From f965c978222cf48e8dd4b7dd6c9a28ccca9bc62f Mon Sep 17 00:00:00 2001 From: Cqoicebordel Date: Sun, 31 May 2015 00:25:59 +0200 Subject: [PATCH] Adds two engines : Youtube with or without API The API needs an API_KEY The NOAPI doesn't have the published dates. --- searx/engines/youtube_api.py | 83 ++++++++++++++++ searx/engines/youtube_noapi.py | 72 ++++++++++++++ searx/settings.yml | 7 +- searx/tests/engines/test_youtube_api.py | 111 ++++++++++++++++++++++ searx/tests/engines/test_youtube_noapi.py | 103 ++++++++++++++++++++ searx/tests/test_engines.py | 2 + 6 files changed, 377 insertions(+), 1 deletion(-) create mode 100644 searx/engines/youtube_api.py create mode 100644 searx/engines/youtube_noapi.py create mode 100644 searx/tests/engines/test_youtube_api.py create mode 100644 searx/tests/engines/test_youtube_noapi.py diff --git a/searx/engines/youtube_api.py b/searx/engines/youtube_api.py new file mode 100644 index 00000000..8fd939a2 --- /dev/null +++ b/searx/engines/youtube_api.py @@ -0,0 +1,83 @@ +# Youtube (Videos) +# +# @website https://www.youtube.com/ +# @provide-api yes (https://developers.google.com/apis-explorer/#p/youtube/v3/youtube.search.list) +# +# @using-api yes +# @results JSON +# @stable yes +# @parse url, title, content, publishedDate, thumbnail, embedded + +from json import loads +from urllib import urlencode +from dateutil import parser + +# engine dependent config +categories = ['videos', 'music'] +paging = False +language_support = True +api_key = None + +# search-url +base_url = 'https://www.googleapis.com/youtube/v3/search' +search_url = base_url + '?part=snippet&{query}&maxResults=20&key={api_key}' + +embedded_url = '' + +base_youtube_url = 'https://www.youtube.com/watch?v=' + + +# do search-request +def request(query, params): + params['url'] = search_url.format(query=urlencode({'q': query}), + api_key=api_key) + + # add language tag if specified + if params['language'] != 'all': + params['url'] += '&relevanceLanguage=' + params['language'].split('_')[0] + + return params + + +# get response from search-request +def response(resp): + results = [] + + search_results = loads(resp.text) + + # return empty array if there are no results + if 'items' not in search_results: + return [] + + # parse results + for result in search_results['items']: + videoid = result['id']['videoId'] + + title = result['snippet']['title'] + content = '' + thumbnail = '' + + pubdate = result['snippet']['publishedAt'] + publishedDate = parser.parse(pubdate) + + thumbnail = result['snippet']['thumbnails']['high']['url'] + + content = result['snippet']['description'] + + url = base_youtube_url + videoid + + embedded = embedded_url.format(videoid=videoid) + + # append result + results.append({'url': url, + 'title': title, + 'content': content, + 'template': 'videos.html', + 'publishedDate': publishedDate, + 'embedded': embedded, + 'thumbnail': thumbnail}) + + # return results + return results diff --git a/searx/engines/youtube_noapi.py b/searx/engines/youtube_noapi.py new file mode 100644 index 00000000..f78e43f0 --- /dev/null +++ b/searx/engines/youtube_noapi.py @@ -0,0 +1,72 @@ +# Youtube (Videos) +# +# @website https://www.youtube.com/ +# @provide-api yes (https://developers.google.com/apis-explorer/#p/youtube/v3/youtube.search.list) +# +# @using-api no +# @results HTML +# @stable no +# @parse url, title, content, publishedDate, thumbnail, embedded + +from urllib import quote_plus +from lxml import html +from searx.engines.xpath import extract_text + +# engine dependent config +categories = ['videos', 'music'] +paging = True +language_support = False + +# search-url +base_url = 'https://www.youtube.com/results' +search_url = base_url + '?search_query={query}&page={page}' + +embedded_url = '' + +base_youtube_url = 'https://www.youtube.com/watch?v=' + +# specific xpath variables +results_xpath = "//ol/li/div[contains(@class, 'yt-lockup yt-lockup-tile yt-lockup-video vve-check')]" +url_xpath = './/h3/a/@href' +title_xpath = './/div[@class="yt-lockup-content"]/h3/a' +content_xpath = './/div[@class="yt-lockup-content"]/div[@class="yt-lockup-description yt-ui-ellipsis yt-ui-ellipsis-2"]' + + +# do search-request +def request(query, params): + params['url'] = search_url.format(query=quote_plus(query), + page=params['pageno']) + + return params + + +# get response from search-request +def response(resp): + results = [] + + dom = html.fromstring(resp.text) + + # parse results + for result in dom.xpath(results_xpath): + videoid = result.xpath('@data-context-item-id')[0] + + url = base_youtube_url + videoid + thumbnail = 'https://i.ytimg.com/vi/' + videoid + '/hqdefault.jpg' + + title = extract_text(result.xpath(title_xpath)[0]) + content = extract_text(result.xpath(content_xpath)[0]) + + embedded = embedded_url.format(videoid=videoid) + + # append result + results.append({'url': url, + 'title': title, + 'content': content, + 'template': 'videos.html', + 'embedded': embedded, + 'thumbnail': thumbnail}) + + # return results + return results diff --git a/searx/settings.yml b/searx/settings.yml index d35b1378..519ea8be 100644 --- a/searx/settings.yml +++ b/searx/settings.yml @@ -242,8 +242,13 @@ engines: shortcut : yhn - name : youtube - engine : youtube shortcut : yt + # You can use the engine using the official stable API, but you need an API key + # See : https://console.developers.google.com/project + # engine : youtube_api + # api_key: 'apikey' # required! + # Or you can use the html non-stable engine, activated by default + engine : youtube_noapi - name : dailymotion engine : dailymotion diff --git a/searx/tests/engines/test_youtube_api.py b/searx/tests/engines/test_youtube_api.py new file mode 100644 index 00000000..0d4d478c --- /dev/null +++ b/searx/tests/engines/test_youtube_api.py @@ -0,0 +1,111 @@ +from collections import defaultdict +import mock +from searx.engines import youtube_api +from searx.testing import SearxTestCase + + +class TestYoutubeAPIEngine(SearxTestCase): + + def test_request(self): + query = 'test_query' + dicto = defaultdict(dict) + dicto['pageno'] = 0 + dicto['language'] = 'fr_FR' + params = youtube_api.request(query, dicto) + self.assertTrue('url' in params) + self.assertTrue(query in params['url']) + self.assertIn('googleapis.com', params['url']) + self.assertIn('youtube', params['url']) + self.assertIn('fr', params['url']) + + dicto['language'] = 'all' + params = youtube_api.request(query, dicto) + self.assertFalse('fr' in params['url']) + + def test_response(self): + self.assertRaises(AttributeError, youtube_api.response, None) + self.assertRaises(AttributeError, youtube_api.response, []) + self.assertRaises(AttributeError, youtube_api.response, '') + self.assertRaises(AttributeError, youtube_api.response, '[]') + + response = mock.Mock(text='{}') + self.assertEqual(youtube_api.response(response), []) + + response = mock.Mock(text='{"data": []}') + self.assertEqual(youtube_api.response(response), []) + + json = """ + { + "kind": "youtube#searchListResponse", + "etag": "xmg9xJZuZD438sF4hb-VcBBREXc/YJQDcTBCDcaBvl-sRZJoXdvy1ME", + "nextPageToken": "CAUQAA", + "pageInfo": { + "totalResults": 1000000, + "resultsPerPage": 20 + }, + "items": [ + { + "kind": "youtube#searchResult", + "etag": "xmg9xJZuZD438sF4hb-VcBBREXc/IbLO64BMhbHIgWLwLw7MDYe7Hs4", + "id": { + "kind": "youtube#video", + "videoId": "DIVZCPfAOeM" + }, + "snippet": { + "publishedAt": "2015-05-29T22:41:04.000Z", + "channelId": "UCNodmx1ERIjKqvcJLtdzH5Q", + "title": "Title", + "description": "Description", + "thumbnails": { + "default": { + "url": "https://i.ytimg.com/vi/DIVZCPfAOeM/default.jpg" + }, + "medium": { + "url": "https://i.ytimg.com/vi/DIVZCPfAOeM/mqdefault.jpg" + }, + "high": { + "url": "https://i.ytimg.com/vi/DIVZCPfAOeM/hqdefault.jpg" + } + }, + "channelTitle": "MinecraftUniverse", + "liveBroadcastContent": "none" + } + } + ] + } + """ + response = mock.Mock(text=json) + results = youtube_api.response(response) + self.assertEqual(type(results), list) + self.assertEqual(len(results), 1) + self.assertEqual(results[0]['title'], 'Title') + self.assertEqual(results[0]['url'], 'https://www.youtube.com/watch?v=DIVZCPfAOeM') + self.assertEqual(results[0]['content'], 'Description') + self.assertEqual(results[0]['thumbnail'], 'https://i.ytimg.com/vi/DIVZCPfAOeM/hqdefault.jpg') + self.assertTrue('DIVZCPfAOeM' in results[0]['embedded']) + + json = """ + { + "kind": "youtube#searchListResponse", + "etag": "xmg9xJZuZD438sF4hb-VcBBREXc/YJQDcTBCDcaBvl-sRZJoXdvy1ME", + "nextPageToken": "CAUQAA", + "pageInfo": { + "totalResults": 1000000, + "resultsPerPage": 20 + } + } + """ + response = mock.Mock(text=json) + results = youtube_api.response(response) + self.assertEqual(type(results), list) + self.assertEqual(len(results), 0) + + json = """ + {"toto":{"entry":[] + } + } + """ + response = mock.Mock(text=json) + results = youtube_api.response(response) + self.assertEqual(type(results), list) + self.assertEqual(len(results), 0) diff --git a/searx/tests/engines/test_youtube_noapi.py b/searx/tests/engines/test_youtube_noapi.py new file mode 100644 index 00000000..b715ed2f --- /dev/null +++ b/searx/tests/engines/test_youtube_noapi.py @@ -0,0 +1,103 @@ +# -*- coding: utf-8 -*- +from collections import defaultdict +import mock +from searx.engines import youtube_noapi +from searx.testing import SearxTestCase + + +class TestYoutubeNoAPIEngine(SearxTestCase): + + def test_request(self): + query = 'test_query' + dicto = defaultdict(dict) + dicto['pageno'] = 0 + params = youtube_noapi.request(query, dicto) + self.assertIn('url', params) + self.assertIn(query, params['url']) + self.assertIn('youtube.com', params['url']) + + def test_response(self): + self.assertRaises(AttributeError, youtube_noapi.response, None) + self.assertRaises(AttributeError, youtube_noapi.response, []) + self.assertRaises(AttributeError, youtube_noapi.response, '') + self.assertRaises(AttributeError, youtube_noapi.response, '[]') + + response = mock.Mock(text='') + self.assertEqual(youtube_noapi.response(response), []) + + html = """ +
    +
  1. +
    +
    + + + +
    +
    +

    + + Title + + - Durée : 11:35. +

    +
    de + MEHAY SUPERBIKE
    +
      +
    • il y a 20 heures
      • +
    • 8 424 vues
      • +
    +
    +
    + Description +
    +
    +
      +
    • + Nouveauté +
      • +
    • HD
      • +
    +
    +
    +
    +
    +
    +
    +
    +
    +
    2. +
+ """ + response = mock.Mock(text=html) + results = youtube_noapi.response(response) + self.assertEqual(type(results), list) + self.assertEqual(len(results), 1) + self.assertEqual(results[0]['title'], 'Title') + self.assertEqual(results[0]['url'], 'https://www.youtube.com/watch?v=DIVZCPfAOeM') + self.assertEqual(results[0]['content'], 'Description') + self.assertEqual(results[0]['thumbnail'], 'https://i.ytimg.com/vi/DIVZCPfAOeM/hqdefault.jpg') + self.assertTrue('DIVZCPfAOeM' in results[0]['embedded']) + + html = """ +
    +
  1. +
    2. +
+ """ + response = mock.Mock(text=html) + results = youtube_noapi.response(response) + self.assertEqual(type(results), list) + self.assertEqual(len(results), 0) diff --git a/searx/tests/test_engines.py b/searx/tests/test_engines.py index 5770458f..d0a4de4b 100644 --- a/searx/tests/test_engines.py +++ b/searx/tests/test_engines.py @@ -39,4 +39,6 @@ from searx.tests.engines.test_www500px import * # noqa from searx.tests.engines.test_yacy import * # noqa from searx.tests.engines.test_yahoo import * # noqa from searx.tests.engines.test_youtube import * # noqa +from searx.tests.engines.test_youtube_api import * # noqa +from searx.tests.engines.test_youtube_noapi import * # noqa from searx.tests.engines.test_yahoo_news import * # noqa