2006eb4 · searx · git.iko.soy

searx/engines/1337x.py+1−2

compare with master

unifiedsplit

@@ -1,7 +1,6 @@

1from urllib.parse import quote, urljoin

2from lxml import html

3−from searx.engines.xpath import extract_text

3+from searx.utils import extract_text, get_torrent_size

4−from searx.utils import get_torrent_size

5

4

6

5

7url = 'https://1337x.to/'

6url = 'https://1337x.to/'

searx/engines/acgsou.py+1−2

@@ -11,8 +11,7 @@

11

12from urllib.parse import urlencode

13from lxml import html

14−from searx.engines.xpath import extract_text

14+from searx.utils import extract_text, get_torrent_size, int_or_zero

15−from searx.utils import get_torrent_size, int_or_zero

16

15

17# engine dependent config

16# engine dependent config

18categories = ['files', 'images', 'videos', 'music']

17categories = ['files', 'images', 'videos', 'music']

searx/engines/apkmirror.py+1−1

@@ -11,7 +11,7 @@

11

12from urllib.parse import urlencode

13from lxml import html

14−from searx.engines.xpath import extract_text

14+from searx.utils import extract_text

15

16

17# engine dependent config

searx/engines/archlinux.py+1−1

@@ -13,7 +13,7 @@

13

14from urllib.parse import urlencode, urljoin

15from lxml import html

16−from searx.engines.xpath import extract_text

16+from searx.utils import extract_text

17

18# engine dependent config

19categories = ['it']

searx/engines/bing.py+1−2

@@ -17,8 +17,7 @@ import re

17from urllib.parse import urlencode

18from lxml import html

19from searx import logger, utils

20−from searx.engines.xpath import extract_text

21−from searx.utils import match_language, gen_useragent, eval_xpath

20+from searx.utils import extract_text, match_language, gen_useragent, eval_xpath

22

21

23logger = logger.getChild('bing engine')

22logger = logger.getChild('bing engine')

24

23

searx/engines/btdigg.py+1−2

@@ -13,8 +13,7 @@

13from lxml import html

14from operator import itemgetter

15from urllib.parse import quote, urljoin

16−from searx.engines.xpath import extract_text

16+from searx.utils import extract_text, get_torrent_size

17−from searx.utils import get_torrent_size

18

17

19# engine dependent config

18# engine dependent config

20categories = ['videos', 'music', 'files']

19categories = ['videos', 'music', 'files']

searx/engines/deviantart.py+1−1

@@ -15,7 +15,7 @@

15from lxml import html

16import re

17from urllib.parse import urlencode

18−from searx.engines.xpath import extract_text

18+from searx.utils import extract_text

19

20

21# engine dependent config

searx/engines/digbt.py+1−2

@@ -12,8 +12,7 @@

12

13from urllib.parse import urljoin

14from lxml import html

15−from searx.engines.xpath import extract_text

15+from searx.utils import extract_text, get_torrent_size

16−from searx.utils import get_torrent_size

17

16

18

17

19categories = ['videos', 'music', 'files']

18categories = ['videos', 'music', 'files']

searx/engines/doku.py+1−2

@@ -11,8 +11,7 @@

11

12from urllib.parse import urlencode

13from lxml.html import fromstring

14−from searx.engines.xpath import extract_text

14+from searx.utils import extract_text, eval_xpath

15−from searx.utils import eval_xpath

16

15

17# engine dependent config

16# engine dependent config

18categories = ['general'] # TODO , 'images', 'music', 'videos', 'files'

17categories = ['general'] # TODO , 'images', 'music', 'videos', 'files'

searx/engines/duckduckgo.py+1−2

@@ -16,9 +16,8 @@

16from lxml.html import fromstring

17from json import loads

18from urllib.parse import urlencode

19−from searx.engines.xpath import extract_text

20from searx.poolrequests import get

19from searx.poolrequests import get

21−from searx.utils import match_language, eval_xpath

20+from searx.utils import extract_text, match_language, eval_xpath

22

21

23# engine dependent config

22# engine dependent config

24categories = ['general']

23categories = ['general']

searx/engines/duckduckgo_definitions.py+1−2

@@ -13,9 +13,8 @@ import json

13from urllib.parse import urlencode

14from lxml import html

15from re import compile

16−from searx.engines.xpath import extract_text

17from searx.engines.duckduckgo import _fetch_supported_languages, supported_languages_url, language_aliases

16from searx.engines.duckduckgo import _fetch_supported_languages, supported_languages_url, language_aliases

18−from searx.utils import html_to_text, match_language

17+from searx.utils import extract_text, html_to_text, match_language

19

18

20url = 'https://api.duckduckgo.com/'\

19url = 'https://api.duckduckgo.com/'\

21 + '?{query}&format=json&pretty=0&no_redirect=1&d=1'

20 + '?{query}&format=json&pretty=0&no_redirect=1&d=1'

searx/engines/duckduckgo_images.py+1−1

@@ -15,12 +15,12 @@

15

16from json import loads

17from urllib.parse import urlencode

18−from searx.engines.xpath import extract_text

19from searx.engines.duckduckgo import (

18from searx.engines.duckduckgo import (

20 _fetch_supported_languages, supported_languages_url,

19 _fetch_supported_languages, supported_languages_url,

21 get_region_code, language_aliases

20 get_region_code, language_aliases

22)

21)

23from searx.poolrequests import get

22from searx.poolrequests import get

23+from searx.utils import extract_text

24

25# engine dependent config

26categories = ['images']

searx/engines/duden.py+1−2

@@ -11,8 +11,7 @@

11from lxml import html, etree

12import re

13from urllib.parse import quote, urljoin

14−from searx.engines.xpath import extract_text

14+from searx.utils import extract_text, eval_xpath

15−from searx.utils import eval_xpath

16from searx import logger

15from searx import logger

17

16

18categories = ['general']

17categories = ['general']

searx/engines/etools.py+1−2

@@ -11,8 +11,7 @@

11

12from lxml import html

13from urllib.parse import quote

14−from searx.engines.xpath import extract_text

14+from searx.utils import extract_text, eval_xpath

15−from searx.utils import eval_xpath

16

15

17categories = ['general']

16categories = ['general']

18paging = False

17paging = False

searx/engines/fdroid.py+1−1

@@ -11,7 +11,7 @@

11

12from urllib.parse import urlencode

13from lxml import html

14−from searx.engines.xpath import extract_text

14+from searx.utils import extract_text

15

16# engine dependent config

17categories = ['files']

searx/engines/framalibre.py+1−1

@@ -13,7 +13,7 @@

13from html import escape

14from urllib.parse import urljoin, urlencode

15from lxml import html

16−from searx.engines.xpath import extract_text

16+from searx.utils import extract_text

17

18# engine dependent config

19categories = ['it']

searx/engines/gentoo.py+1−1

@@ -13,7 +13,7 @@

13

14from urllib.parse import urlencode, urljoin

15from lxml import html

16−from searx.engines.xpath import extract_text

16+from searx.utils import extract_text

17

18# engine dependent config

19categories = ['it']

searx/engines/google.py+1−2

@@ -21,9 +21,8 @@ Definitions`_.

21from urllib.parse import urlencode, urlparse

22from lxml import html

23from flask_babel import gettext

24−from searx.engines.xpath import extract_text

25from searx import logger

24from searx import logger

26−from searx.utils import match_language, eval_xpath

25+from searx.utils import match_language, extract_text, eval_xpath

27

26

28logger = logger.getChild('google engine')

27logger = logger.getChild('google engine')

29

28

searx/engines/google_images.py+1−2

@@ -28,8 +28,7 @@ from urllib.parse import urlencode, urlparse, unquote

28from lxml import html

29from flask_babel import gettext

30from searx import logger

31−from searx.utils import eval_xpath

31+from searx.utils import extract_text, eval_xpath

32−from searx.engines.xpath import extract_text

33

32

34# pylint: disable=unused-import

33# pylint: disable=unused-import

35from searx.engines.google import (

34from searx.engines.google import (

searx/engines/google_videos.py+1−1

@@ -14,7 +14,7 @@ from datetime import date, timedelta

14from json import loads

15from urllib.parse import urlencode

16from lxml import html

17−from searx.engines.xpath import extract_text

17+from searx.utils import extract_text

18import re

19

20# engine dependent config

searx/engines/ina.py+1−1

@@ -16,7 +16,7 @@ from urllib.parse import urlencode

16from lxml import html

17from dateutil import parser

18from html.parser import HTMLParser

19−from searx.engines.xpath import extract_text

19+from searx.utils import extract_text

20

21

22# engine dependent config

searx/engines/kickass.py+1−2

@@ -13,8 +13,7 @@

13from lxml import html

14from operator import itemgetter

15from urllib.parse import quote, urljoin

16−from searx.engines.xpath import extract_text

17−from searx.utils import get_torrent_size, convert_str_to_int

16+from searx.utils import extract_text, get_torrent_size, convert_str_to_int

18

17

19# engine dependent config

18# engine dependent config

20categories = ['videos', 'music', 'files']

19categories = ['videos', 'music', 'files']

searx/engines/nyaa.py+1−2

@@ -11,8 +11,7 @@

11

12from lxml import html

13from urllib.parse import urlencode

14−from searx.engines.xpath import extract_text

14+from searx.utils import extract_text, get_torrent_size, int_or_zero

15−from searx.utils import get_torrent_size, int_or_zero

16

15

17# engine dependent config

16# engine dependent config

18categories = ['files', 'images', 'videos', 'music']

17categories = ['files', 'images', 'videos', 'music']

searx/engines/piratebay.py+1−2

@@ -13,8 +13,7 @@ from datetime import datetime

13from operator import itemgetter

14

15from urllib.parse import quote, urljoin

16−from searx.engines.xpath import extract_text

16+from searx.utils import extract_text, get_torrent_size

17−from searx.utils import get_torrent_size

18

17

19# engine dependent config

18# engine dependent config

20categories = ["videos", "music", "files"]

19categories = ["videos", "music", "files"]

searx/engines/seedpeer.py+1−1

@@ -12,7 +12,7 @@ from lxml import html

12from json import loads

13from operator import itemgetter

14from urllib.parse import quote, urljoin

15−from searx.engines.xpath import extract_text

15+from searx.utils import extract_text

16

17

18url = 'https://seedpeer.me/'

searx/engines/stackoverflow.py+1−1

@@ -12,7 +12,7 @@

12

13from urllib.parse import urlencode, urljoin

14from lxml import html

15−from searx.engines.xpath import extract_text

15+from searx.utils import extract_text

16

17# engine dependent config

18categories = ['it']

searx/engines/startpage.py+1−2

@@ -17,9 +17,8 @@ import re

17from unicodedata import normalize, combining

18from babel import Locale

19from babel.localedata import locale_identifiers

20−from searx.engines.xpath import extract_text

21from searx.languages import language_codes

20from searx.languages import language_codes

22−from searx.utils import eval_xpath, match_language

21+from searx.utils import extract_text, eval_xpath, match_language

23

22

24# engine dependent config

23# engine dependent config

25categories = ['general']

24categories = ['general']

searx/engines/tokyotoshokan.py+1−2

@@ -13,9 +13,8 @@

13import re

14from urllib.parse import urlencode

15from lxml import html

16−from searx.engines.xpath import extract_text

17from datetime import datetime

16from datetime import datetime

18−from searx.utils import get_torrent_size, int_or_zero

17+from searx.utils import extract_text, get_torrent_size, int_or_zero

19

18

20# engine dependent config

19# engine dependent config

21categories = ['files', 'videos', 'music']

20categories = ['files', 'videos', 'music']

searx/engines/torrentz.py+1−2

@@ -15,8 +15,7 @@ import re

15from urllib.parse import urlencode

16from lxml import html

17from datetime import datetime

18−from searx.engines.xpath import extract_text

18+from searx.utils import extract_text, get_torrent_size

19−from searx.utils import get_torrent_size

20

19

21# engine dependent config

20# engine dependent config

22categories = ['files', 'videos', 'music']

21categories = ['files', 'videos', 'music']

searx/engines/twitter.py+1−1

@@ -15,7 +15,7 @@

15from urllib.parse import urlencode, urljoin

16from lxml import html

17from datetime import datetime

18−from searx.engines.xpath import extract_text

18+from searx.utils import extract_text

19

20# engine dependent config

21categories = ['social media']

searx/engines/wikidata.py+1−2

@@ -13,9 +13,8 @@

13

14from searx import logger

15from searx.poolrequests import get

16−from searx.engines.xpath import extract_text

17from searx.engines.wikipedia import _fetch_supported_languages, supported_languages_url

16from searx.engines.wikipedia import _fetch_supported_languages, supported_languages_url

18−from searx.utils import match_language, eval_xpath

17+from searx.utils import extract_text, match_language, eval_xpath

19

18

20from urllib.parse import urlencode

19from urllib.parse import urlencode

21from json import loads

20from json import loads

searx/engines/www1x.py+1−1

@@ -12,7 +12,7 @@

12

13from lxml import html

14from urllib.parse import urlencode, urljoin

15−from searx.engines.xpath import extract_text

15+from searx.utils import extract_text

16

17# engine dependent config

18categories = ['images']

searx/engines/xpath.py+2−73

@@ -1,7 +1,6 @@

1−from urllib.parse import unquote, urlencode, urljoin, urlparse

2from lxml import html

1from lxml import html

3−from lxml.etree import _ElementStringResult, _ElementUnicodeResult

2+from urllib.parse import urlencode

4−from searx.utils import html_to_text, eval_xpath

3+from searx.utils import extract_text, extract_url, eval_xpath

5

4

6search_url = None

5search_url = None

7url_xpath = None

6url_xpath = None

@@ -21,76 +20,6 @@ page_size = 1

21first_page_num = 1

20first_page_num = 1

22

21

23

22

24−'''

25−if xpath_results is list, extract the text from each result and concat the list

26−if xpath_results is a xml element, extract all the text node from it

27− ( text_content() method from lxml )

28−if xpath_results is a string element, then it's already done

29−'''

30−

31−

32−def extract_text(xpath_results):

33− if type(xpath_results) == list:

34− # it's list of result : concat everything using recursive call

35− result = ''

36− for e in xpath_results:

37− result = result + extract_text(e)

38− return result.strip()

39− elif type(xpath_results) in [_ElementStringResult, _ElementUnicodeResult]:

40− # it's a string

41− return ''.join(xpath_results)

42− else:

43− # it's a element

44− text = html.tostring(

45− xpath_results, encoding='unicode', method='text', with_tail=False

46− )

47− text = text.strip().replace('\n', ' ')

48− return ' '.join(text.split())

49−

50−

51−def extract_url(xpath_results, search_url):

52− if xpath_results == []:

53− raise Exception('Empty url resultset')

54− url = extract_text(xpath_results)

55−

56− if url.startswith('//'):

57− # add http or https to this kind of url //example.com/

58− parsed_search_url = urlparse(search_url)

59− url = '{0}:{1}'.format(parsed_search_url.scheme or 'http', url)

60− elif url.startswith('/'):

61− # fix relative url to the search engine

62− url = urljoin(search_url, url)

63−

64− # fix relative urls that fall through the crack

65− if '://' not in url:

66− url = urljoin(search_url, url)

67−

68− # normalize url

69− url = normalize_url(url)

70−

71− return url

72−

73−

74−def normalize_url(url):

75− parsed_url = urlparse(url)

76−

77− # add a / at this end of the url if there is no path

78− if not parsed_url.netloc:

79− raise Exception('Cannot parse url')

80− if not parsed_url.path:

81− url += '/'

82−

83− # FIXME : hack for yahoo

84− if parsed_url.hostname == 'search.yahoo.com'\

85− and parsed_url.path.startswith('/r'):

86− p = parsed_url.path

87− mark = p.find('/**')

88− if mark != -1:

89− return unquote(p[mark + 3:]).decode()

90−

91− return url

92−

93−

94def request(query, params):

23def request(query, params):

95 query = urlencode({'q': query})[2:]

24 query = urlencode({'q': query})[2:]

96

25

searx/engines/yahoo.py+1−2

@@ -13,8 +13,7 @@

13

14from urllib.parse import unquote, urlencode

15from lxml import html

16−from searx.engines.xpath import extract_text, extract_url

16+from searx.utils import extract_text, extract_url, match_language, eval_xpath

17−from searx.utils import match_language, eval_xpath

18

17

19# engine dependent config

18# engine dependent config

20categories = ['general']

19categories = ['general']

searx/engines/yahoo_news.py+1−2

@@ -13,12 +13,11 @@ import re

13from datetime import datetime, timedelta

14from urllib.parse import urlencode

15from lxml import html

16−from searx.engines.xpath import extract_text, extract_url

17from searx.engines.yahoo import (

16from searx.engines.yahoo import (

18 parse_url, _fetch_supported_languages, supported_languages_url, language_aliases

17 parse_url, _fetch_supported_languages, supported_languages_url, language_aliases

19)

18)

20from dateutil import parser

19from dateutil import parser

21−from searx.utils import match_language

20+from searx.utils import extract_text, extract_url, match_language

22

21

23# engine dependent config

22# engine dependent config

24categories = ['news']

23categories = ['news']

searx/engines/yggtorrent.py+1−2

@@ -12,8 +12,7 @@ from lxml import html

12from operator import itemgetter

13from datetime import datetime

14from urllib.parse import quote

15−from searx.engines.xpath import extract_text

15+from searx.utils import extract_text, get_torrent_size

16−from searx.utils import get_torrent_size

17from searx.poolrequests import get as http_get

16from searx.poolrequests import get as http_get

18

17

19# engine dependent config

18# engine dependent config

searx/engines/youtube_noapi.py+1−2

@@ -11,8 +11,7 @@

11from functools import reduce

12from json import loads

13from urllib.parse import quote_plus

14−from searx.engines.xpath import extract_text

14+from searx.utils import extract_text, list_get

15−from searx.utils import list_get

16

15

17# engine dependent config

16# engine dependent config

18categories = ['videos', 'music']

17categories = ['videos', 'music']

searx/utils.py+73−1

@@ -10,9 +10,13 @@ from os.path import splitext, join

10from io import open

11from random import choice

12from html.parser import HTMLParser

13+from urllib.parse import urljoin, urlparse, unquote

14+

15+from lxml import html

13−from lxml.etree import XPath

16+from lxml.etree import XPath, _ElementStringResult, _ElementUnicodeResult

14from babel.core import get_global

17from babel.core import get_global

15

18

19+

16from searx import settings

20from searx import settings

17from searx.version import VERSION_STRING

21from searx.version import VERSION_STRING

18from searx.languages import language_codes

22from searx.languages import language_codes

@@ -106,6 +110,74 @@ def html_to_text(html):

106 return s.get_text()

110 return s.get_text()

107

111

108

112

113+def extract_text(xpath_results):

114+ '''

115+ if xpath_results is list, extract the text from each result and concat the list

116+ if xpath_results is a xml element, extract all the text node from it

117+ ( text_content() method from lxml )

118+ if xpath_results is a string element, then it's already done

119+ '''

120+ if type(xpath_results) == list:

121+ # it's list of result : concat everything using recursive call

122+ result = ''

123+ for e in xpath_results:

124+ result = result + extract_text(e)

125+ return result.strip()

126+ elif type(xpath_results) in [_ElementStringResult, _ElementUnicodeResult]:

127+ # it's a string

128+ return ''.join(xpath_results)

129+ else:

130+ # it's a element

131+ text = html.tostring(

132+ xpath_results, encoding='unicode', method='text', with_tail=False

133+ )

134+ text = text.strip().replace('\n', ' ')

135+ return ' '.join(text.split())

136+

137+

138+def extract_url(xpath_results, search_url):

139+ if xpath_results == []:

140+ raise Exception('Empty url resultset')

141+ url = extract_text(xpath_results)

142+

143+ if url.startswith('//'):

144+ # add http or https to this kind of url //example.com/

145+ parsed_search_url = urlparse(search_url)

146+ url = '{0}:{1}'.format(parsed_search_url.scheme or 'http', url)

147+ elif url.startswith('/'):

148+ # fix relative url to the search engine

149+ url = urljoin(search_url, url)

150+

151+ # fix relative urls that fall through the crack

152+ if '://' not in url:

153+ url = urljoin(search_url, url)

154+

155+ # normalize url

156+ url = normalize_url(url)

157+

158+ return url

159+

160+

161+def normalize_url(url):

162+ parsed_url = urlparse(url)

163+

164+ # add a / at this end of the url if there is no path

165+ if not parsed_url.netloc:

166+ raise Exception('Cannot parse url')

167+ if not parsed_url.path:

168+ url += '/'

169+

170+ # FIXME : hack for yahoo

171+ if parsed_url.hostname == 'search.yahoo.com'\

172+ and parsed_url.path.startswith('/r'):

173+ p = parsed_url.path

174+ mark = p.find('/**')

175+ if mark != -1:

176+ return unquote(p[mark + 3:]).decode()

177+

178+ return url

179+

180+

109def dict_subset(d, properties):

181def dict_subset(d, properties):

110 result = {}

182 result = {}

111 for k in properties:

183 for k in properties:

tests/unit/test_utils.py+45−4

@@ -1,4 +1,7 @@

1# -*- coding: utf-8 -*-

2+import lxml.etree

3+from lxml import html

4+

2from searx.testing import SearxTestCase

5from searx.testing import SearxTestCase

3from searx import utils

6from searx import utils

4

7

@@ -16,7 +19,30 @@ class TestUtils(SearxTestCase):

16 self.assertTrue(utils.searx_useragent().startswith('searx'))

19 self.assertTrue(utils.searx_useragent().startswith('searx'))

17

20

18 def test_html_to_text(self):

21 def test_html_to_text(self):

19− html = """

22+ html_str = """

23+ <a href="/testlink" class="link_access_account">

24+ <style>

25+ .toto {

26+ color: red;

27+ }

28+ </style>

29+

30+

31+ <img src="test.jpg" />

32+

33+

34+

35+ Test text

36+

37+ <script>value='dummy';</script>

38+ </a>

39+ """

40+ self.assertIsInstance(utils.html_to_text(html_str), str)

41+ self.assertIsNotNone(utils.html_to_text(html_str))

42+ self.assertEqual(utils.html_to_text(html_str), "Test text")

43+

44+ def test_extract_text(self):

45+ html_str = """

20 <a href="/testlink" class="link_access_account">

46 <a href="/testlink" class="link_access_account">

21

47

22

48

@@ -28,9 +54,24 @@ class TestUtils(SearxTestCase):

28

54

29 </a>

55 </a>

30 """

56 """

31− self.assertIsInstance(utils.html_to_text(html), str)

57+ dom = html.fromstring(html_str)

32− self.assertIsNotNone(utils.html_to_text(html))

58+ self.assertEqual(utils.extract_text(dom), 'Test text')

59+ self.assertEqual(utils.extract_text(dom.xpath('//span')), 'Test text')

60+ self.assertEqual(utils.extract_text(dom.xpath('//img/@src')), 'test.jpg')

33− self.assertEqual(utils.html_to_text(html), "Test text")

61+ self.assertEqual(utils.extract_text(dom.xpath('//unexistingtag')), '')

62+

63+ def test_extract_url(self):

64+ def f(html_str, search_url):

65+ return utils.extract_url(html.fromstring(html_str), search_url)

66+ self.assertEqual(f('https://example.com', 'http://example.com/'), 'https://example.com/')

67+ self.assertEqual(f('https://example.com', 'http://example.com/'), 'https://example.com/')

68+ self.assertEqual(f('//example.com', 'http://example.com/'), 'http://example.com/')

69+ self.assertEqual(f('//example.com', 'https://example.com/'), 'https://example.com/')

70+ self.assertEqual(f('/path?a=1', 'https://example.com'), 'https://example.com/path?a=1')

71+ with self.assertRaises(lxml.etree.ParserError):

72+ f('', 'https://example.com')

73+ with self.assertRaises(Exception):

74+ utils.extract_url([], 'https://example.com')

34

75

35 def test_html_to_text_invalid(self):

76 def test_html_to_text_invalid(self):

36 html = 'Lorem ipsumdolor sit amet'

77 html = 'Lorem ipsumdolor sit amet'