2013-12-30 21:42:37 +00:00
|
|
|
from urllib import urlencode
|
|
|
|
from json import loads
|
2014-02-05 19:24:31 +00:00
|
|
|
from lxml import html
|
2013-12-30 21:42:37 +00:00
|
|
|
|
|
|
|
categories = ['videos']
|
2014-01-05 12:55:17 +00:00
|
|
|
locale = 'en_US'
|
2013-12-30 21:42:37 +00:00
|
|
|
|
|
|
|
# see http://www.dailymotion.com/doc/api/obj-video.html
|
2014-01-29 23:01:42 +00:00
|
|
|
search_url = 'https://api.dailymotion.com/videos?fields=title,description,duration,url,thumbnail_360_url&sort=relevance&limit=25&page={pageno}&{query}' # noqa
|
2014-01-20 01:31:20 +00:00
|
|
|
|
|
|
|
# TODO use video result template
|
|
|
|
content_tpl = '<a href="{0}" title="{0}" ><img src="{1}" /></a><br />'
|
2013-12-30 21:42:37 +00:00
|
|
|
|
2014-01-29 23:01:42 +00:00
|
|
|
paging = True
|
|
|
|
|
2014-01-19 21:59:01 +00:00
|
|
|
|
2013-12-30 21:42:37 +00:00
|
|
|
def request(query, params):
|
2014-01-19 21:59:01 +00:00
|
|
|
params['url'] = search_url.format(
|
2014-01-29 23:01:42 +00:00
|
|
|
query=urlencode({'search': query, 'localization': locale}),
|
|
|
|
pageno=params['pageno'])
|
2013-12-30 21:42:37 +00:00
|
|
|
return params
|
|
|
|
|
|
|
|
|
|
|
|
def response(resp):
|
|
|
|
results = []
|
|
|
|
search_res = loads(resp.text)
|
|
|
|
if not 'list' in search_res:
|
|
|
|
return results
|
|
|
|
for res in search_res['list']:
|
|
|
|
title = res['title']
|
|
|
|
url = res['url']
|
|
|
|
if res['thumbnail_360_url']:
|
2014-01-20 01:31:20 +00:00
|
|
|
content = content_tpl.format(url, res['thumbnail_360_url'])
|
2013-12-30 21:42:37 +00:00
|
|
|
else:
|
|
|
|
content = ''
|
|
|
|
if res['description']:
|
2014-01-05 12:55:17 +00:00
|
|
|
description = text_content_from_html(res['description'])
|
|
|
|
content += description[:500]
|
2013-12-30 21:42:37 +00:00
|
|
|
results.append({'url': url, 'title': title, 'content': content})
|
|
|
|
return results
|
2014-01-05 12:55:17 +00:00
|
|
|
|
2014-01-19 21:59:01 +00:00
|
|
|
|
2014-01-05 12:55:17 +00:00
|
|
|
def text_content_from_html(html_string):
|
|
|
|
desc_html = html.fragment_fromstring(html_string, create_parent=True)
|
|
|
|
return desc_html.text_content()
|