zaclys
/
searxng
mirror of https://github.com/searxng/searxng.git
1
0
Fork 1
Code Issues Packages Projects Releases Wiki Activity

[enh] xpath engine absolute xpath support

This commit is contained in:
asciimoo 2013-10-26 13:45:43 +02:00
parent c09d69bd2c
commit 5d764f95cf
1 changed files with 12 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
searx/engines/xpath.py
View File

@ -5,10 +5,10 @@ from cgi import escape
from lxml.etree import _ElementStringResult from lxml.etree import _ElementStringResult
search_url = None search_url = None
results_xpath = None
url_xpath = None url_xpath = None
content_xpath = None content_xpath = None
title_xpath = None title_xpath = None
results_xpath = ''
def extract_url(xpath_results): def extract_url(xpath_results):
url = '' url = ''
@ -26,7 +26,7 @@ def extract_url(xpath_results):
else: else:
url = xpath_results[0].attrib.get('href') url = xpath_results[0].attrib.get('href')
else: else:
raise Exception('Cannot handle xpath url resultset') url = xpath_results.attrib.get('href')
if not url.startswith('http://') or not url.startswith('https://'): if not url.startswith('http://') or not url.startswith('https://'):
url = 'http://'+url url = 'http://'+url
parsed_url = urlparse(url) parsed_url = urlparse(url)
@ -45,10 +45,15 @@ def response(resp):
results = [] results = []
dom = html.fromstring(resp.text) dom = html.fromstring(resp.text)
query = resp.search_params['query'] query = resp.search_params['query']
for result in dom.xpath(results_xpath): if results_xpath:
url = extract_url(result.xpath(url_xpath)) for result in dom.xpath(results_xpath):
title = ' '.join(result.xpath(title_xpath)) url = extract_url(result.xpath(url_xpath))
content = escape(' '.join(result.xpath(content_xpath))).replace(query, '<b>{0}</b>'.format(query)) title = ' '.join(result.xpath(title_xpath))
results.append({'url': url, 'title': title, 'content': content}) content = escape(' '.join(result.xpath(content_xpath))).replace(query, '<b>{0}</b>'.format(query))
results.append({'url': url, 'title': title, 'content': content})
else:
for content, url, title in zip(dom.xpath(content_xpath), map(extract_url, dom.xpath(url_xpath)), dom.xpath(title_xpath)):
results.append({'url': url, 'title': title, 'content': content})
return results return results