From f6e9c074bbe8b4237ee361befa8dcb2c6d31a11a Mon Sep 17 00:00:00 2001 From: Adam Tauber Date: Sun, 11 Dec 2016 02:33:04 +0100 Subject: [PATCH] [fix] vimeo engine change follow-up --- searx/engines/vimeo.py | 32 +++++-------- tests/unit/engines/test_vimeo.py | 79 +++----------------------------- 2 files changed, 19 insertions(+), 92 deletions(-) diff --git a/searx/engines/vimeo.py b/searx/engines/vimeo.py index 517ac1c4..5d531054 100644 --- a/searx/engines/vimeo.py +++ b/searx/engines/vimeo.py @@ -12,10 +12,8 @@ # @todo rewrite to api # @todo set content-parameter with correct data +from json import loads from urllib import urlencode -from lxml import html -from HTMLParser import HTMLParser -from searx.engines.xpath import extract_text from dateutil import parser # engine dependent config @@ -23,17 +21,10 @@ categories = ['videos'] paging = True # search-url -base_url = 'https://vimeo.com' +base_url = 'https://vimeo.com/' search_url = base_url + '/search/page:{pageno}?{query}' -# specific xpath variables -results_xpath = '//div[contains(@class,"results_grid")]/ul/li' -url_xpath = './/a/@href' -title_xpath = './/span[@class="title"]' -thumbnail_xpath = './/img[@class="js-clip_thumbnail_image"]/@src' -publishedDate_xpath = './/time/attribute::datetime' - -embedded_url = '' @@ -49,17 +40,18 @@ def request(query, params): # get response from search-request def response(resp): results = [] - - dom = html.fromstring(resp.text) - p = HTMLParser() + data_start_pos = resp.text.find('{"filtered"') + data_end_pos = resp.text.find(';\n', data_start_pos + 1) + data = loads(resp.text[data_start_pos:data_end_pos]) # parse results - for result in dom.xpath(results_xpath): - videoid = result.xpath(url_xpath)[0] + for result in data['filtered']['data']: + result = result[result['type']] + videoid = result['uri'].split('/')[-1] url = base_url + videoid - title = p.unescape(extract_text(result.xpath(title_xpath))) - thumbnail = extract_text(result.xpath(thumbnail_xpath)[0]) - publishedDate = parser.parse(extract_text(result.xpath(publishedDate_xpath)[0])) + title = result['name'] + thumbnail = result['pictures']['sizes'][-1]['link'] + publishedDate = parser.parse(result['created_time']) embedded = embedded_url.format(videoid=videoid) # append result diff --git a/tests/unit/engines/test_vimeo.py b/tests/unit/engines/test_vimeo.py index 50b1cb56..c86b50a1 100644 --- a/tests/unit/engines/test_vimeo.py +++ b/tests/unit/engines/test_vimeo.py @@ -22,80 +22,15 @@ class TestVimeoEngine(SearxTestCase): self.assertRaises(AttributeError, vimeo.response, '') self.assertRaises(AttributeError, vimeo.response, '[]') - response = mock.Mock(text='') - self.assertEqual(vimeo.response(response), []) + json = u""" +{"filtered":{"total":274641,"page":1,"per_page":18,"paging":{"next":"?sizes=590x332&page=2","previous":null,"first":"?sizes=590x332&page=1","last":"?sizes=590x332&page=15258"},"data":[{"is_staffpick":false,"is_featured":true,"type":"clip","clip":{"uri":"\\/videos\\/106557563","name":"Hot Rod Revue: The South","link":"https:\\/\\/vimeo.com\\/106557563","duration":4069,"created_time":"2014-09-19T03:38:04+00:00","privacy":{"view":"ptv"},"pictures":{"sizes":[{"width":"590","height":"332","link":"https:\\/\\/i.vimeocdn.com\\/video\\/489717884_590x332.jpg?r=pad","link_with_play_button":"https:\\/\\/i.vimeocdn.com\\/filter\\/overlay?src0=https%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F489717884_590x332.jpg&src1=http%3A%2F%2Ff.vimeocdn.com%2Fp%2Fimages%2Fcrawler_play.png"}]},"stats":{"plays":null},"metadata":{"connections":{"comments":{"total":0},"likes":{"total":5}},"interactions":[]},"user":{"name":"Cal Thorley","link":"https:\\/\\/vimeo.com\\/calthorley","pictures":{"sizes":[{"width":30,"height":30,"link":"https:\\/\\/i.vimeocdn.com\\/portrait\\/2545308_30x30?r=pad"},{"width":75,"height":75,"link":"https:\\/\\/i.vimeocdn.com\\/portrait\\/2545308_75x75?r=pad"},{"width":100,"height":100,"link":"https:\\/\\/i.vimeocdn.com\\/portrait\\/2545308_100x100?r=pad"},{"width":300,"height":300,"link":"https:\\/\\/i.vimeocdn.com\\/portrait\\/2545308_300x300?r=pad"}]}}}}]}}; - html = """ -
- -
- """ - response = mock.Mock(text=html) +""" # noqa + response = mock.Mock(text=json) results = vimeo.response(response) self.assertEqual(type(results), list) self.assertEqual(len(results), 1) - self.assertEqual(results[0]['title'], 'This is the title') - self.assertEqual(results[0]['url'], 'https://vimeo.com/videoid') + self.assertEqual(results[0]['title'], u'Hot Rod Revue: The South') + self.assertEqual(results[0]['url'], 'https://vimeo.com/106557563') self.assertEqual(results[0]['content'], '') - self.assertEqual(results[0]['thumbnail'], 'http://image.url.webp') - self.assertIn('/videoid', results[0]['embedded']) - - html = """ -
    -
  1. - - -
    -

    - This is the title -

    -

    - -

    -
    -     -
    2. -
- """ - response = mock.Mock(text=html) - results = vimeo.response(response) - self.assertEqual(type(results), list) - self.assertEqual(len(results), 0) + self.assertEqual(results[0]['thumbnail'], 'https://i.vimeocdn.com/video/489717884_590x332.jpg?r=pad')