[fix] yahoo url extraction

Browse Source
This commit is contained in:
Adam Tauber 2014-03-08 19:09:03 +01:00
parent aa9df5bf16
commit 0fc481f47e
2 changed files with 7 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
searx/engines/yahoo.py
View File

@ -15,6 +15,10 @@ suggestion_xpath = '//div[@id="satat"]//a'
paging = True
def parse_url(url_string):
start = url_string.find('http', url_string.find('/RU=')+1)
end = min(url_string.rfind('/RS'), url_string.rfind('/RK'))
return unquote(url_string[start:end])
def request(query, params):
offset = (params['pageno'] - 1) * 10 + 1
@ -34,10 +38,7 @@ def response(resp):
dom = html.fromstring(resp.text)
for result in dom.xpath(results_xpath):
url_string = extract_url(result.xpath(url_xpath), search_url)
start = url_string.find('http', url_string.find('/RU=')+1)
end = url_string.rfind('/RS')
url = unquote(url_string[start:end])
url = parse_url(extract_url(result.xpath(url_xpath), search_url))
title = extract_text(result.xpath(title_xpath)[0])
content = extract_text(result.xpath(content_xpath)[0])
results.append({'url': url, 'title': title, 'content': content})

7
searx/engines/yahoo_news.py
View File

@ -1,9 +1,9 @@
#!/usr/bin/env python
from urllib import urlencode
from urlparse import unquote
from lxml import html
from searx.engines.xpath import extract_text, extract_url
from searx.engines.yahoo import parse_url
categories = ['news']
search_url = 'http://news.search.yahoo.com/search?{query}&b={offset}'
@ -34,10 +34,7 @@ def response(resp):
dom = html.fromstring(resp.text)
for result in dom.xpath(results_xpath):
url_string = extract_url(result.xpath(url_xpath), search_url)
start = url_string.find('http', url_string.find('/RU=')+1)
end = url_string.rfind('/RS')
url = unquote(url_string[start:end])
url = parse_url(extract_url(result.xpath(url_xpath), search_url))
title = extract_text(result.xpath(title_xpath)[0])
content = extract_text(result.xpath(content_xpath)[0])
results.append({'url': url, 'title': title, 'content': content})