mirror of
https://github.com/searxng/searxng.git
synced 2025-07-19 03:09:25 +02:00
[mod] bing: add time_range support & upgrade to data_type: traits_v1
Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
This commit is contained in:
parent
c9cd376186
commit
d0f465e6fa
6 changed files with 366 additions and 1968 deletions
|
@ -4,16 +4,28 @@
|
|||
|
||||
- https://github.com/searx/searx/issues/2019#issuecomment-648227442
|
||||
"""
|
||||
# pylint: disable=too-many-branches
|
||||
|
||||
# pylint: disable=too-many-branches, invalid-name
|
||||
|
||||
from typing import TYPE_CHECKING
|
||||
import datetime
|
||||
import re
|
||||
from urllib.parse import urlencode, urlparse, parse_qs
|
||||
import uuid
|
||||
from urllib.parse import urlencode
|
||||
from lxml import html
|
||||
from searx.utils import eval_xpath, extract_text, eval_xpath_list, match_language, eval_xpath_getindex
|
||||
from searx.network import multi_requests, Request
|
||||
import babel
|
||||
import babel.languages
|
||||
|
||||
from searx.utils import eval_xpath, extract_text, eval_xpath_list, eval_xpath_getindex
|
||||
from searx import network
|
||||
from searx.locales import language_tag, region_tag
|
||||
from searx.enginelib.traits import EngineTraits
|
||||
|
||||
if TYPE_CHECKING:
|
||||
import logging
|
||||
|
||||
logger: logging.Logger
|
||||
|
||||
traits: EngineTraits
|
||||
|
||||
about = {
|
||||
|
@ -25,56 +37,124 @@ about = {
|
|||
"results": 'HTML',
|
||||
}
|
||||
|
||||
send_accept_language_header = True
|
||||
"""Bing tries to guess user's language and territory from the HTTP
|
||||
Accept-Language. Optional the user can select a search-language (can be
|
||||
different to the UI language) and a region (market code)."""
|
||||
|
||||
# engine dependent config
|
||||
categories = ['general', 'web']
|
||||
paging = True
|
||||
time_range_support = False
|
||||
safesearch = False
|
||||
send_accept_language_header = True
|
||||
supported_languages_url = 'https://www.bing.com/account/general'
|
||||
language_aliases = {}
|
||||
time_range_support = True
|
||||
safesearch = True
|
||||
safesearch_types = {2: 'STRICT', 1: 'DEMOTE', 0: 'OFF'} # cookie: ADLT=STRICT
|
||||
|
||||
# search-url
|
||||
base_url = 'https://www.bing.com/'
|
||||
base_url = 'https://www.bing.com/search'
|
||||
"""Bing (Web) search URL"""
|
||||
|
||||
# initial query: https://www.bing.com/search?q=foo&search=&form=QBLH
|
||||
inital_query = 'search?{query}&search=&form=QBLH'
|
||||
|
||||
# following queries: https://www.bing.com/search?q=foo&search=&first=11&FORM=PERE
|
||||
page_query = 'search?{query}&search=&first={offset}&FORM=PERE'
|
||||
bing_traits_url = 'https://learn.microsoft.com/en-us/bing/search-apis/bing-web-search/reference/market-codes'
|
||||
"""Bing (Web) search API description"""
|
||||
|
||||
|
||||
def _get_offset_from_pageno(pageno):
|
||||
return (pageno - 1) * 10 + 1
|
||||
|
||||
|
||||
def set_bing_cookies(params, engine_language, engine_region, SID):
|
||||
|
||||
# set cookies
|
||||
# -----------
|
||||
|
||||
params['cookies']['_EDGE_V'] = '1'
|
||||
|
||||
# _EDGE_S: F=1&SID=3A5253BD6BCA609509B741876AF961CA&mkt=zh-tw
|
||||
_EDGE_S = [
|
||||
'F=1',
|
||||
'SID=%s' % SID,
|
||||
'mkt=%s' % engine_region.lower(),
|
||||
'ui=%s' % engine_language.lower(),
|
||||
]
|
||||
params['cookies']['_EDGE_S'] = '&'.join(_EDGE_S)
|
||||
logger.debug("cookie _EDGE_S=%s", params['cookies']['_EDGE_S'])
|
||||
|
||||
# "_EDGE_CD": "m=zh-tw",
|
||||
|
||||
_EDGE_CD = [ # pylint: disable=invalid-name
|
||||
'm=%s' % engine_region.lower(), # search region: zh-cn
|
||||
'u=%s' % engine_language.lower(), # UI: en-us
|
||||
]
|
||||
|
||||
params['cookies']['_EDGE_CD'] = '&'.join(_EDGE_CD) + ';'
|
||||
logger.debug("cookie _EDGE_CD=%s", params['cookies']['_EDGE_CD'])
|
||||
|
||||
SRCHHPGUSR = [ # pylint: disable=invalid-name
|
||||
'SRCHLANG=%s' % engine_language,
|
||||
# Trying to set ADLT cookie here seems not to have any effect, I assume
|
||||
# there is some age verification by a cookie (and/or session ID) needed,
|
||||
# to disable the SafeSearch.
|
||||
'ADLT=%s' % safesearch_types.get(params['safesearch'], 'DEMOTE'),
|
||||
]
|
||||
params['cookies']['SRCHHPGUSR'] = '&'.join(SRCHHPGUSR)
|
||||
logger.debug("cookie SRCHHPGUSR=%s", params['cookies']['SRCHHPGUSR'])
|
||||
|
||||
|
||||
def request(query, params):
|
||||
"""Assemble a Bing-Web request."""
|
||||
|
||||
offset = _get_offset_from_pageno(params.get('pageno', 1))
|
||||
engine_region = traits.get_region(params['searxng_locale'], 'en-US')
|
||||
engine_language = traits.get_language(params['searxng_locale'], 'en')
|
||||
|
||||
# logger.debug("params['pageno'] --> %s", params.get('pageno'))
|
||||
# logger.debug(" offset --> %s", offset)
|
||||
SID = uuid.uuid1().hex.upper()
|
||||
CVID = uuid.uuid1().hex.upper()
|
||||
|
||||
search_string = page_query
|
||||
if offset == 1:
|
||||
search_string = inital_query
|
||||
set_bing_cookies(params, engine_language, engine_region, SID)
|
||||
|
||||
if params['language'] == 'all':
|
||||
lang = 'EN'
|
||||
else:
|
||||
lang = match_language(params['language'], supported_languages, language_aliases)
|
||||
# build URL query
|
||||
# ---------------
|
||||
|
||||
query = 'language:{} {}'.format(lang.split('-')[0].upper(), query)
|
||||
# query term
|
||||
page = int(params.get('pageno', 1))
|
||||
query_params = {
|
||||
# fmt: off
|
||||
'q': query,
|
||||
'pq': query,
|
||||
'cvid': CVID,
|
||||
'qs': 'n',
|
||||
'sp': '-1'
|
||||
# fmt: on
|
||||
}
|
||||
|
||||
search_path = search_string.format(query=urlencode({'q': query}), offset=offset)
|
||||
|
||||
if offset > 1:
|
||||
referer = base_url + inital_query.format(query=urlencode({'q': query}))
|
||||
# page
|
||||
if page > 1:
|
||||
referer = base_url + '?' + urlencode(query_params)
|
||||
params['headers']['Referer'] = referer
|
||||
logger.debug("headers.Referer --> %s", referer)
|
||||
|
||||
params['url'] = base_url + search_path
|
||||
params['headers']['Accept'] = 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8'
|
||||
query_params['first'] = _get_offset_from_pageno(page)
|
||||
|
||||
if page == 2:
|
||||
query_params['FORM'] = 'PERE'
|
||||
elif page > 2:
|
||||
query_params['FORM'] = 'PERE%s' % (page - 2)
|
||||
|
||||
filters = ''
|
||||
if params['time_range']:
|
||||
query_params['filt'] = 'custom'
|
||||
|
||||
if params['time_range'] == 'day':
|
||||
filters = 'ex1:"ez1"'
|
||||
elif params['time_range'] == 'week':
|
||||
filters = 'ex1:"ez2"'
|
||||
elif params['time_range'] == 'month':
|
||||
filters = 'ex1:"ez3"'
|
||||
elif params['time_range'] == 'year':
|
||||
epoch_1970 = datetime.date(1970, 1, 1)
|
||||
today_no = (datetime.date.today() - epoch_1970).days
|
||||
filters = 'ex1:"ez5_%s_%s"' % (today_no - 365, today_no)
|
||||
|
||||
params['url'] = base_url + '?' + urlencode(query_params)
|
||||
if filters:
|
||||
params['url'] = params['url'] + '&filters=' + filters
|
||||
return params
|
||||
|
||||
|
||||
|
@ -111,7 +191,8 @@ def response(resp):
|
|||
url_cite = extract_text(eval_xpath(result, './/div[@class="b_attribution"]/cite'))
|
||||
# Bing can shorten the URL either at the end or in the middle of the string
|
||||
if (
|
||||
url_cite.startswith('https://')
|
||||
url_cite
|
||||
and url_cite.startswith('https://')
|
||||
and '…' not in url_cite
|
||||
and '...' not in url_cite
|
||||
and '›' not in url_cite
|
||||
|
@ -131,9 +212,9 @@ def response(resp):
|
|||
|
||||
# resolve all Bing redirections in parallel
|
||||
request_list = [
|
||||
Request.get(u, allow_redirects=False, headers=resp.search_params['headers']) for u in url_to_resolve
|
||||
network.Request.get(u, allow_redirects=False, headers=resp.search_params['headers']) for u in url_to_resolve
|
||||
]
|
||||
response_list = multi_requests(request_list)
|
||||
response_list = network.multi_requests(request_list)
|
||||
for i, redirect_response in enumerate(response_list):
|
||||
if not isinstance(redirect_response, Exception):
|
||||
results[url_to_resolve_index[i]]['url'] = redirect_response.headers['location']
|
||||
|
@ -161,94 +242,43 @@ def response(resp):
|
|||
return results
|
||||
|
||||
|
||||
# get supported languages from their site
|
||||
def _fetch_supported_languages(resp):
|
||||
|
||||
lang_tags = set()
|
||||
|
||||
dom = html.fromstring(resp.text)
|
||||
lang_links = eval_xpath(dom, '//div[@id="language-section"]//li')
|
||||
|
||||
for _li in lang_links:
|
||||
|
||||
href = eval_xpath(_li, './/@href')[0]
|
||||
(_scheme, _netloc, _path, _params, query, _fragment) = urlparse(href)
|
||||
query = parse_qs(query, keep_blank_values=True)
|
||||
|
||||
# fmt: off
|
||||
setlang = query.get('setlang', [None, ])[0]
|
||||
# example: 'mn-Cyrl-MN' --> '['mn', 'Cyrl-MN']
|
||||
lang, nation = (setlang.split('-', maxsplit=1) + [None,])[:2] # fmt: skip
|
||||
# fmt: on
|
||||
|
||||
tag = lang + '-' + nation if nation else lang
|
||||
lang_tags.add(tag)
|
||||
|
||||
return list(lang_tags)
|
||||
|
||||
|
||||
def fetch_traits(engine_traits: EngineTraits):
|
||||
"""Fetch languages and regions from bing."""
|
||||
"""Fetch languages and regions from Bing-Web."""
|
||||
|
||||
# pylint: disable=import-outside-toplevel, disable=too-many-branches,
|
||||
# pylint: disable=too-many-locals, too-many-statements
|
||||
xpath_market_codes = '//table[1]/tbody/tr/td[3]'
|
||||
# xpath_country_codes = '//table[2]/tbody/tr/td[2]'
|
||||
xpath_language_codes = '//table[3]/tbody/tr/td[2]'
|
||||
|
||||
engine_traits.data_type = 'supported_languages' # deprecated
|
||||
_fetch_traits(engine_traits, bing_traits_url, xpath_language_codes, xpath_market_codes)
|
||||
|
||||
import babel
|
||||
import babel.languages
|
||||
from searx import network
|
||||
from searx.locales import get_offical_locales, language_tag, region_tag
|
||||
from searx.utils import gen_useragent
|
||||
|
||||
headers = {
|
||||
'User-Agent': gen_useragent(),
|
||||
'Accept-Language': "en-US,en;q=0.5", # bing needs to set the English language
|
||||
}
|
||||
resp = network.get('https://www.bing.com/account/general', headers=headers)
|
||||
def _fetch_traits(engine_traits: EngineTraits, url: str, xpath_language_codes: str, xpath_market_codes: str):
|
||||
|
||||
# insert alias to map from a language (zh) to a language + script (zh_Hans)
|
||||
engine_traits.languages['zh'] = 'zh-hans'
|
||||
|
||||
resp = network.get(url)
|
||||
|
||||
if not resp.ok:
|
||||
print("ERROR: response from peertube is not OK.")
|
||||
|
||||
dom = html.fromstring(resp.text)
|
||||
|
||||
# Selector to get items from "Display language"
|
||||
map_lang = {'jp': 'ja'}
|
||||
for td in eval_xpath(dom, xpath_language_codes):
|
||||
eng_lang = td.text
|
||||
|
||||
lang_map = {
|
||||
'prs': 'fa', # Persian
|
||||
'pt_BR': 'pt', # Portuguese (Brasil)
|
||||
'pt_PT': 'pt', # Portuguese (Portugal)
|
||||
'ca-ES-VALENCIA': 'ca', # Catalan (Spain, Valencian)
|
||||
}
|
||||
|
||||
unknow_langs = [
|
||||
'quc', # K'iche'
|
||||
'nso', # Sesotho sa Leboa
|
||||
'tn', # Setswana
|
||||
]
|
||||
|
||||
for div in eval_xpath(dom, '//div[@id="limit-languages"]//input/..'):
|
||||
|
||||
eng_lang = eval_xpath(div, './/input/@value')[0]
|
||||
if eng_lang in unknow_langs:
|
||||
if eng_lang in ('en-gb', 'pt-br'):
|
||||
# language 'en' is already in the list and a language 'en-gb' can't
|
||||
# be handled in SearXNG, same with pt-br which is covered by pt-pt.
|
||||
continue
|
||||
|
||||
eng_lang = lang_map.get(eng_lang, eng_lang)
|
||||
label = extract_text(eval_xpath(div, './/label'))
|
||||
|
||||
# The 'language:xx' query string in the request function (above) does
|
||||
# only support the language codes from the "Display languages" list.
|
||||
# Examples of items from the "Display languages" not sopported in the
|
||||
# query string: zh_Hans --> zh / sr_latn --> sr
|
||||
#
|
||||
# eng_lang = eng_lang.split('_')[0]
|
||||
|
||||
babel_lang = map_lang.get(eng_lang, eng_lang).replace('-', '_')
|
||||
try:
|
||||
sxng_tag = language_tag(babel.Locale.parse(eng_lang.replace('-', '_'), sep='_'))
|
||||
sxng_tag = language_tag(babel.Locale.parse(babel_lang))
|
||||
except babel.UnknownLocaleError:
|
||||
print("ERROR: %s (%s) is unknown by babel" % (label, eng_lang))
|
||||
print("ERROR: language (%s) is unknown by babel" % (eng_lang))
|
||||
continue
|
||||
|
||||
conflict = engine_traits.languages.get(sxng_tag)
|
||||
if conflict:
|
||||
if conflict != eng_lang:
|
||||
|
@ -256,25 +286,27 @@ def fetch_traits(engine_traits: EngineTraits):
|
|||
continue
|
||||
engine_traits.languages[sxng_tag] = eng_lang
|
||||
|
||||
engine_traits.languages['zh'] = 'zh_Hans'
|
||||
map_region = {
|
||||
'en-ID': 'id_ID',
|
||||
'no-NO': 'nb_NO',
|
||||
}
|
||||
|
||||
# regiones
|
||||
for td in eval_xpath(dom, xpath_market_codes):
|
||||
eng_region = td.text
|
||||
babel_region = map_region.get(eng_region, eng_region).replace('-', '_')
|
||||
|
||||
for a in eval_xpath(dom, '//div[@id="region-section-content"]//li/a'):
|
||||
href = eval_xpath(a, './/@href')[0]
|
||||
# lang_name = extract_text(a)
|
||||
query = urlparse(href)[4]
|
||||
query = parse_qs(query, keep_blank_values=True)
|
||||
cc = query.get('cc')[0] # pylint:disable=invalid-name
|
||||
if cc == 'clear':
|
||||
if eng_region == 'en-WW':
|
||||
engine_traits.all_locale = eng_region
|
||||
continue
|
||||
|
||||
# Assert babel supports this locales
|
||||
sxng_locales = get_offical_locales(cc.upper(), engine_traits.languages.keys())
|
||||
|
||||
if not sxng_locales:
|
||||
# print("ERROR: can't map from bing country %s (%s) to a babel region." % (a.text_content().strip(), cc))
|
||||
try:
|
||||
sxng_tag = region_tag(babel.Locale.parse(babel_region))
|
||||
except babel.UnknownLocaleError:
|
||||
print("ERROR: region (%s) is unknown by babel" % (eng_region))
|
||||
continue
|
||||
|
||||
for sxng_locale in sxng_locales:
|
||||
engine_traits.regions[region_tag(sxng_locale)] = cc
|
||||
conflict = engine_traits.regions.get(sxng_tag)
|
||||
if conflict:
|
||||
if conflict != eng_region:
|
||||
print("CONFLICT: babel %s --> %s, %s" % (sxng_tag, conflict, eng_region))
|
||||
continue
|
||||
engine_traits.regions[sxng_tag] = eng_region
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue