searx

commit: 2d42208e835105119db3bddce1efdf7b90757985
parent: 46277888340ceaa76233c57be23fb21591188e26
Author: Adam Tauber <asciimoo@gmail.com>
Date:   Tue,  4 Mar 2014 14:16:29 +0100

Merge pull request #43 from pointhi/news

Adding News-sites
Diffstat:
A
searx/engines/bing_news.py
51
+++++++++++++++++++++++++++++++++++++++++++++++++++
A
searx/engines/google_news.py
37
+++++++++++++++++++++++++++++++++++++
A
searx/engines/yahoo_news.py
51
+++++++++++++++++++++++++++++++++++++++++++++++++++
M
searx/settings.yml
13
+++++++++++++

4 files changed, 152 insertions(+), 0 deletions(-)
diff --git a/searx/engines/bing_news.py b/searx/engines/bing_news.py
@@ -0,0 +1,51 @@
+from urllib import urlencode
+from cgi import escape
+from lxml import html
+
+categories = ['news']
+
+base_url = 'http://www.bing.com/'
+search_string = 'news/search?{query}&first={offset}'
+paging = True
+language_support = True
+
+
+def request(query, params):
+    offset = (params['pageno'] - 1) * 10 + 1
+    if params['language'] == 'all':
+        language = 'en-US'
+    else:
+        language = params['language'].replace('_', '-')
+    search_path = search_string.format(
+        query=urlencode({'q': query, 'setmkt': language}),
+        offset=offset)
+
+    params['cookies']['SRCHHPGUSR'] = \
+        'NEWWND=0&NRSLT=-1&SRCHLANG=' + language.split('-')[0]
+    #if params['category'] == 'images':
+    # params['url'] = base_url + 'images/' + search_path
+    params['url'] = base_url + search_path
+    return params
+
+
+def response(resp):
+    global base_url
+    results = []
+    dom = html.fromstring(resp.content)
+    for result in dom.xpath('//div[@class="sa_cc"]'):
+        link = result.xpath('.//h3/a')[0]
+        url = link.attrib.get('href')
+        title = ' '.join(link.xpath('.//text()'))
+        content = escape(' '.join(result.xpath('.//p//text()')))
+        results.append({'url': url, 'title': title, 'content': content})
+
+    if results:
+        return results
+
+    for result in dom.xpath('//li[@class="b_algo"]'):
+        link = result.xpath('.//h2/a')[0]
+        url = link.attrib.get('href')
+        title = ' '.join(link.xpath('.//text()'))
+        content = escape(' '.join(result.xpath('.//p//text()')))
+        results.append({'url': url, 'title': title, 'content': content})
+    return results
diff --git a/searx/engines/google_news.py b/searx/engines/google_news.py
@@ -0,0 +1,37 @@
+#!/usr/bin/env python
+
+from urllib import urlencode
+from json import loads
+
+categories = ['news']
+
+url = 'https://ajax.googleapis.com/'
+search_url = url + 'ajax/services/search/news?v=2.0&start={offset}&rsz=large&safe=off&filter=off&{query}&hl={language}' # noqa
+
+paging = True
+language_support = True
+
+
+def request(query, params):
+    offset = (params['pageno'] - 1) * 8
+    language = 'en-US'
+    if params['language'] != 'all':
+        language = params['language'].replace('_', '-')
+    params['url'] = search_url.format(offset=offset,
+                                      query=urlencode({'q': query}),
+                                      language=language)
+    return params
+
+
+def response(resp):
+    results = []
+    search_res = loads(resp.text)
+
+    if not search_res.get('responseData', {}).get('results'):
+        return []
+
+    for result in search_res['responseData']['results']:
+        results.append({'url': result['unescapedUrl'],
+                        'title': result['titleNoFormatting'],
+                        'content': result['content']})
+    return results
diff --git a/searx/engines/yahoo_news.py b/searx/engines/yahoo_news.py
@@ -0,0 +1,51 @@
+#!/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
+
+categories = ['news']
+search_url = 'http://news.search.yahoo.com/search?{query}&b={offset}'
+results_xpath = '//div[@class="res"]'
+url_xpath = './/h3/a/@href'
+title_xpath = './/h3/a'
+content_xpath = './/div[@class="abstr"]'
+suggestion_xpath = '//div[@id="satat"]//a'
+
+paging = True
+
+
+def request(query, params):
+    offset = (params['pageno'] - 1) * 10 + 1
+    if params['language'] == 'all':
+        language = 'en'
+    else:
+        language = params['language'].split('_')[0]
+    params['url'] = search_url.format(offset=offset,
+                                      query=urlencode({'p': query}))
+    params['cookies']['sB'] = 'fl=1&vl=lang_{lang}&sh=1&rw=new&v=1'\
+        .format(lang=language)
+    return params
+
+
+def response(resp):
+    results = []
+    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('/RU=')+4
+        end = url_string.rfind('/RS')
+        url = unquote(url_string[start:end])
+        title = extract_text(result.xpath(title_xpath)[0])
+        content = extract_text(result.xpath(content_xpath)[0])
+        results.append({'url': url, 'title': title, 'content': content})
+
+    if not suggestion_xpath:
+        return results
+
+    for suggestion in dom.xpath(suggestion_xpath):
+        results.append({'suggestion': extract_text(suggestion)})
+
+    return results
diff --git a/searx/settings.yml b/searx/settings.yml
@@ -17,6 +17,11 @@ engines:
     locale : en-US
     shortcut : bi
 
+  - name : bing news
+    engine : bing_news
+    locale : en-US
+    shortcut : bin
+    
   - name : currency
     engine : currency_convert
     categories : general
@@ -61,6 +66,10 @@ engines:
     engine : google_images
     shortcut : goi
 
+  - name : google news
+    engine : google_news
+    shortcut : gon
+    
   - name : piratebay
     engine : piratebay
     categories : videos, music, files
@@ -112,6 +121,10 @@ engines:
     engine : yahoo
     shortcut : yh
 
+  - name : yahoo news
+    engine : yahoo_news
+    shortcut : yhn
+    
   - name : youtube
     engine : youtube
     categories : videos