commit: 0a71525ab6d4fe4cbc4b33b4653bdb39ae4d55e9
parent 6bfd5663539052a64c984f5bdb7135d0d652c923
Author: Dalf <alex@al-f.net>
Date: Sun, 28 Sep 2014 16:53:30 +0200
[enh] add infoboxes and answers (clean up)
Diffstat:
4 files changed, 74 insertions(+), 20 deletions(-)
diff --git a/searx/engines/wikidata.py b/searx/engines/wikidata.py
@@ -1,13 +1,12 @@
import json
-from datetime import datetime
from requests import get
from urllib import urlencode
+from datetime import datetime
resultCount=2
-urlSearch = 'https://www.wikidata.org/w/api.php?action=query&list=search&format=json&srnamespace=0&srprop=sectionsnippet&{query}'
+urlSearch = 'https://www.wikidata.org/w/api.php?action=query&list=search&format=json&srnamespace=0&srprop=sectiontitle&{query}'
urlDetail = 'https://www.wikidata.org/w/api.php?action=wbgetentities&format=json&props=labels%7Cinfo%7Csitelinks%7Csitelinks%2Furls%7Cdescriptions%7Cclaims&{query}'
-# find the right URL for urlMap
-urlMap = 'http://www.openstreetmap.org/?lat={latitude}&lon={longitude}&zoom={zoom}&layers=M'
+urlMap = 'https://www.openstreetmap.org/?lat={latitude}&lon={longitude}&zoom={zoom}&layers=M'
def request(query, params):
params['url'] = urlSearch.format(query=urlencode({'srsearch': query, 'srlimit': resultCount}))
@@ -18,24 +17,27 @@ def request(query, params):
def response(resp):
results = []
search_res = json.loads(resp.text)
- # TODO parallel http queries
- before = datetime.now()
+
+ wikidata_ids = set()
for r in search_res.get('query', {}).get('search', {}):
- wikidata_id = r.get('title', '')
- results = results + getDetail(wikidata_id)
- after = datetime.now()
- print str(after - before) + " second(s)"
+ wikidata_ids.add(r.get('title', ''))
- return results
+ language = resp.search_params['language'].split('_')[0]
+ if language == 'all':
+ language = 'en'
+ url = urlDetail.format(query=urlencode({'ids': '|'.join(wikidata_ids), 'languages': language + '|en'}))
+
+ before = datetime.now()
+ htmlresponse = get(url)
+ print datetime.now() - before
+ jsonresponse = json.loads(htmlresponse.content)
+ for wikidata_id in wikidata_ids:
+ results = results + getDetail(jsonresponse, wikidata_id, language)
-def getDetail(wikidata_id):
- language = 'fr'
+ return results
- url = urlDetail.format(query=urlencode({'ids': wikidata_id, 'languages': language + '|en'}))
- print url
- response = get(url)
- result = json.loads(response.content)
- result = result.get('entities', {}).get(wikidata_id, {})
+def getDetail(jsonresponse, wikidata_id, language):
+ result = jsonresponse.get('entities', {}).get(wikidata_id, {})
title = result.get('labels', {}).get(language, {}).get('value', None)
if title == None:
@@ -50,7 +52,6 @@ def getDetail(wikidata_id):
claims = result.get('claims', {})
official_website = get_string(claims, 'P856', None)
- print official_website
if official_website != None:
urls.append({ 'title' : 'Official site', 'url': official_website })
results.append({ 'title': title, 'url' : official_website })
@@ -98,10 +99,12 @@ def getDetail(wikidata_id):
return results
+
def add_url(urls, title, url):
if url != None:
urls.append({'title' : title, 'url' : url})
+
def get_mainsnak(claims, propertyName):
propValue = claims.get(propertyName, {})
if len(propValue) == 0:
@@ -110,6 +113,7 @@ def get_mainsnak(claims, propertyName):
propValue = propValue[0].get('mainsnak', None)
return propValue
+
def get_string(claims, propertyName, defaultValue=None):
propValue = claims.get(propertyName, {})
if len(propValue) == 0:
@@ -129,6 +133,7 @@ def get_string(claims, propertyName, defaultValue=None):
else:
return ', '.join(result)
+
def get_time(claims, propertyName, defaultValue=None):
propValue = claims.get(propertyName, {})
if len(propValue) == 0:
@@ -149,6 +154,7 @@ def get_time(claims, propertyName, defaultValue=None):
else:
return ', '.join(result)
+
def get_geolink(claims, propertyName, defaultValue=''):
mainsnak = get_mainsnak(claims, propertyName)
@@ -182,6 +188,7 @@ def get_geolink(claims, propertyName, defaultValue=''):
return url
+
def get_wikilink(result, wikiid):
url = result.get('sitelinks', {}).get(wikiid, {}).get('url', None)
if url == None:
diff --git a/searx/search.py b/searx/search.py
@@ -76,7 +76,6 @@ def make_callback(engine_name, results, suggestions, answers, infoboxes, callbac
# if it is an infobox, add it to list of infoboxes
if 'infobox' in result:
infoboxes.append(result)
- print result
continue
# append result
diff --git a/searx/settings.yml b/searx/settings.yml
@@ -44,6 +44,10 @@ engines:
engine : duckduckgo_definitions
shortcut : ddd
+ - name : wikidata
+ engine : wikidata
+ shortcut : wd
+
- name : duckduckgo
engine : duckduckgo
shortcut : ddg
diff --git a/searx/templates/default/infobox.html b/searx/templates/default/infobox.html
@@ -0,0 +1,44 @@
+<div class="infobox">
+ <h2>{{ infobox.infobox }}</h2>
+ {% if infobox.img_src %}<img src="{{ infobox.img_src }}" />{% endif %}
+ <p>{{ infobox.entity }}</p>
+ <p>{{ infobox.content }}</p>
+ {% if infobox.attributes %}
+ <div class="attributes">
+ <table>
+ {% for attribute in infobox.attributes %}
+ <tr><td>{{ attribute.label }}</td><td>{{ attribute.value }}</td></tr>
+ {% endfor %}
+ </table>
+ </div>
+ {% endif %}
+
+ {% if infobox.urls %}
+ <div class="urls">
+ <ul>
+ {% for url in infobox.urls %}
+ <li class="url"><a href="{{ url.url }}">{{ url.title }}</a></li>
+ {% endfor %}
+ </ul>
+ </div>
+ {% endif %}
+
+ {% if infobox.relatedTopics %}
+ <div class="relatedTopics">
+ {% for topic in infobox.relatedTopics %}
+ <div>
+ <h3>{{ topic.name }}</h3>
+ {% for suggestion in topic.suggestions %}
+ <form method="{{ method or 'POST' }}" action="{{ url_for('index') }}">
+ <input type="hidden" name="q" value="{{ suggestion }}">
+ <input type="submit" value="{{ suggestion }}" />
+ </form>
+ {% endfor %}
+ </div>
+ {% endfor %}
+ </div>
+ {% endif %}
+
+ <br />
+
+</div>
