diff options
Diffstat (limited to 'searx/engines/yahoo_news.py')
| -rw-r--r-- | searx/engines/yahoo_news.py | 20 |
1 files changed, 19 insertions, 1 deletions
diff --git a/searx/engines/yahoo_news.py b/searx/engines/yahoo_news.py index 35e323917..13a8a6024 100644 --- a/searx/engines/yahoo_news.py +++ b/searx/engines/yahoo_news.py @@ -4,6 +4,8 @@ from urllib import urlencode from lxml import html from searx.engines.xpath import extract_text, extract_url from searx.engines.yahoo import parse_url +from datetime import datetime, timedelta +import re categories = ['news'] search_url = 'http://news.search.yahoo.com/search?{query}&b={offset}' @@ -11,6 +13,7 @@ results_xpath = '//div[@class="res"]' url_xpath = './/h3/a/@href' title_xpath = './/h3/a' content_xpath = './/div[@class="abstr"]' +publishedDate_xpath = './/span[@class="timestamp"]' suggestion_xpath = '//div[@id="satat"]//a' paging = True @@ -37,7 +40,22 @@ def response(resp): 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}) + publishedDate = extract_text(result.xpath(publishedDate_xpath)[0]) + + if re.match("^[0-9]+ minute(s|) ago$", publishedDate): + publishedDate = datetime.now() - timedelta(minutes=int(re.match(r'\d+', publishedDate).group())) + else: + if re.match("^[0-9]+ hour(s|), [0-9]+ minute(s|) ago$", publishedDate): + timeNumbers = re.findall(r'\d+', publishedDate) + publishedDate = datetime.now() - timedelta(hours=int(timeNumbers[0])) - timedelta(minutes=int(timeNumbers[1])) + else: + # TODO year in string possible? + publishedDate = datetime.strptime(publishedDate,"%b %d %H:%M%p") + + if publishedDate.year == 1900: + publishedDate = publishedDate.replace(year=datetime.now().year) + + results.append({'url': url, 'title': title, 'content': content,'publishedDate':publishedDate}) if not suggestion_xpath: return results |