-10-17 21:21:23 +02:00
categories = ['music']
2014-09-02 18:12:30 +02:00
paging = True
2013-10-17 21:21:23 +02:00
2014-09-02 18:12:30 +02:00
# search-url
# missing attribute: user_id, app_version, app_locale
url = 'https://api-v2.soundcloud.com/'
2014-12-16 17:26:16 +01:00
search_url = url + 'search?{query}'\
'&variant_ids='\
2014-12-16 17:26:16 +01:00
'&facet=model'\
'&limit=20'\
'&offset={offset}'\
'&linked_partitioning=1'\
'&client_id={client_id}' # noqa
2014-01-20 02:31:20 +01:00
embedded_url = '<iframe width="100%" height="166" ' +\
'scrolling="no" frameborder="no" ' +\
'data-src="https://w.soundcloud.com/player/?url={uri}"></iframe>'
2016-11-30 18:43:03 +01:00
cid_re = re.compile(r'client_id:"([^"]*)"', re.I | re.U)
guest_client_id = ''
2016-11-30 18:43:03 +01:00
2013-10-17 21:21:23 +02:00
def get_client_id():
response = http_get("https://soundcloud.com")
if response.ok:
2016-11-30 18:43:03 +01:00
tree = html.fromstring(response.content)
script_tags = tree.xpath("//script[contains(@src, '/assets/app')]")
app_js_urls = [script_tag.get('src') for script_tag in script_tags if script_tag is not None]
# extracts valid app_js urls from soundcloud.com content
for app_js_url in app_js_urls:
# gets app_js and searches for the clientid
response = http_get(app_js_url)
if response.ok:
cids = cid_re.search(response.content.decode("utf-8"))
if cids is not None and len(cids.groups()):
return cids.groups()[0]
logger.warning("Unable to fetch guest client_id from SoundCloud, check parser!")
return ""
def init(engine_settings=None):
global guest_client_id
# api-key
guest_client_id = get_client_id()
2014-09-02 18:12:30 +02:00
# do search-request
2013-10-17 21:21:23 +02:00
def request(query, params):
2014-01-30 01:50:15 +01:00
offset = (params['pageno'] - 1) * 20
2014-09-02 18:12:30 +02:00
2014-01-30 01:50:15 +01:00
params['url'] = search_url.format(query=urlencode({'q': query}),
2014-09-02 18:12:30 +02:00
offset=offset,
client_id=guest_client_id)
2013-10-17 21:21:23 +02:00
return params
2014-09-02 18:12:30 +02:00
# get response from search-request
2013-10-17 21:21:23 +02:00
def response(resp):
results = []
2014-09-02 18:12:30 +02:00
2013-10-17 21:21:23 +02:00
search_res = loads(resp.text)
2014-09-02 18:12:30 +02:00
# parse results
2013-10-17 21:21:23 +02:00
for result in search_res.get('collection', []):
2013-10-20 00:52:32 +02:00
if result['kind'] in ('track', 'playlist'):
2013-10-17 21:21:23 +02:00
title = result['title']
content = result['description']
publishedDate = parser.parse(result['last_modified'])
uri = quote_plus(result['uri'])
embedded = embedded_url.format(uri=uri)
2014-09-02 18:12:30 +02:00
# append result
2014-01-20 02:31:20 +01:00
results.append({'url': result['permalink_url'],
'title': title,
'publishedDate': publishedDate,
'embedded': embedded,
2014-01-20 02:31:20 +01:00
'content': content})
2014-09-02 18:12:30 +02:00
# return results
2013-10-17 21:21:23 +02:00
return results