commit: 018a14431bd3612db4e8840ce24f3e60026ece0f
parent: faed14b2c691746ba6cf98d164a5e6b1ca3ee4c9
Author: Adam Tauber <asciimoo@gmail.com>
Date: Tue, 18 Mar 2014 18:20:10 +0100
Merge pull request #57 from pointhi/results
improving publishDate extraction and output of it
Diffstat:
9 files changed, 26 insertions(+), 15 deletions(-)
diff --git a/requirements.txt b/requirements.txt
@@ -3,3 +3,4 @@ flask-babel
grequests
lxml
pyyaml
+python-dateutil
diff --git a/searx/engines/google_news.py b/searx/engines/google_news.py
@@ -2,6 +2,7 @@
from urllib import urlencode
from json import loads
+from dateutil import parser
from datetime import datetime
categories = ['news']
@@ -32,16 +33,9 @@ def response(resp):
return []
for result in search_res['responseData']['results']:
-# S.149 (159), library.pdf
-# datetime.strptime("Mon, 10 Mar 2014 16:26:15 -0700",
-# "%a, %d %b %Y %H:%M:%S %z")
-# publishedDate = parse(result['publishedDate'])
- publishedDate = datetime.strptime(
- str.join(' ', result['publishedDate'].split(None)[0:5]),
- "%a, %d %b %Y %H:%M:%S")
- #utc_offset = timedelta(result['publishedDate'].split(None)[5])
- # local = utc + offset
- #publishedDate = publishedDate + utc_offset
+
+# Mon, 10 Mar 2014 16:26:15 -0700
+ publishedDate = parser.parse(result['publishedDate'])
results.append({'url': result['unescapedUrl'],
'title': result['titleNoFormatting'],
diff --git a/searx/engines/vimeo.py b/searx/engines/vimeo.py
@@ -2,6 +2,8 @@ from urllib import urlencode
from HTMLParser import HTMLParser
from lxml import html
from xpath import extract_text
+from datetime import datetime
+from dateutil import parser
base_url = 'http://vimeo.com'
search_url = base_url + '/search?{query}'
@@ -10,6 +12,7 @@ content_xpath = None
title_xpath = None
results_xpath = ''
content_tpl = '<a href="{0}"> <img src="{2}"/> </a>'
+publishedDate_xpath = './/p[@class="meta"]//attribute::datetime'
# the cookie set by vimeo contains all the following values,
# but only __utma seems to be requiered
@@ -40,9 +43,12 @@ def response(resp):
url = base_url + result.xpath(url_xpath)[0]
title = p.unescape(extract_text(result.xpath(title_xpath)))
thumbnail = extract_text(result.xpath(content_xpath)[0])
+ publishedDate = parser.parse(extract_text(result.xpath(publishedDate_xpath)[0]))
+
results.append({'url': url,
'title': title,
'content': content_tpl.format(url, title, thumbnail),
'template': 'videos.html',
+ 'publishedDate': publishedDate,
'thumbnail': thumbnail})
return results
diff --git a/searx/engines/yahoo_news.py b/searx/engines/yahoo_news.py
@@ -6,6 +6,7 @@ from searx.engines.xpath import extract_text, extract_url
from searx.engines.yahoo import parse_url
from datetime import datetime, timedelta
import re
+from dateutil import parser
categories = ['news']
search_url = 'http://news.search.yahoo.com/search?{query}&b={offset}'
@@ -52,9 +53,7 @@ def response(resp):
- 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")
+ publishedDate = parser.parse(publishedDate)
if publishedDate.year == 1900:
publishedDate = publishedDate.replace(year=datetime.now().year)
diff --git a/searx/engines/youtube.py b/searx/engines/youtube.py
@@ -1,5 +1,7 @@
from json import loads
from urllib import urlencode
+from dateutil import parser
+from datetime import datetime
categories = ['videos']
@@ -35,6 +37,10 @@ def response(resp):
content = ''
thumbnail = ''
+#"2013-12-31T15:22:51.000Z"
+ pubdate = result['published']['$t']
+ publishedDate = parser.parse(pubdate)
+
if result['media$group']['media$thumbnail']:
thumbnail = result['media$group']['media$thumbnail'][0]['url']
content += '<a href="{0}" title="{0}" ><img src="{1}" /></a>'.format(url, thumbnail) # noqa
@@ -48,6 +54,7 @@ def response(resp):
'title': title,
'content': content,
'template': 'videos.html',
+ 'publishedDate': publishedDate,
'thumbnail': thumbnail})
return results
diff --git a/searx/templates/opensearch_response_rss.xml b/searx/templates/opensearch_response_rss.xml
@@ -16,6 +16,7 @@
<title>{{ r.title }}</title>
<link>{{ r.url }}</link>
<description>{{ r.content }}</description>
+ {% if r.pubdate %}<pubDate>{{ r.pubdate }}</pubDate>{% endif %}
</item>
{% endfor %}
</channel>
diff --git a/searx/templates/result_templates/videos.html b/searx/templates/result_templates/videos.html
@@ -5,6 +5,7 @@
<p>
<h3 class="result_title"><a href="{{ result.url }}">{{ result.title|safe }}</a></h3>
+ {% if result.publishedDate %}<p class="published_date">{{ result.publishedDate }}</p>{% endif %}
<a href="{{ result.url }}"><img width="400px" src="{{ result.thumbnail }}" title={{ result.title }} alt=" {{ result.title }}"/></a>
<p class="url">{{ result.url }}</p>
</p>
diff --git a/searx/webapp.py b/searx/webapp.py
@@ -159,8 +159,8 @@ def index():
# TODO, check if timezone is calculated right
if 'publishedDate' in result:
- if result['publishedDate'] >= datetime.now() - timedelta(days=1):
- timedifference = datetime.now() - result['publishedDate']
+ if result['publishedDate'].replace(tzinfo=None) >= datetime.now() - timedelta(days=1):
+ timedifference = datetime.now() - result['publishedDate'].replace(tzinfo=None)
minutes = int((timedifference.seconds / 60) % 60)
hours = int(timedifference.seconds / 60 / 60)
if hours == 0:
@@ -168,6 +168,7 @@ def index():
else:
result['publishedDate'] = gettext(u'{hours} hour(s), {minutes} minute(s) ago').format(hours=hours, minutes=minutes) # noqa
else:
+ result['pubdate'] = result['publishedDate'].strftime('%a, %d %b %Y %H:%M:%S %z')
result['publishedDate'] = format_date(result['publishedDate'])
if search.request_data.get('format') == 'json':
diff --git a/setup.py b/setup.py
@@ -35,6 +35,7 @@ setup(
'lxml',
'pyyaml',
'setuptools',
+ 'python-dateutil',
],
extras_require={
'test': [
