Unverified Commit a51b2b6c authored by Mathieu Brunot's avatar Mathieu Brunot Committed by GitHub

Merge branch 'master' into feature/accessibility

parents 4d17d453 12f42d15
......@@ -11,7 +11,9 @@ ARG TIMESTAMP_UWSGI=0
ARG LABEL_VCS_REF=
ARG LABEL_VCS_URL=
ENV BASE_URL= \
ENV INSTANCE_NAME=searx \
AUTOCOMPLETE= \
BASE_URL= \
MORTY_KEY= \
MORTY_URL=
EXPOSE 8080
......
......@@ -29,6 +29,8 @@ do
printf " -f Always update on the configuration files (existing files are renamed with the .old suffix)\n"
printf " Without this option, new configuration files are copied with the .new suffix\n"
printf "\nEnvironment variables:\n\n"
printf " INSTANCE_NAME settings.yml : general.instance_name\n"
printf " AUTOCOMPLETE settings.yml : search.autocomplete\n"
printf " BASE_URL settings.yml : server.base_url\n"
printf " MORTY_URL settings.yml : result_proxy.url\n"
printf " MORTY_KEY settings.yml : result_proxy.key\n"
......@@ -53,6 +55,8 @@ patch_searx_settings() {
# update settings.yml
sed -i -e "s|base_url : False|base_url : ${BASE_URL}|g" \
-e "s/instance_name : \"searx\"/instance_name : \"${INSTANCE_NAME}\"/g" \
-e "s/autocomplete : \"\"/autocomplete : \"${AUTOCOMPLETE}\"/g" \
-e "s/ultrasecretkey/$(openssl rand -hex 32)/g" \
"${CONF}"
......
......@@ -27,7 +27,7 @@ from json import loads
from requests import get
from searx import settings
from searx import logger
from searx.utils import load_module, match_language
from searx.utils import load_module, match_language, get_engine_from_settings
logger = logger.getChild('engines')
......@@ -53,7 +53,8 @@ engine_default_args = {'paging': False,
'disabled': False,
'suspend_end_time': 0,
'continuous_errors': 0,
'time_range_support': False}
'time_range_support': False,
'offline': False}
def load_engine(engine_data):
......@@ -128,14 +129,16 @@ def load_engine(engine_data):
engine.stats = {
'result_count': 0,
'search_count': 0,
'page_load_time': 0,
'page_load_count': 0,
'engine_time': 0,
'engine_time_count': 0,
'score_count': 0,
'errors': 0
}
if not engine.offline:
engine.stats['page_load_time'] = 0
engine.stats['page_load_count'] = 0
for category_name in engine.categories:
categories.setdefault(category_name, []).append(engine)
......@@ -173,11 +176,6 @@ def get_engines_stats():
results_num = \
engine.stats['result_count'] / float(engine.stats['search_count'])
if engine.stats['page_load_count'] != 0:
load_times = engine.stats['page_load_time'] / float(engine.stats['page_load_count']) # noqa
else:
load_times = 0
if engine.stats['engine_time_count'] != 0:
this_engine_time = engine.stats['engine_time'] / float(engine.stats['engine_time_count']) # noqa
else:
......@@ -189,14 +187,19 @@ def get_engines_stats():
else:
score = score_per_result = 0.0
max_pageload = max(load_times, max_pageload)
if not engine.offline:
load_times = 0
if engine.stats['page_load_count'] != 0:
load_times = engine.stats['page_load_time'] / float(engine.stats['page_load_count']) # noqa
max_pageload = max(load_times, max_pageload)
pageloads.append({'avg': load_times, 'name': engine.name})
max_engine_times = max(this_engine_time, max_engine_times)
max_results = max(results_num, max_results)
max_score = max(score, max_score)
max_score_per_result = max(score_per_result, max_score_per_result)
max_errors = max(max_errors, engine.stats['errors'])
pageloads.append({'avg': load_times, 'name': engine.name})
engine_times.append({'avg': this_engine_time, 'name': engine.name})
results.append({'avg': results_num, 'name': engine.name})
scores.append({'avg': score, 'name': engine.name})
......@@ -255,7 +258,7 @@ def initialize_engines(engine_list):
load_engines(engine_list)
def engine_init(engine_name, init_fn):
init_fn()
init_fn(get_engine_from_settings(engine_name))
logger.debug('%s engine: Initialized', engine_name)
for engine_name, engine in engines.items():
......
......@@ -17,6 +17,7 @@ from searx.url_utils import urlencode
categories = ['science']
paging = True
base_url = 'http://export.arxiv.org/api/query?search_query=all:'\
+ '{query}&start={offset}&max_results={number_of_results}'
......
......@@ -24,7 +24,7 @@ time_range_support = True
# search-url
base_url = 'https://www.deviantart.com/'
search_url = base_url + 'browse/all/?offset={offset}&{query}'
search_url = base_url + 'search?page={page}&{query}'
time_range_url = '&order={range}'
time_range_dict = {'day': 11,
......@@ -37,9 +37,7 @@ def request(query, params):
if params['time_range'] and params['time_range'] not in time_range_dict:
return params
offset = (params['pageno'] - 1) * 24
params['url'] = search_url.format(offset=offset,
params['url'] = search_url.format(page=params['pageno'],
query=urlencode({'q': query}))
if params['time_range'] in time_range_dict:
params['url'] += time_range_url.format(range=time_range_dict[params['time_range']])
......@@ -57,28 +55,27 @@ def response(resp):
dom = html.fromstring(resp.text)
regex = re.compile(r'\/200H\/')
# parse results
for result in dom.xpath('.//span[@class="thumb wide"]'):
link = result.xpath('.//a[@class="torpedo-thumb-link"]')[0]
url = link.attrib.get('href')
title = extract_text(result.xpath('.//span[@class="title"]'))
thumbnail_src = link.xpath('.//img')[0].attrib.get('src')
img_src = regex.sub('/', thumbnail_src)
# http to https, remove domain sharding
thumbnail_src = re.sub(r"https?://(th|fc)\d+.", "https://th01.", thumbnail_src)
thumbnail_src = re.sub(r"http://", "https://", thumbnail_src)
url = re.sub(r"http://(.*)\.deviantart\.com/", "https://\\1.deviantart.com/", url)
# append result
results.append({'url': url,
'title': title,
'img_src': img_src,
'thumbnail_src': thumbnail_src,
'template': 'images.html'})
for row in dom.xpath('//div[contains(@data-hook, "content_row")]'):
for result in row.xpath('./div'):
link = result.xpath('.//a[@data-hook="deviation_link"]')[0]
url = link.attrib.get('href')
title = link.attrib.get('title')
thumbnail_src = result.xpath('.//img')[0].attrib.get('src')
img_src = thumbnail_src
# http to https, remove domain sharding
thumbnail_src = re.sub(r"https?://(th|fc)\d+.", "https://th01.", thumbnail_src)
thumbnail_src = re.sub(r"http://", "https://", thumbnail_src)
url = re.sub(r"http://(.*)\.deviantart\.com/", "https://\\1.deviantart.com/", url)
# append result
results.append({'url': url,
'title': title,
'img_src': img_src,
'thumbnail_src': thumbnail_src,
'template': 'images.html'})
# return results
return results
......@@ -15,7 +15,8 @@ import string
from dateutil import parser
from json import loads
from lxml import html
from searx.url_utils import quote_plus
from searx.url_utils import urlencode
from datetime import datetime
# engine dependent config
categories = ['news', 'social media']
......@@ -23,7 +24,7 @@ paging = True
# search-url
base_url = 'https://digg.com/'
search_url = base_url + 'api/search/{query}.json?position={position}&format=html'
search_url = base_url + 'api/search/?{query}&from={position}&size=20&format=html'
# specific xpath variables
results_xpath = '//article'
......@@ -38,9 +39,9 @@ digg_cookie_chars = string.ascii_uppercase + string.ascii_lowercase +\
# do search-request
def request(query, params):
offset = (params['pageno'] - 1) * 10
offset = (params['pageno'] - 1) * 20
params['url'] = search_url.format(position=offset,
query=quote_plus(query))
query=urlencode({'q': query}))
params['cookies']['frontend.auid'] = ''.join(random.choice(
digg_cookie_chars) for _ in range(22))
return params
......@@ -52,30 +53,17 @@ def response(resp):
search_result = loads(resp.text)
if 'html' not in search_result or search_result['html'] == '':
return results
dom = html.fromstring(search_result['html'])
# parse results
for result in dom.xpath(results_xpath):
url = result.attrib.get('data-contenturl')
thumbnail = result.xpath('.//img')[0].attrib.get('src')
title = ''.join(result.xpath(title_xpath))
content = ''.join(result.xpath(content_xpath))
pubdate = result.xpath(pubdate_xpath)[0].attrib.get('datetime')
publishedDate = parser.parse(pubdate)
# http to https
thumbnail = thumbnail.replace("http://static.digg.com", "https://static.digg.com")
for result in search_result['mapped']:
published = datetime.strptime(result['created']['ISO'], "%Y-%m-%d %H:%M:%S")
# append result
results.append({'url': url,
'title': title,
'content': content,
results.append({'url': result['url'],
'title': result['title'],
'content': result['excerpt'],
'template': 'videos.html',
'publishedDate': publishedDate,
'thumbnail': thumbnail})
'publishedDate': published,
'thumbnail': result['images']['thumbImage']})
# return results
return results
......@@ -65,21 +65,36 @@ def get_region_code(lang, lang_list=[]):
def request(query, params):
if params['time_range'] and params['time_range'] not in time_range_dict:
if params['time_range'] not in (None, 'None', '') and params['time_range'] not in time_range_dict:
return params
offset = (params['pageno'] - 1) * 30
region_code = get_region_code(params['language'], supported_languages)
if region_code:
params['url'] = url.format(
query=urlencode({'q': query, 'kl': region_code}), offset=offset, dc_param=offset)
params['url'] = 'https://duckduckgo.com/html/'
if params['pageno'] > 1:
params['method'] = 'POST'
params['data']['q'] = query
params['data']['s'] = offset
params['data']['dc'] = 30
params['data']['nextParams'] = ''
params['data']['v'] = 'l'
params['data']['o'] = 'json'
params['data']['api'] = '/d.js'
if params['time_range'] in time_range_dict:
params['data']['df'] = time_range_dict[params['time_range']]
if region_code:
params['data']['kl'] = region_code
else:
params['url'] = url.format(
query=urlencode({'q': query}), offset=offset, dc_param=offset)
if region_code:
params['url'] = url.format(
query=urlencode({'q': query, 'kl': region_code}), offset=offset, dc_param=offset)
else:
params['url'] = url.format(
query=urlencode({'q': query}), offset=offset, dc_param=offset)
if params['time_range'] in time_range_dict:
params['url'] += time_range_url.format(range=time_range_dict[params['time_range']])
if params['time_range'] in time_range_dict:
params['url'] += time_range_url.format(range=time_range_dict[params['time_range']])
return params
......@@ -91,7 +106,9 @@ def response(resp):
doc = fromstring(resp.text)
# parse results
for r in doc.xpath(result_xpath):
for i, r in enumerate(doc.xpath(result_xpath)):
if i >= 30:
break
try:
res_url = r.xpath(url_xpath)[-1]
except:
......
......@@ -35,8 +35,8 @@ search_string = 'search?{query}'\
'&ff={safesearch}'\
'&rxiec={rxieu}'\
'&ulse={ulse}'\
'&rand={rxikd}' # current unix timestamp
'&rand={rxikd}'\
'&dbez={dbez}'
# specific xpath variables
results_xpath = '//response//result'
url_xpath = './/url'
......@@ -70,7 +70,8 @@ def request(query, params):
rxieu=random.randint(1000000000, 9999999999),
ulse=random.randint(100000000, 999999999),
lang=language,
safesearch=safesearch)
safesearch=safesearch,
dbez=random.randint(100000000, 999999999))
params['url'] = base_url + search_path
......
......@@ -66,7 +66,7 @@ def get_client_id():
return ""
def init():
def init(engine_settings=None):
global guest_client_id
# api-key
guest_client_id = get_client_id()
......
......@@ -15,6 +15,7 @@ from dateutil import parser
from datetime import datetime, timedelta
import re
from searx.engines.xpath import extract_text
from searx.languages import language_codes
# engine dependent config
categories = ['general']
......@@ -22,7 +23,7 @@ categories = ['general']
# (probably the parameter qid), require
# storing of qid's between mulitble search-calls
# paging = False
paging = True
language_support = True
# search-url
......@@ -32,23 +33,32 @@ search_url = base_url + 'do/search'
# specific xpath variables
# ads xpath //div[@id="results"]/div[@id="sponsored"]//div[@class="result"]
# not ads: div[@class="result"] are the direct childs of div[@id="results"]
results_xpath = '//li[contains(@class, "search-result") and contains(@class, "search-item")]'
link_xpath = './/h3/a'
content_xpath = './p[@class="search-item__body"]'
results_xpath = '//div[@class="w-gl__result"]'
link_xpath = './/a[@class="w-gl__result-title"]'
content_xpath = './/p[@class="w-gl__description"]'
# do search-request
def request(query, params):
offset = (params['pageno'] - 1) * 10
params['url'] = search_url
params['method'] = 'POST'
params['data'] = {'query': query,
'startat': offset}
params['data'] = {
'query': query,
'page': params['pageno'],
'cat': 'web',
'cmd': 'process_search',
'engine0': 'v1all',
}
# set language if specified
if params['language'] != 'all':
params['data']['with_language'] = ('lang_' + params['language'].split('-')[0])
language = 'english'
for lc, _, _, lang in language_codes:
if lc == params['language']:
language = lang
params['data']['language'] = language
params['data']['lui'] = language
return params
......
......@@ -55,7 +55,7 @@ def obtain_token():
return token
def init():
def init(engine_settings=None):
obtain_token()
......
......@@ -11,8 +11,8 @@
"""
from lxml import html
import re
from searx.url_utils import urlencode, urljoin
from searx.engines.xpath import extract_text
# engine dependent config
categories = ['images']
......@@ -34,41 +34,18 @@ def request(query, params):
def response(resp):
results = []
# get links from result-text
regex = re.compile('(</a>|<a)')
results_parts = re.split(regex, resp.text)
cur_element = ''
# iterate over link parts
for result_part in results_parts:
dom = html.fromstring(resp.text)
for res in dom.xpath('//div[@class="List-item MainListing"]'):
# processed start and end of link
if result_part == '<a':
cur_element = result_part
continue
elif result_part != '</a>':
cur_element += result_part
continue
cur_element += result_part
# fix xml-error
cur_element = cur_element.replace('"></a>', '"/></a>')
dom = html.fromstring(cur_element)
link = dom.xpath('//a')[0]
link = res.xpath('//a')[0]
url = urljoin(base_url, link.attrib.get('href'))
title = link.attrib.get('title', '')
title = extract_text(link)
thumbnail_src = urljoin(base_url, link.xpath('.//img')[0].attrib['src'])
thumbnail_src = urljoin(base_url, res.xpath('.//img')[0].attrib['src'])
# TODO: get image with higher resolution
img_src = thumbnail_src
# check if url is showing to a photo
if '/photo/' not in url:
continue
# append result
results.append({'url': url,
'title': title,
......
......@@ -28,5 +28,6 @@ class SearxParameterException(SearxException):
else:
message = 'Invalid value "' + value + '" for parameter ' + name
super(SearxParameterException, self).__init__(message)
self.message = message
self.parameter_name = name
self.parameter_value = value
......@@ -225,6 +225,9 @@ def https_url_rewrite(result):
def on_result(request, search, result):
if 'parsed_url' not in result:
return True
if result['parsed_url'].scheme == 'http':
https_url_rewrite(result)
return True
......
......@@ -35,6 +35,9 @@ def get_doi_resolver(args, preference_doi_resolver):
def on_result(request, search, result):
if 'parsed_url' not in result:
return True
doi = extract_doi(result['parsed_url'])
if doi and len(doi) < 50:
for suffix in ('/', '.pdf', '/full', '/meta', '/abstract'):
......
......@@ -17,10 +17,10 @@ along with searx. If not, see < http://www.gnu.org/licenses/ >.
from flask_babel import gettext
import re
from searx.url_utils import urlunparse
from searx.url_utils import urlunparse, parse_qsl, urlencode
regexes = {re.compile(r'utm_[^&]+&?'),
re.compile(r'(wkey|wemail)[^&]+&?'),
regexes = {re.compile(r'utm_[^&]+'),
re.compile(r'(wkey|wemail)[^&]*'),
re.compile(r'&$')}
name = gettext('Tracker URL remover')
......@@ -30,16 +30,25 @@ preference_section = 'privacy'
def on_result(request, search, result):
if 'parsed_url' not in result:
return True
query = result['parsed_url'].query
if query == "":
return True
for reg in regexes:
query = reg.sub('', query)
if query != result['parsed_url'].query:
result['parsed_url'] = result['parsed_url']._replace(query=query)
result['url'] = urlunparse(result['parsed_url'])
parsed_query = parse_qsl(query)
changed = False
for i, (param_name, _) in enumerate(list(parsed_query)):
for reg in regexes:
if reg.match(param_name):
parsed_query.pop(i)
changed = True
break
if changed:
result['parsed_url'] = result['parsed_url']._replace(query=urlencode(parsed_query))
result['url'] = urlunparse(result['parsed_url'])
return True
......@@ -184,7 +184,7 @@ class SearchQuery(object):
self.lang = lang
self.safesearch = safesearch
self.pageno = pageno
self.time_range = time_range
self.time_range = None if time_range in ('', 'None', None) else time_range
self.timeout_limit = timeout_limit
def __str__(self):
......
......@@ -197,6 +197,13 @@ class ResultContainer(object):
self.infoboxes.append(infobox)
def _merge_result(self, result, position):
if 'url' in result:
self.__merge_url_result(result, position)
return
self.__merge_result_no_url(result, position)
def __merge_url_result(self, result, position):
result['parsed_url'] = urlparse(result['url'])
# if the result has no scheme, use http as default
......@@ -210,51 +217,60 @@ class ResultContainer(object):
if result.get('content'):
result['content'] = WHITESPACE_REGEX.sub(' ', result['content'])
# check for duplicates
duplicated = False
duplicated = self.__find_duplicated_http_result(result)
if duplicated:
self.__merge_duplicated_http_result(duplicated, result, position)
return
# if there is no duplicate found, append result
result['positions'] = [position]
with RLock():
self._merged_results.append(result)
def __find_duplicated_http_result(self, result):
result_template = result.get('template')
for merged_result in self._merged_results:
if 'parsed_url' not in merged_result:
continue
if compare_urls(result['parsed_url'], merged_result['parsed_url'])\
and result_template == merged_result.get('template'):
if result_template != 'images.html':
# not an image, same template, same url : it's a duplicate
duplicated = merged_result
break
return merged_result
else:
# it's an image
# it's a duplicate if the parsed_url, template and img_src are differents
if result.get('img_src', '') == merged_result.get('img_src', ''):
duplicated = merged_result
break
return merged_result
return None
# merge duplicates together
if duplicated:
# using content with more text
if result_content_len(result.get('content', '')) >\
result_content_len(duplicated.get('content', '')):
duplicated['content'] = result['content']
def __merge_duplicated_http_result(self, duplicated, result, position):
# using content with more text
if result_content_len(result.get('content', '')) >\
result_content_len(duplicated.get('content', '')):
duplicated['content'] = result['content']
# merge all result's parameters not found in duplicate
for key in result.keys():
if not duplicated.get(key):
duplicated[key] = result.get(key)
# merge all result's parameters not found in duplicate
for key in result.keys():
if not duplicated.get(key):
duplicated[key] = result.get(key)
# add the new position
duplicated['positions'].append(position)
# add the new position
duplicated['positions'].append(position)
# add engine to list of result-engines
duplicated['engines'].add(result['engine'])
# add engine to list of result-engines
duplicated['engines'].add(result['engine'])
# using https if possible
if duplicated['parsed_url'].scheme != 'https' and result['parsed_url'].scheme == 'https':
duplicated['url'] = result['parsed_url'].geturl()
duplicated['parsed_url'] = result['parsed_url']
# using https if possible
if duplicated['parsed_url'].scheme != 'https' and result['parsed_url'].scheme == 'https':
duplicated['url'] = result['parsed_url'].geturl()
duplicated['parsed_url'] = result['parsed_url']
# if there is no duplicate found, append result
else:
result['positions'] = [position]
with RLock():
self._merged_results.append(result)
def __merge_result_no_url(self, result, position):
result['engines'] = set([result['engine']])
result['positions'] = [position]
with RLock():
self._merged_results.append(result)
def order_results(self):
for result in self._merged_results:
......
......@@ -77,7 +77,7 @@ def send_http_request(engine, request_params):
return req(request_params['url'], **request_args)
def search_one_request(engine, query, request_params):
def search_one_http_request(engine, query, request_params):
# update request parameters dependent on
# search-engine (contained in engines folder)
engine.request(query, request_params)
......@@ -97,7 +97,53 @@ def search_one_request(engine, query, request_params):
return engine.response(response)
def search_one_offline_request(engine, query, request_params):
return engine.search(query, request_params)
def search_one_request_safe(engine_name, query, request_params, result_container, start_time, timeout_limit):
if engines[engine_name].offline:
return search_one_offline_request_safe(engine_name, query, request_params, result_container, start_time, timeout_limit) # noqa
return search_one_http_request_safe(engine_name, query, request_params, result_container, start_time, timeout_limit)
def search_one_offline_request_safe(engine_name, query, request_params, result_container, start_time, timeout_limit):
engine = engines[engine_name]
try:
search_results = search_one_offline_request(engine, query, request_params)
if search_results:
result_container.extend(engine_name, search_results)
engine_time = time() - start_time
result_container.add_timing(engine_name, engine_time, engine_time)
with threading.RLock():
engine.stats['engine_time'] += engine_time
engine.stats['engine_time_count'] += 1
except ValueError as e:
record_offline_engine_stats_on_error(engine, result_container, start_time)
logger.exception('engine {0} : invalid input : {1}'.format(engine_name, e))
except Exception as e:
record_offline_engine_stats_on_error(engine, result_container, start_time)
result_container.add_unresponsive_engine((
engine_name,
u'{0}: {1}'.format(gettext('unexpected crash'), e),
))
logger.exception('engine {0} : exception : {1}'.format(engine_name, e))
def record_offline_engine_stats_on_error(engine, result_container, start_time):
engine_time = time() - start_time
result_container.add_timing(engine.name, engine_time, engine_time)
with threading.RLock():
engine.stats['errors'] += 1
def search_one_http_request_safe(engine_name, query, request_params, result_container, start_time, timeout_limit):
# set timeout for all HTTP requests
requests_lib.set_timeout_for_thread(timeout_limit, start_time=start_time)
# reset the HTTP total time
......@@ -111,7 +157,7 @@ def search_one_request_safe(engine_name, query, request_params, result_container
try:
# send requests and parse the results
search_results = search_one_request(engine, query, request_params)
search_results = search_one_http_request(engine, query, request_params)
# check if the engine accepted the request
if search_results is not None:
......@@ -427,19 +473,21 @@ class Search(object):
continue
# set default request parameters
request_params = default_request_params()
request_params['headers']['User-Agent'] = user_agent
request_params['category'] = selected_engine['category']
request_params['pageno'] = search_query.pageno
request_params = {}
if not engine.offline:
request_params = default_request_params()
request_params['headers']['User-Agent'] = user_agent
if hasattr(engine, 'language') and engine.language:
request_params['language'] = engine.language
else:
request_params['language'] = search_query.lang
if hasattr(engine, 'language') and engine.language:
request_params['language'] = engine.language
else:
request_params['language'] = search_query.lang
# 0 = None, 1 = Moderate, 2 = Strict
request_params['safesearch'] = search_query.safesearch
request_params['time_range'] = search_query.time_range
request_params['safesearch'] = search_query.safesearch
request_params['time_range'] = search_query.time_range
request_params['category'] = selected_engine['category']
request_params['pageno'] = search_query.pageno
# append request to list
requests.append((selected_engine['name'], search_query.query, request_params))
......
......@@ -161,11 +161,12 @@ engines:
weight : 2
disabled : True
- name : digbt
engine : digbt
shortcut : dbt
timeout : 6.0
disabled : True
# cloudflare protected
# - name : digbt
# engine : digbt
# shortcut : dbt
# timeout : 6.0
# disabled : True
- name : digg
engine : digg
......@@ -703,9 +704,9 @@ engines:
shortcut: vo
categories: social media
search_url : https://searchvoat.co/?t={query}
url_xpath : //div[@class="entry"]/p/a[contains(@class, "title")]/@href
title_xpath : //div[@class="entry"]/p/a[contains(@class, "title")]
content_xpath : //div[@class="entry"]/p/span[@class="domain"]/a/text()
url_xpath : //div[@class="entry"]//p[@class="title"]/a/@href
title_xpath : //div[@class="entry"]//p[@class="title"]/a/text()
content_xpath : //div[@class="entry"]//span[@class="domain"]/a/text()
timeout : 10.0
disabled : True
......
This diff is collapsed.
......@@ -325,6 +325,10 @@ a {
font-size: 0.9em;
}
.result .engines {
text-align: right;
}
.result .content {
margin: 0;
color: #666;
......
This diff is collapsed.
......@@ -376,6 +376,10 @@ table {
width: 100%;
}
.result-table {
margin-bottom: 10px;
}
td {
padding: 0 4px;
}
......
<div class="result">
<table>
{% for key, value in result.items() %}
{% if key in ['engine', 'engines', 'template', 'score', 'category', 'positions'] %}
{% continue %}
{% endif %}
<tr>
<td><b>{{ key|upper }}</b>: {{ value|safe }}</td>
</tr>
{% endfor %}
</table>
<p class="engines">{{ result.engines|join(', ') }}</p>
</div>
<table class="result-table">
{% for key, value in result.items() %}
{% if key in ['engine', 'engines', 'template', 'score', 'category', 'positions'] %}
{% continue %}
{% endif %}
<tr>
<td><b>{{ key|upper }}</b>: {{ value|safe }}</td>
</tr>
{% endfor %}
<tr>
<td><b>ENGINES</b>: {{ result.engines|join(', ') }}</td>
</tr>
</table>
......@@ -14,7 +14,7 @@
<!-- Draw result header -->
{% macro result_header(result, favicons) -%}
<h4 class="result_header">{% if result.engine~".png" in favicons %}{{ draw_favicon(result.engine) }} {% endif %}{{ result_link(result.url, result.title|safe) }}</h4>
<h4 class="result_header">{% if result.engine~".png" in favicons %}{{ draw_favicon(result.engine) }} {% endif %}{% if result.url %}{{ result_link(result.url, result.title|safe) }}{% else %}{{ result.title|safe}}{% endif %}</h4>
{%- endmacro %}
<!-- Draw result sub header -->
......@@ -31,12 +31,16 @@
{% for engine in result.engines %}
<span class="label label-default">{{ engine }}</span>
{% endfor %}
{% if result.url %}
<small>{{ result_link("https://web.archive.org/web/" + result.url, icon('link') + _('cached'), "text-info") }}</small>
{% endif %}
{% if proxify %}
<small>{{ result_link(proxify(result.url), icon('sort') + _('proxied'), "text-info") }}</small>
{% endif %}
</div>
{% if result.pretty_url %}
<div class="external-link">{{ result.pretty_url }}</div>
{% endif %}
{%- endmacro %}
<!-- Draw result footer -->
......@@ -45,11 +49,15 @@
{% for engine in result.engines %}
<span class="label label-default">{{ engine }}</span>
{% endfor %}
{% if result.url %}
<small>{{ result_link("https://web.archive.org/web/" + result.url, icon('link') + _('cached'), "text-info") }}</small>
{% endif %}
{% if proxify %}
<small>{{ result_link(proxify(result.url), icon('sort') + _('proxied'), "text-info") }}</small>
{% endif %}
{% if result.pretty_url %}
<div class="external-link">{{ result.pretty_url }}</div>
{% endif %}
{%- endmacro %}
{% macro preferences_item_header(info, label, rtl) -%}
......
{% from 'oscar/macros.html' import result_footer, result_footer_rtl with context %}
<div class="panel panel-default">
<table class="table table-responsive table-bordered table-condensed">
{% for key, value in result.items() %}
{% if key in ['engine', 'engines', 'template', 'score', 'category', 'positions'] %}
{% continue %}
{% endif %}
<tr>
<td><b>{{ key|upper }}</b>: {{ value }}</td>
</tr>
{% endfor %}
</table>
{% if rtl %}
{{ result_footer_rtl(result) }}
{% else %}
{{ result_footer(result) }}
{% endif %}
</div>
<table>
{% for key, value in result.items() %}
{% if key in ['engine', 'engines', 'template', 'score', 'category', 'positions'] %}
{% continue %}
{% endif %}
<tr>
<td><b>{{ key|upper }}</b>: {{ value }}</td>
</tr>
{% endfor %}
</table>
<div class="engines">{% for engine in result.engines %}<span>{{ engine }}</span>{% endfor %}</div>{{- '' -}}
......@@ -308,14 +308,15 @@ def int_or_zero(num):
def is_valid_lang(lang):
is_abbr = (len(lang) == 2)
lang = lang.lower().decode('utf-8')
if is_abbr:
for l in language_codes:
if l[0][:2] == lang.lower():
if l[0][:2] == lang:
return (True, l[0][:2], l[3].lower())
return False
else:
for l in language_codes:
if l[1].lower() == lang.lower():
if l[1].lower() == lang or l[3].lower() == lang:
return (True, l[0][:2], l[3].lower())
return False
......@@ -434,3 +435,18 @@ def ecma_unescape(s):
# "%20" becomes " ", "%F3" becomes "ó"
s = ecma_unescape2_re.sub(lambda e: unichr(int(e.group(1), 16)), s)
return s
def get_engine_from_settings(name):
"""Return engine configuration from settings.yml of a given engine name"""
if 'engines' not in settings:
return {}
for engine in settings['engines']:
if 'name' not in engine:
continue
if name == engine['name']:
return engine
return {}
......@@ -124,6 +124,7 @@ app = Flask(
app.jinja_env.trim_blocks = True
app.jinja_env.lstrip_blocks = True
app.jinja_env.add_extension('jinja2.ext.loopcontrols')
app.secret_key = settings['server']['secret_key']
if not searx_debug \
......@@ -538,14 +539,16 @@ def index():
if output_format == 'html':
if 'content' in result and result['content']:
result['content'] = highlight_content(escape(result['content'][:1024]), search_query.query)
result['title'] = highlight_content(escape(result['title'] or u''), search_query.query)
if 'title' in result and result['title']:
result['title'] = highlight_content(escape(result['title'] or u''), search_query.query)
else:
if result.get('content'):
result['content'] = html_to_text(result['content']).strip()
# removing html content and whitespace duplications
result['title'] = ' '.join(html_to_text(result['title']).strip().split())
result['pretty_url'] = prettify_url(result['url'])
if 'url' in result:
result['pretty_url'] = prettify_url(result['url'])
# TODO, check if timezone is calculated right
if 'publishedDate' in result:
......
......@@ -22,74 +22,3 @@ class TestDeviantartEngine(SearxTestCase):
dicto['time_range'] = 'year'
params = deviantart.request(query, dicto)
self.assertEqual({}, params['url'])
def test_response(self):
self.assertRaises(AttributeError, deviantart.response, None)
self.assertRaises(AttributeError, deviantart.response, [])
self.assertRaises(AttributeError, deviantart.response, '')
self.assertRaises(AttributeError, deviantart.response, '[]')
response = mock.Mock(text='<html></html>')
self.assertEqual(deviantart.response(response), [])
response = mock.Mock(status_code=302)
self.assertEqual(deviantart.response(response), [])
html = """
<div id="page-1-results" class="page-results results-page-thumb torpedo-container">
<span class="thumb wide" href="http://amai911.deviantart.com/art/Horse-195212845"
data-super-full-width="900" data-super-full-height="600">
<a class="torpedo-thumb-link" href="https://url.of.image">
<img data-sigil="torpedo-img" src="https://url.of.thumbnail" />
</a>
<span class="info"><span class="title-wrap"><span class="title">Title of image</span></span>
</div>
"""
response = mock.Mock(text=html)
results = deviantart.response(response)
self.assertEqual(type(results), list)
self.assertEqual(len(results), 1)
self.assertEqual(results[0]['title'], 'Title of image')
self.assertEqual(results[0]['url'], 'https://url.of.image')
self.assertNotIn('content', results[0])
self.assertEqual(results[0]['thumbnail_src'], 'https://url.of.thumbnail')
html = """
<span class="tt-fh-tc" style="width: 202px;">
<span class="tt-bb" style="width: 202px;">
</span>
<span class="shadow">
<a class="thumb" href="http://url.of.result/2nd.part.of.url"
title="Behoimi BE Animation Test by test-0, Jan 4,
2010 in Digital Art &gt; Animation"> <i></i>
<img width="200" height="200" alt="Test"
src="http://url.of.thumbnail" data-src="http://th08.deviantart.net/test.jpg">
</a>
</span>
<!-- ^TTT -->
</span>
<span class="details">
<a href="http://test-0.deviantart.com/art/Test" class="t"
title="Behoimi BE Animation Test by test-0, Jan 4, 2010">
<span class="tt-fh-oe">Title of image</span> </a>
<small>
<span class="category">
<span class="age">
5 years ago
</span>
in <a title="Behoimi BE Animation Test by test-0, Jan 4, 2010"
href="http://www.deviantart.com/browse/all/digitalart/animation/">Animation</a>
</span>
<div class="commentcount">
<a href="http://test-0.deviantart.com/art/Test#comments">
<span class="iconcommentsstats"></span>9 Comments</a>
</div>
<a class="mlt-link" href="http://www.deviantart.com/morelikethis/149167425">
<span class="mlt-icon"></span> <span class="mlt-text">More Like This</span> </a>
</span>
</small> <!-- TTT$ -->
"""
response = mock.Mock(text=html)
results = deviantart.response(response)
self.assertEqual(type(results), list)
self.assertEqual(len(results), 0)
......@@ -14,88 +14,3 @@ class TestDiggEngine(SearxTestCase):
self.assertIn('url', params)
self.assertIn(query, params['url'])
self.assertIn('digg.com', params['url'])
def test_response(self):
self.assertRaises(AttributeError, digg.response, None)
self.assertRaises(AttributeError, digg.response, [])
self.assertRaises(AttributeError, digg.response, '')
self.assertRaises(AttributeError, digg.response, '[]')
response = mock.Mock(text='{}')
self.assertEqual(digg.response(response), [])
response = mock.Mock(text='{"data": []}')
self.assertEqual(digg.response(response), [])
json = """
{
"status": "ok",
"num": 10,
"next_position": 20,
"html": "<article itemscope itemtype=\\"http://schema.org/Article\\"
class=\\"story-container digg-story-el hentry entry story-1sRANah col-1\\"
data-content-id=\\"1sRANah\\" data-contenturl=\\"http://url.of.link\\"
data-position=\\"0\\" data-diggs=\\"24\\" data-tweets=\\"69\\"
data-digg-score=\\"1190\\"> <div class=\\"story-image story-image-thumb\\">
<a data-position=\\"0\\" data-content-id=\\"1sRANah\\"
class=\\"story-link\\" href=\\"http://www.thedailybeast.com/\\"
target=\\"_blank\\"><img class=\\"story-image-img\\"
src=\\"http://url.of.image.jpeg\\" width=\\"312\\" height=\\"170\\"
alt=\\"\\" /> </a> </div> <div class=\\"story-content\\"><header
class=\\"story-header\\"> <div itemprop=\\"alternativeHeadline\\"
class=\\"story-kicker\\" >Kicker</div> <h2 itemprop=\\"headline\\"
class=\\"story-title entry-title\\"><a class=\\"story-title-link story-link\\"
rel=\\"bookmark\\" itemprop=\\"url\\" href=\\"http://www.thedailybeast.com/\\"
target=\\"_blank\\">Title of article</h2> <div class=\\"story-meta\\">
<div class=\\"story-score \\">
<div class=\\"story-score-diggscore diggscore-1sRANah\\">1190</div>
<div class=\\"story-score-details\\"> <div class=\\"arrow\\"></div>
<ul class=\\"story-score-details-list\\"> <li
class=\\"story-score-detail story-score-diggs\\"><span
class=\\"label\\">Diggs:</span> <span class=\\"count diggs-1sRANah\\">24</span>
</li> <li class=\\"story-score-detail story-score-twitter\\"><span
class=\\"label\\">Tweets:</span> <span class=\\"count tweets-1sRANah\\">69</span>
</li> <li class=\\"story-score-detail story-score-facebook\\"><span
class=\\"label\\">Facebook Shares:</span> <span
class=\\"count fb_shares-1sRANah\\">1097</span></li> </ul> </div> </div>
<span class=\\"story-meta-item story-source\\"> <a
itemprop=\\"publisher copyrightHolder sourceOrganization provider\\"
class=\\"story-meta-item-link story-source-link\\"
href=\\"/source/thedailybeast.com\\">The Daily Beast </a> </span>
<span class=\\"story-meta-item story-tag first-tag\\"> <a
itemprop=\\"keywords\\" rel=\\"tag\\"
class=\\"story-meta-item-link story-tag-link\\" href=\\"/tag/news\\">News</a>
</span> <abbr class=\\"published story-meta-item story-timestamp\\"
title=\\"2014-10-18 14:53:45\\"> <time datetime=\\"2014-10-18 14:53:45\\">18 Oct 2014</time>
</abbr> </div> </header> </div> <ul class=\\"story-actions\\"> <li
class=\\"story-action story-action-digg btn-story-action-container\\">
<a class=\\"target digg-1sRANah\\" href=\\"#\\">Digg</a></li> <li
class=\\"story-action story-action-save btn-story-action-container\\">
<a class=\\"target save-1sRANah\\" href=\\"#\\">Save</a></li> <li
class=\\"story-action story-action-share\\"><a
class=\\"target share-facebook\\" href=\\"https://www.facebook.com/\\">Facebook</a></li>
<li class=\\"story-action story-action-share\\"><a class=\\"target share-twitter\\"
href=\\"https://twitter.com/\\">Twitter</a></li> </ul> </article>"
}
"""
json = json.replace('\r\n', '').replace('\n', '').replace('\r', '')
response = mock.Mock(text=json)
results = digg.response(response)
self.assertEqual(type(results), list)
self.assertEqual(len(results), 1)
self.assertEqual(results[0]['title'], 'Title of article')
self.assertEqual(results[0]['url'], 'http://url.of.link')
self.assertEqual(results[0]['thumbnail'], 'http://url.of.image.jpeg')
self.assertEqual(results[0]['content'], '')
json = """
{
"status": "error",
"num": 10,
"next_position": 20
}
"""
response = mock.Mock(text=json)
results = digg.response(response)
self.assertEqual(type(results), list)
self.assertEqual(len(results), 0)
......@@ -18,12 +18,9 @@ class TestStartpageEngine(SearxTestCase):
self.assertIn('data', params)
self.assertIn('query', params['data'])
self.assertIn(query, params['data']['query'])
self.assertIn('with_language', params['data'])
self.assertIn('lang_fr', params['data']['with_language'])
dicto['language'] = 'all'
params = startpage.request(query, dicto)
self.assertNotIn('with_language', params['data'])
def test_response(self):
self.assertRaises(AttributeError, startpage.response, None)
......@@ -35,33 +32,32 @@ class TestStartpageEngine(SearxTestCase):
self.assertEqual(startpage.response(response), [])
html = """
<li class="search-result search-item">
<h3>
<a href='http://this.should.be.the.link/' id='title_2' name='title_2' >
This should be the title
<div class="w-gl__result">
<a
class="w-gl__result-title"
href="http://this.should.be.the.link/"
data-onw="1"
rel="noopener noreferrer"
target="_blank">
<h3>This should be the title</h3>
</a>
<span id='title_stars_2' name='title_stars_2'> </span>
</h3>
<p class="search-item__body">
This should be the content.
</p>
<p>
<span class='url'>www.speed<b>test</b>.net/fr/
</span>
-
<A class="proxy" id="proxy_link" HREF="https://ixquick-proxy.com/do/spg/proxy?ep=&edata=&ek=&ekdata="
class='proxy'>
Navigation avec Ixquick Proxy
</A>
-
<A HREF="https://ixquick-proxy.com/do/spg/highlight.pl?l=francais&c=hf&cat=web&q=test&rl=NONE&rid=
&hlq=https://startpage.com/do/search&mtabp=-1&mtcmd=process_search&mtlanguage=francais&mtengine0=
&mtcat=web&u=http:%2F%2Fwww.speedtest.net%2Ffr%2F" class='proxy'>
Mis en surbrillance
</A>
</p>
</li>
"""
<div class="w-gl__result-second-line-container">
<div class="w-gl__result-url-container">
<a
class="w-gl__result-url"
href="http://this.should.be.the.link/"
rel="noopener noreferrer"
target="_blank">https://www.cnbc.com/2019/10/12/dj-zedd-banned-in-china-for-liking-a-south-park-tweet.html</a>
</div>
<a
class="w-gl__anonymous-view-url"
href="https://eu-browse.startpage.com/do/proxy?ep=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&amp;ek=4q58686o5047786n6343527259445247576p6o38&amp;ekdata=84abd523dc13cba5c65164d04d7d7263"
target="_blank">Anonymous View</a>
</div>
<p class="w-gl__description">This should be the content.</p>
</div>
""" # noqa
response = mock.Mock(text=html.encode('utf-8'))
results = startpage.response(response)
self.assertEqual(type(results), list)
......@@ -69,72 +65,3 @@ class TestStartpageEngine(SearxTestCase):
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.')
html = """
<li class="search-result search-item">
<h3>
<a href='http://www.google.com/aclk?sa=l&ai=C' id='title_2' name='title_2' >
This should be the title
</a>
<span id='title_stars_2' name='title_stars_2'> </span>
</h3>
<p class="search-item__body">
This should be the content.
</p>
<p>
<span class='url'>www.speed<b>test</b>.net/fr/
</span>
-
<A class="proxy" id="proxy_link" HREF="https://ixquick-proxy.com/do/spg/proxy?ep=&edata=&ek=&ekdata="
class='proxy'>
Navigation avec Ixquick Proxy
</A>
-
<A HREF="https://ixquick-proxy.com/do/spg/highlight.pl?l=francais&c=hf&cat=web&q=test&rl=NONE&rid=
&hlq=https://startpage.com/do/search&mtabp=-1&mtcmd=process_search&mtlanguage=francais&mtengine0=
&mtcat=web&u=http:%2F%2Fwww.speedtest.net%2Ffr%2F" class='proxy'>
Mis en surbrillance
</A>
</p>
</li>
<li class="search-result search-item">
<h3>
<span id='title_stars_2' name='title_stars_2'> </span>
</h3>
<p class="search-item__body">
This should be the content.
</p>
<p>
<span class='url'>www.speed<b>test</b>.net/fr/
</span>
</p>
</li>
<li class="search-result search-item">
<h3>
<a href='http://this.should.be.the.link/' id='title_2' name='title_2' >
This should be the title
</a>
<span id='title_stars_2' name='title_stars_2'> </span>
</h3>
<p>
<span class='url'>www.speed<b>test</b>.net/fr/
</span>
-
<A class="proxy" id="proxy_link" HREF="https://ixquick-proxy.com/do/spg/proxy?ep=&edata=&ek=&ekdata="
class='proxy'>
Navigation avec Ixquick Proxy
</A>
-
<A HREF="https://ixquick-proxy.com/do/spg/highlight.pl?l=francais&c=hf&cat=web&q=test&rl=NONE&rid=
&hlq=https://startpage.com/do/search&mtabp=-1&mtcmd=process_search&mtlanguage=francais&mtengine0=
&mtcat=web&u=http:%2F%2Fwww.speedtest.net%2Ffr%2F" class='proxy'>
Mis en surbrillance
</A>
</p>
</li>
"""
response = mock.Mock(text=html.encode('utf-8'))
results = startpage.response(response)
self.assertEqual(type(results), list)
self.assertEqual(len(results), 1)
self.assertEqual(results[0]['content'], '')
......@@ -12,46 +12,3 @@ class TestWww1xEngine(SearxTestCase):
self.assertTrue('url' in params)
self.assertTrue(query in params['url'])
self.assertTrue('1x.com' in params['url'])
def test_response(self):
self.assertRaises(AttributeError, www1x.response, None)
self.assertRaises(AttributeError, www1x.response, [])
self.assertRaises(AttributeError, www1x.response, '')
self.assertRaises(AttributeError, www1x.response, '[]')
response = mock.Mock(text='<html></html>')
self.assertEqual(www1x.response(response), [])
html = """
<?xml version="1.0" encoding="UTF-8"?><!DOCTYPE characters
[
<!ELEMENT characters (character*) >
<!ELEMENT character (#PCDATA ) >
<!ENTITY iexcl "&#161;" >
<!ENTITY cent "&#162;" >
<!ENTITY pound "&#163;" >
]
><root><searchresult><![CDATA[<table border="0" cellpadding="0" cellspacing="0" width="100%">
<tr>
<td style="min-width: 220px;" valign="top">
<div style="font-size: 30px; margin: 0px 0px 20px 0px;">Photos</div>
<div>
<a href="/photo/123456" class="dynamiclink">
<img border="0" class="searchresult" src="/images/user/testimage-123456.jpg" style="width: 125px; height: 120px;">
</a>
<a title="sjoerd lammers street photography" href="/member/sjoerdlammers" class="dynamiclink">
<img border="0" class="searchresult" src="/images/profile/60c48b394c677d2fa4d9e7d263aabf44-square.jpg">
</a>
</div>
</td>
</table>
]]></searchresult></root>
"""
response = mock.Mock(text=html)
results = www1x.response(response)
self.assertEqual(type(results), list)
self.assertEqual(len(results), 1)
self.assertEqual(results[0]['url'], 'https://1x.com/photo/123456')
self.assertEqual(results[0]['thumbnail_src'], 'https://1x.com/images/user/testimage-123456.jpg')
self.assertEqual(results[0]['content'], '')
self.assertEqual(results[0]['template'], 'images.html')
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment