diff --git a/searx/engines/google_images.py b/searx/engines/google_images.py index 66a543e57..d9a49e9cc 100644 --- a/searx/engines/google_images.py +++ b/searx/engines/google_images.py @@ -11,9 +11,9 @@ """ from datetime import date, timedelta +from json import loads from lxml import html -from searx.url_utils import urlencode, urlparse, parse_qs - +from searx.url_utils import urlencode # engine dependent config categories = ['images'] @@ -25,8 +25,7 @@ number_of_results = 100 search_url = 'https://www.google.com/search'\ '?{query}'\ '&tbm=isch'\ - '&gbv=1'\ - '&sa=G'\ + '&yv=2'\ '&{search_options}' time_range_attr = "qdr:{range}" time_range_custom_attr = "cdr:1,cd_min:{start},cd_max{end}" @@ -38,6 +37,7 @@ time_range_dict = {'day': 'd', # do search-request def request(query, params): search_options = { + 'ijn': params['pageno'] - 1, 'start': (params['pageno'] - 1) * number_of_results } @@ -51,7 +51,7 @@ def request(query, params): search_options['tbs'] = time_range_custom_attr.format(start=start, end=end) if safesearch and params['safesearch']: - search_options['safe'] = 'active' + search_options['safe'] = 'on' params['url'] = search_url.format(query=urlencode({'q': query}), search_options=urlencode(search_options)) @@ -61,30 +61,27 @@ def request(query, params): # get response from search-request def response(resp): + results = [] + dom = html.fromstring(resp.text) - results = [] - for element in dom.xpath('//div[@id="search"] //td'): - link = element.xpath('./a')[0] + # parse results + for result in dom.xpath('//div[contains(@class, "rg_meta")]/text()'): - google_url = urlparse(link.xpath('.//@href')[0]) - query = parse_qs(google_url.query) - source_url = next(iter(query.get('q', [])), None) + try: + metadata = loads(result) + img_format = "{0} {1}x{2}".format(metadata['ity'], str(metadata['ow']), str(metadata['oh'])) + source = "{0} ({1})".format(metadata['st'], metadata['isu']) + results.append({'url': metadata['ru'], + 'title': metadata['pt'], + 'content': metadata['s'], + 'source': source, + 'img_format': img_format, + 'thumbnail_src': metadata['tu'], + 'img_src': metadata['ou'], + 'template': 'images.html'}) - title_parts = element.xpath('./cite//following-sibling::*/text()') - title_parts.extend(element.xpath('./cite//following-sibling::text()')[:-1]) - - result = { - 'title': ''.join(title_parts), - 'content': '', - 'template': 'images.html', - 'url': source_url, - 'img_src': source_url, - 'thumbnail_src': next(iter(link.xpath('.//img //@src')), None) - } - - if not source_url or not result['thumbnail_src']: + except: continue - results.append(result) return results diff --git a/searx/engines/seedpeer.py b/searx/engines/seedpeer.py deleted file mode 100644 index 3770dacac..000000000 --- a/searx/engines/seedpeer.py +++ /dev/null @@ -1,75 +0,0 @@ -# Seedpeer (Videos, Music, Files) -# -# @website http://seedpeer.eu -# @provide-api no (nothing found) -# -# @using-api no -# @results HTML (using search portal) -# @stable yes (HTML can change) -# @parse url, title, content, seed, leech, magnetlink - -from lxml import html -from operator import itemgetter -from searx.url_utils import quote, urljoin - - -url = 'http://www.seedpeer.eu/' -search_url = url + 'search/{search_term}/7/{page_no}.html' -# specific xpath variables -torrent_xpath = '//*[@id="body"]/center/center/table[2]/tr/td/a' -alternative_torrent_xpath = '//*[@id="body"]/center/center/table[1]/tr/td/a' -title_xpath = '//*[@id="body"]/center/center/table[2]/tr/td/a/text()' -alternative_title_xpath = '//*[@id="body"]/center/center/table/tr/td/a' -seeds_xpath = '//*[@id="body"]/center/center/table[2]/tr/td[4]/font/text()' -alternative_seeds_xpath = '//*[@id="body"]/center/center/table/tr/td[4]/font/text()' -peers_xpath = '//*[@id="body"]/center/center/table[2]/tr/td[5]/font/text()' -alternative_peers_xpath = '//*[@id="body"]/center/center/table/tr/td[5]/font/text()' -age_xpath = '//*[@id="body"]/center/center/table[2]/tr/td[2]/text()' -alternative_age_xpath = '//*[@id="body"]/center/center/table/tr/td[2]/text()' -size_xpath = '//*[@id="body"]/center/center/table[2]/tr/td[3]/text()' -alternative_size_xpath = '//*[@id="body"]/center/center/table/tr/td[3]/text()' - - -# do search-request -def request(query, params): - params['url'] = search_url.format(search_term=quote(query), - page_no=params['pageno'] - 1) - return params - - -# get response from search-request -def response(resp): - results = [] - dom = html.fromstring(resp.text) - torrent_links = dom.xpath(torrent_xpath) - if len(torrent_links) > 0: - seeds = dom.xpath(seeds_xpath) - peers = dom.xpath(peers_xpath) - titles = dom.xpath(title_xpath) - sizes = dom.xpath(size_xpath) - ages = dom.xpath(age_xpath) - else: # under ~5 results uses a different xpath - torrent_links = dom.xpath(alternative_torrent_xpath) - seeds = dom.xpath(alternative_seeds_xpath) - peers = dom.xpath(alternative_peers_xpath) - titles = dom.xpath(alternative_title_xpath) - sizes = dom.xpath(alternative_size_xpath) - ages = dom.xpath(alternative_age_xpath) - # return empty array if nothing is found - if not torrent_links: - return [] - - # parse results - for index, result in enumerate(torrent_links): - link = result.attrib.get('href') - href = urljoin(url, link) - results.append({'url': href, - 'title': titles[index].text_content(), - 'content': '{}, {}'.format(sizes[index], ages[index]), - 'seed': seeds[index], - 'leech': peers[index], - - 'template': 'torrent.html'}) - - # return results sorted by seeder - return sorted(results, key=itemgetter('seed'), reverse=True) diff --git a/searx/engines/subtitleseeker.py b/searx/engines/subtitleseeker.py deleted file mode 100644 index 2cbc991b3..000000000 --- a/searx/engines/subtitleseeker.py +++ /dev/null @@ -1,86 +0,0 @@ -""" - Subtitleseeker (Video) - - @website http://www.subtitleseeker.com - @provide-api no - - @using-api no - @results HTML - @stable no (HTML can change) - @parse url, title, content -""" - -from lxml import html -from searx.languages import language_codes -from searx.engines.xpath import extract_text -from searx.url_utils import quote_plus - -# engine dependent config -categories = ['videos'] -paging = True -language = "" - -# search-url -url = 'http://www.subtitleseeker.com/' -search_url = url + 'search/TITLES/{query}?p={pageno}' - -# specific xpath variables -results_xpath = '//div[@class="boxRows"]' - - -# do search-request -def request(query, params): - params['url'] = search_url.format(query=quote_plus(query), - pageno=params['pageno']) - return params - - -# get response from search-request -def response(resp): - results = [] - - dom = html.fromstring(resp.text) - - search_lang = "" - - # dirty fix for languages named differenly in their site - if resp.search_params['language'][:2] == 'fa': - search_lang = 'Farsi' - elif resp.search_params['language'] == 'pt-BR': - search_lang = 'Brazilian' - elif resp.search_params['language'] != 'all': - search_lang = [lc[3] - for lc in language_codes - if lc[0].split('-')[0] == resp.search_params['language'].split('-')[0]] - search_lang = search_lang[0].split(' (')[0] - - # parse results - for result in dom.xpath(results_xpath): - link = result.xpath(".//a")[0] - href = link.attrib.get('href') - - if language is not "": - href = href + language + '/' - elif search_lang: - href = href + search_lang + '/' - - title = extract_text(link) - - content = extract_text(result.xpath('.//div[contains(@class,"red")]')) - content = content + " - " - text = extract_text(result.xpath('.//div[contains(@class,"grey-web")]')[0]) - content = content + text - - if result.xpath(".//span") != []: - content = content +\ - " - (" +\ - extract_text(result.xpath(".//span")) +\ - ")" - - # append result - results.append({'url': href, - 'title': title, - 'content': content}) - - # return results - return results diff --git a/searx/engines/swisscows.py b/searx/engines/swisscows.py deleted file mode 100644 index e451bcffc..000000000 --- a/searx/engines/swisscows.py +++ /dev/null @@ -1,125 +0,0 @@ -""" - Swisscows (Web, Images) - - @website https://swisscows.ch - @provide-api no - - @using-api no - @results HTML (using search portal) - @stable no (HTML can change) - @parse url, title, content -""" - -from json import loads -import re -from lxml.html import fromstring -from searx.url_utils import unquote, urlencode -from searx.utils import match_language - -# engine dependent config -categories = ['general', 'images'] -paging = True -language_support = True - -# search-url -base_url = 'https://swisscows.ch/' -search_string = '?{query}&page={page}' - -supported_languages_url = base_url - -# regex -regex_json = re.compile(b'initialData: {"Request":(.|\n)*},\s*environment') -regex_json_remove_start = re.compile(b'^initialData:\s*') -regex_json_remove_end = re.compile(b',\s*environment$') -regex_img_url_remove_start = re.compile(b'^https?://i\.swisscows\.ch/\?link=') - - -# do search-request -def request(query, params): - if params['language'] == 'all': - ui_language = 'browser' - region = 'browser' - else: - region = match_language(params['language'], supported_languages, language_aliases) - ui_language = region.split('-')[0] - - search_path = search_string.format( - query=urlencode({'query': query, 'uiLanguage': ui_language, 'region': region}), - page=params['pageno'] - ) - - # image search query is something like 'image?{query}&page={page}' - if params['category'] == 'images': - search_path = 'image' + search_path - - params['url'] = base_url + search_path - - return params - - -# get response from search-request -def response(resp): - results = [] - - json_regex = regex_json.search(resp.text) - - # check if results are returned - if not json_regex: - return [] - - json_raw = regex_json_remove_end.sub(b'', regex_json_remove_start.sub(b'', json_regex.group())) - json = loads(json_raw.decode('utf-8')) - - # parse results - for result in json['Results'].get('items', []): - result_title = result['Title'].replace(u'\uE000', '').replace(u'\uE001', '') - - # parse image results - if result.get('ContentType', '').startswith('image'): - img_url = unquote(regex_img_url_remove_start.sub(b'', result['Url'].encode('utf-8')).decode('utf-8')) - - # append result - results.append({'url': result['SourceUrl'], - 'title': result['Title'], - 'content': '', - 'img_src': img_url, - 'template': 'images.html'}) - - # parse general results - else: - result_url = result['Url'].replace(u'\uE000', '').replace(u'\uE001', '') - result_content = result['Description'].replace(u'\uE000', '').replace(u'\uE001', '') - - # append result - results.append({'url': result_url, - 'title': result_title, - 'content': result_content}) - - # parse images - for result in json.get('Images', []): - # decode image url - img_url = unquote(regex_img_url_remove_start.sub(b'', result['Url'].encode('utf-8')).decode('utf-8')) - - # append result - results.append({'url': result['SourceUrl'], - 'title': result['Title'], - 'content': '', - 'img_src': img_url, - 'template': 'images.html'}) - - # return results - return results - - -# get supported languages from their site -def _fetch_supported_languages(resp): - supported_languages = [] - dom = fromstring(resp.text) - options = dom.xpath('//div[@id="regions-popup"]//ul/li/a') - for option in options: - code = option.xpath('./@data-search-language')[0] - if code.startswith('nb-'): - code = code.replace('nb', 'no', 1) - supported_languages.append(code) - - return supported_languages diff --git a/searx/settings.yml b/searx/settings.yml index 506d73cb5..0be07b1e5 100644 --- a/searx/settings.yml +++ b/searx/settings.yml @@ -568,24 +568,12 @@ engines: engine : spotify shortcut : stf - - name : subtitleseeker - engine : subtitleseeker - shortcut : ss -# The language is an option. You can put any language written in english -# Examples : English, French, German, Hungarian, Chinese... -# language : English - - name : startpage engine : startpage shortcut : sp timeout : 6.0 disabled : True - - name : swisscows - engine : swisscows - shortcut : sw - disabled : True - - name : tokyotoshokan engine : tokyotoshokan shortcut : tt @@ -664,12 +652,6 @@ engines: timeout: 6.0 categories : science - - name : seedpeer - engine : seedpeer - shortcut: speu - categories: files, music, videos - disabled: True - - name : dictzone engine : dictzone shortcut : dc diff --git a/tests/unit/engines/test_seedpeer.py b/tests/unit/engines/test_seedpeer.py deleted file mode 100644 index 37b2de8e4..000000000 --- a/tests/unit/engines/test_seedpeer.py +++ /dev/null @@ -1,51 +0,0 @@ -import mock -from collections import defaultdict -from searx.engines import seedpeer -from searx.testing import SearxTestCase -from datetime import datetime - - -class TestSeedPeerEngine(SearxTestCase): - - html = '' - with open('./tests/unit/engines/seedpeer_fixture.html') as fixture: - html += fixture.read() - - def test_request(self): - query = 'test_query' - dicto = defaultdict(dict) - dicto['pageno'] = 1 - params = seedpeer.request(query, dicto) - self.assertIn('url', params) - self.assertIn(query, params['url']) - self.assertIn('seedpeer.eu', params['url']) - - def test_response_raises_attr_error_on_empty_response(self): - self.assertRaises(AttributeError, seedpeer.response, None) - self.assertRaises(AttributeError, seedpeer.response, []) - self.assertRaises(AttributeError, seedpeer.response, '') - self.assertRaises(AttributeError, seedpeer.response, '[]') - - def test_response_returns_empty_list(self): - response = mock.Mock(text='') - self.assertEqual(seedpeer.response(response), []) - - def test_response_returns_all_results(self): - response = mock.Mock(text=self.html) - results = seedpeer.response(response) - self.assertTrue(isinstance(results, list)) - self.assertEqual(len(results), 2) - - def test_response_returns_correct_results(self): - response = mock.Mock(text=self.html) - results = seedpeer.response(response) - self.assertEqual( - results[0]['title'], 'Narcos - Season 2 - 720p WEBRiP - x265 HEVC - ShAaNiG ' - ) - self.assertEqual( - results[0]['url'], - 'http://www.seedpeer.eu/details/11685972/Narcos---Season-2---720p-WEBRiP---x265-HEVC---ShAaNiG.html' - ) - self.assertEqual(results[0]['content'], '2.48 GB, 1 day') - self.assertEqual(results[0]['seed'], '861') - self.assertEqual(results[0]['leech'], '332') diff --git a/tests/unit/engines/test_subtitleseeker.py b/tests/unit/engines/test_subtitleseeker.py deleted file mode 100644 index a22ee74b9..000000000 --- a/tests/unit/engines/test_subtitleseeker.py +++ /dev/null @@ -1,174 +0,0 @@ -from collections import defaultdict -import mock -from searx.engines import subtitleseeker -from searx.testing import SearxTestCase - - -class TestSubtitleseekerEngine(SearxTestCase): - - def test_request(self): - query = 'test_query' - dicto = defaultdict(dict) - dicto['pageno'] = 1 - dicto['language'] = 'fr-FR' - params = subtitleseeker.request(query, dicto) - self.assertTrue('url' in params) - self.assertTrue(query in params['url']) - self.assertTrue('subtitleseeker.com' in params['url']) - - def test_response(self): - dicto = defaultdict(dict) - dicto['language'] = 'fr-FR' - response = mock.Mock(search_params=dicto) - - self.assertRaises(AttributeError, subtitleseeker.response, None) - self.assertRaises(AttributeError, subtitleseeker.response, []) - self.assertRaises(AttributeError, subtitleseeker.response, '') - self.assertRaises(AttributeError, subtitleseeker.response, '[]') - - response = mock.Mock(text='', search_params=dicto) - self.assertEqual(subtitleseeker.response(response), []) - - html = """ -
-
- - - This is the Title - -

- - "Alternative Title" - -
-
- 1998 -
-
- - 1039 Subs -
-
- - 1 hours ago -
-
-
- """ - response = mock.Mock(text=html, search_params=dicto) - results = subtitleseeker.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'], 'http://this.is.the.url/French/') - self.assertIn('1998', results[0]['content']) - self.assertIn('1039 Subs', results[0]['content']) - self.assertIn('Alternative Title', results[0]['content']) - - dicto['language'] = 'pt-BR' - results = subtitleseeker.response(response) - self.assertEqual(results[0]['url'], 'http://this.is.the.url/Brazilian/') - - html = """ -
-
- - - This is the Title - -
-
- 1998 -
-
- - 1039 Subs -
-
- - 1 hours ago -
-
-
- """ - dicto['language'] = 'all' - response = mock.Mock(text=html, search_params=dicto) - results = subtitleseeker.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'], 'http://this.is.the.url/') - self.assertIn('1998', results[0]['content']) - self.assertIn('1039 Subs', results[0]['content']) - - html = """ -
-
- - - This is the Title - -
-
- 1998 -
-
- - 1039 Subs -
-
- - 1 hours ago -
-
-
- """ - subtitleseeker.language = 'English' - response = mock.Mock(text=html, search_params=dicto) - results = subtitleseeker.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'], 'http://this.is.the.url/English/') - self.assertIn('1998', results[0]['content']) - self.assertIn('1039 Subs', results[0]['content']) - - html = """ -
- - - This is the Title - -
-
- 1998 -
-
- - 1039 Subs -
-
- - 1 hours ago -
- """ - response = mock.Mock(text=html, search_params=dicto) - results = subtitleseeker.response(response) - self.assertEqual(type(results), list) - self.assertEqual(len(results), 0) diff --git a/tests/unit/engines/test_swisscows.py b/tests/unit/engines/test_swisscows.py deleted file mode 100644 index fbfcc94cc..000000000 --- a/tests/unit/engines/test_swisscows.py +++ /dev/null @@ -1,157 +0,0 @@ -from collections import defaultdict -import mock -from searx.engines import swisscows -from searx.testing import SearxTestCase - - -class TestSwisscowsEngine(SearxTestCase): - - def test_request(self): - swisscows.supported_languages = ['de-AT', 'de-DE'] - swisscows.language_aliases = {} - query = 'test_query' - dicto = defaultdict(dict) - dicto['pageno'] = 1 - dicto['language'] = 'de-DE' - params = swisscows.request(query, dicto) - self.assertTrue('url' in params) - self.assertTrue(query in params['url']) - self.assertTrue('swisscows.ch' in params['url']) - self.assertTrue('uiLanguage=de' in params['url']) - self.assertTrue('region=de-DE' in params['url']) - - dicto['language'] = 'all' - params = swisscows.request(query, dicto) - self.assertTrue('uiLanguage=browser' in params['url']) - self.assertTrue('region=browser' in params['url']) - - dicto['category'] = 'images' - params = swisscows.request(query, dicto) - self.assertIn('image', params['url']) - - def test_response(self): - self.assertRaises(AttributeError, swisscows.response, None) - self.assertRaises(AttributeError, swisscows.response, []) - self.assertRaises(AttributeError, swisscows.response, '') - self.assertRaises(AttributeError, swisscows.response, '[]') - - response = mock.Mock(text=b'') - self.assertEqual(swisscows.response(response), []) - - response = mock.Mock(text=b'') - self.assertEqual(swisscows.response(response), []) - - html = b""" - - """ - response = mock.Mock(text=html) - results = swisscows.response(response) - self.assertEqual(type(results), list) - self.assertEqual(len(results), 3) - self.assertEqual(results[0]['title'], 'This should be the title') - self.assertEqual(results[0]['url'], 'http://this.should.be.the.link/') - self.assertEqual(results[0]['content'], 'This should be the content.') - self.assertEqual(results[1]['title'], 'Datei:This should1.svg') - self.assertEqual(results[1]['url'], 'http://de.wikipedia.org/wiki/Datei:This should1.svg') - self.assertEqual(results[1]['img_src'], 'http://ts2.mm.This/should1.png') - self.assertEqual(results[1]['template'], 'images.html') - self.assertEqual(results[2]['title'], 'Datei:This should.svg') - self.assertEqual(results[2]['url'], 'http://de.wikipedia.org/wiki/Datei:This should.svg') - self.assertEqual(results[2]['img_src'], 'http://ts2.mm.This/should.png') - self.assertEqual(results[2]['template'], 'images.html') - - def test_fetch_supported_languages(self): - html = """""" - response = mock.Mock(text=html) - languages = swisscows._fetch_supported_languages(response) - self.assertEqual(type(languages), list) - self.assertEqual(len(languages), 0) - - html = """ - -
-
- -
-
- - """ - response = mock.Mock(text=html) - languages = swisscows._fetch_supported_languages(response) - self.assertEqual(type(languages), list) - self.assertEqual(len(languages), 3) - self.assertIn('de-CH', languages) - self.assertIn('fr-CH', languages)