def test_simple_query(self): query_text = 'the query' query = RawTextQuery(query_text, []) query.parse_query() self.assertEquals(query.getFullQuery(), query_text) self.assertEquals(len(query.query_parts), 1) self.assertEquals(len(query.languages), 0) self.assertFalse(query.specific)
def test_timeout_above1000(self): query_text = '<3500 the query' query = RawTextQuery(query_text, []) query.parse_query() self.assertEquals(query.getFullQuery(), query_text) self.assertEquals(len(query.query_parts), 3) self.assertEquals(query.timeout_limit, 3.5) self.assertFalse(query.specific)
def autocompleter(): """Return autocompleter results""" # set blocked engines disabled_engines = request.preferences.engines.get_disabled() # parse query if PY3: raw_text_query = RawTextQuery(request.form.get('q', b''), disabled_engines) else: raw_text_query = RawTextQuery( request.form.get('q', u'').encode('utf-8'), disabled_engines) raw_text_query.parse_query() # check if search query is set if not raw_text_query.getSearchQuery(): return '', 400 # run autocompleter completer = autocomplete_backends.get( request.preferences.get_value('autocomplete')) # parse searx specific autocompleter results like !bang raw_results = searx_bang(raw_text_query) # normal autocompletion results only appear if no inner results returned # and there is a query part besides the engine and language bangs if len(raw_results) == 0 and completer and ( len(raw_text_query.query_parts) > 1 or (len(raw_text_query.languages) == 0 and not raw_text_query.specific)): # get language from cookie language = request.preferences.get_value('language') if not language or language == 'all': language = 'en' else: language = language.split('-')[0] # run autocompletion raw_results.extend(completer(raw_text_query.getSearchQuery(), language)) # parse results (write :language and !engine back to result string) results = [] for result in raw_results: raw_text_query.changeSearchQuery(result) # add parsed result results.append(raw_text_query.getFullQuery()) # return autocompleter results if request.form.get('format') == 'x-suggestions': return Response(json.dumps([raw_text_query.query, results]), mimetype='application/json') return Response(json.dumps(results), mimetype='application/json')
def test_invalid_language_code(self): language = 'not_a_language' query_text = 'the query' full_query = ':' + language + ' ' + query_text query = RawTextQuery(full_query, []) query.parse_query() self.assertEquals(query.getFullQuery(), full_query) self.assertEquals(len(query.query_parts), 1) self.assertEquals(len(query.languages), 0) self.assertFalse(query.specific)
def test_unlisted_language_code(self): language = 'all' query_text = 'the query' full_query = ':' + language + ' ' + query_text query = RawTextQuery(full_query, []) query.parse_query() self.assertEquals(query.getFullQuery(), full_query) self.assertEquals(len(query.query_parts), 3) self.assertIn('all', query.languages) self.assertFalse(query.specific)
def test_timeout_invalid(self): # invalid number: it is not bang but it is part of the query query_text = '<xxx the query' query = RawTextQuery(query_text, []) query.parse_query() self.assertEquals(query.getFullQuery(), query_text) self.assertEquals(len(query.query_parts), 1) self.assertEquals(query.query_parts[0], query_text) self.assertEquals(query.timeout_limit, None) self.assertFalse(query.specific)
def autocompleter(): """Return autocompleter results""" # set blocked engines disabled_engines = request.preferences.engines.get_disabled() # parse query if PY3: raw_text_query = RawTextQuery(request.form.get('q', b''), disabled_engines) else: raw_text_query = RawTextQuery(request.form.get('q', u'').encode('utf-8'), disabled_engines) raw_text_query.parse_query() # check if search query is set if not raw_text_query.getSearchQuery(): return '', 400 # run autocompleter completer = autocomplete_backends.get(request.preferences.get_value('autocomplete')) # parse searx specific autocompleter results like !bang raw_results = searx_bang(raw_text_query) # normal autocompletion results only appear if max 3 inner results returned if len(raw_results) <= 3 and completer: # get language from cookie language = request.preferences.get_value('language') if not language or language == 'all': language = 'en' else: language = language.split('-')[0] # run autocompletion raw_results.extend(completer(raw_text_query.getSearchQuery(), language)) # parse results (write :language and !engine back to result string) results = [] for result in raw_results: raw_text_query.changeSearchQuery(result) # add parsed result results.append(raw_text_query.getFullQuery()) # return autocompleter results if request.form.get('format') == 'x-suggestions': return Response(json.dumps([raw_text_query.query, results]), mimetype='application/json') return Response(json.dumps(results), mimetype='application/json')
def autocompleter(): """Return autocompleter results""" # set blocked engines disabled_engines = request.preferences.engines.get_disabled() # parse query raw_text_query = RawTextQuery(request.form.get("q", "").encode("utf-8"), disabled_engines) raw_text_query.parse_query() # check if search query is set if not raw_text_query.getSearchQuery(): return "", 400 # run autocompleter completer = autocomplete_backends.get(request.preferences.get_value("autocomplete")) # parse searx specific autocompleter results like !bang raw_results = searx_bang(raw_text_query) # normal autocompletion results only appear if max 3 inner results returned if len(raw_results) <= 3 and completer: # get language from cookie language = request.preferences.get_value("language") if not language or language == "all": language = "en" else: language = language.split("_")[0] # run autocompletion raw_results.extend(completer(raw_text_query.getSearchQuery(), language)) # parse results (write :language and !engine back to result string) results = [] for result in raw_results: raw_text_query.changeSearchQuery(result) # add parsed result results.append(raw_text_query.getFullQuery()) # return autocompleter results if request.form.get("format") == "x-suggestions": return Response(json.dumps([raw_text_query.query, results]), mimetype="application/json") return Response(json.dumps(results), mimetype="application/json")
def get_search_query_from_webapp(preferences, form): # no text for the query ? if not form.get('q'): raise SearxParameterException('q', '') # set blocked engines disabled_engines = preferences.engines.get_disabled() # parse query, if tags are set, which change # the serch engine or search-language raw_text_query = RawTextQuery(form['q'], disabled_engines) raw_text_query.parse_query() # set query query = raw_text_query.getSearchQuery() # get and check page number pageno_param = form.get('pageno', '1') if not pageno_param.isdigit() or int(pageno_param) < 1: raise SearxParameterException('pageno', pageno_param) query_pageno = int(pageno_param) # get language # set specific language if set on request, query or preferences # TODO support search with multible languages if len(raw_text_query.languages): query_lang = raw_text_query.languages[-1] elif 'language' in form: query_lang = form.get('language') else: query_lang = preferences.get_value('language') # provides backwards compatibility for requests using old language default if query_lang == 'all': query_lang = settings['search']['language'] # check language if not VALID_LANGUAGE_CODE.match(query_lang): raise SearxParameterException('language', query_lang) # get safesearch if 'safesearch' in form: query_safesearch = form.get('safesearch') # first check safesearch if not query_safesearch.isdigit(): raise SearxParameterException('safesearch', query_safesearch) query_safesearch = int(query_safesearch) else: query_safesearch = preferences.get_value('safesearch') # safesearch : second check if query_safesearch < 0 or query_safesearch > 2: raise SearxParameterException('safesearch', query_safesearch) # get time_range query_time_range = form.get('time_range') # check time_range if query_time_range not in ('None', None, '', 'day', 'week', 'month', 'year'): raise SearxParameterException('time_range', query_time_range) # query_engines query_engines = raw_text_query.engines # query_categories query_categories = [] # if engines are calculated from query, # set categories by using that informations if query_engines and raw_text_query.specific: query_categories = list(set(engine['category'] for engine in query_engines)) # otherwise, using defined categories to # calculate which engines should be used else: # set categories/engines load_default_categories = True for pd_name, pd in form.items(): if pd_name == 'categories': query_categories.extend(categ for categ in map(unicode.strip, pd.split(',')) if categ in categories) elif pd_name == 'engines': pd_engines = [{'category': engines[engine].categories[0], 'name': engine} for engine in map(unicode.strip, pd.split(',')) if engine in engines] if pd_engines: query_engines.extend(pd_engines) load_default_categories = False elif pd_name.startswith('category_'): category = pd_name[9:] # if category is not found in list, skip if category not in categories: continue if pd != 'off': # add category to list query_categories.append(category) elif category in query_categories: # remove category from list if property is set to 'off' query_categories.remove(category) if not load_default_categories: if not query_categories: query_categories = list(set(engine['category'] for engine in query_engines)) else: # if no category is specified for this search, # using user-defined default-configuration which # (is stored in cookie) if not query_categories: cookie_categories = preferences.get_value('categories') for ccateg in cookie_categories: if ccateg in categories: query_categories.append(ccateg) # if still no category is specified, using general # as default-category if not query_categories: query_categories = ['general'] # using all engines for that search, which are # declared under the specific categories for categ in query_categories: query_engines.extend({'category': categ, 'name': engine.name} for engine in categories[categ] if (engine.name, categ) not in disabled_engines) return SearchQuery(query, query_engines, query_categories, query_lang, query_safesearch, query_pageno, query_time_range)
def get_search_query_from_webapp(preferences, form): query = None query_engines = [] query_categories = [] query_paging = False query_pageno = 1 query_lang = 'all' query_time_range = None # set blocked engines disabled_engines = preferences.engines.get_disabled() # set specific language if set query_lang = preferences.get_value('language') # safesearch query_safesearch = preferences.get_value('safesearch') # TODO better exceptions if not form.get('q'): raise Exception('noquery') # set pagenumber pageno_param = form.get('pageno', '1') if not pageno_param.isdigit() or int(pageno_param) < 1: pageno_param = 1 query_pageno = int(pageno_param) # parse query, if tags are set, which change # the serch engine or search-language raw_text_query = RawTextQuery(form['q'], disabled_engines) raw_text_query.parse_query() # set query query = raw_text_query.getSearchQuery() # get last selected language in query, if possible # TODO support search with multible languages if len(raw_text_query.languages): query_lang = raw_text_query.languages[-1] query_time_range = form.get('time_range') query_engines = raw_text_query.engines # if engines are calculated from query, # set categories by using that informations if query_engines and raw_text_query.specific: query_categories = list(set(engine['category'] for engine in query_engines)) # otherwise, using defined categories to # calculate which engines should be used else: # set categories/engines load_default_categories = True for pd_name, pd in form.items(): if pd_name == 'categories': query_categories.extend(categ for categ in map(unicode.strip, pd.split(',')) if categ in categories) elif pd_name == 'engines': pd_engines = [{'category': engines[engine].categories[0], 'name': engine} for engine in map(unicode.strip, pd.split(',')) if engine in engines] if pd_engines: query_engines.extend(pd_engines) load_default_categories = False elif pd_name.startswith('category_'): category = pd_name[9:] # if category is not found in list, skip if category not in categories: continue if pd != 'off': # add category to list query_categories.append(category) elif category in query_categories: # remove category from list if property is set to 'off' query_categories.remove(category) if not load_default_categories: if not query_categories: query_categories = list(set(engine['category'] for engine in engines)) else: # if no category is specified for this search, # using user-defined default-configuration which # (is stored in cookie) if not query_categories: cookie_categories = preferences.get_value('categories') for ccateg in cookie_categories: if ccateg in categories: query_categories.append(ccateg) # if still no category is specified, using general # as default-category if not query_categories: query_categories = ['general'] # using all engines for that search, which are # declared under the specific categories for categ in query_categories: query_engines.extend({'category': categ, 'name': engine.name} for engine in categories[categ] if (engine.name, categ) not in disabled_engines) return SearchQuery(query, query_engines, query_categories, query_lang, query_safesearch, query_pageno, query_time_range)
def get_search_query_from_webapp(preferences, form): query = None query_engines = [] query_categories = [] query_paging = False query_pageno = 1 query_lang = 'all' query_time_range = None # set blocked engines disabled_engines = preferences.engines.get_disabled() # set specific language if set query_lang = preferences.get_value('language') # safesearch query_safesearch = preferences.get_value('safesearch') # TODO better exceptions if not form.get('q'): raise Exception('noquery') # set pagenumber pageno_param = form.get('pageno', '1') if not pageno_param.isdigit() or int(pageno_param) < 1: pageno_param = 1 query_pageno = int(pageno_param) # parse query, if tags are set, which change # the serch engine or search-language raw_text_query = RawTextQuery(form['q'], disabled_engines) raw_text_query.parse_query() # set query query = raw_text_query.getSearchQuery() # get last selected language in query, if possible # TODO support search with multible languages if len(raw_text_query.languages): query_lang = raw_text_query.languages[-1] query_time_range = form.get('time_range') query_engines = raw_text_query.engines # if engines are calculated from query, # set categories by using that informations if query_engines and raw_text_query.specific: query_categories = list( set(engine['category'] for engine in query_engines)) # otherwise, using defined categories to # calculate which engines should be used else: # set categories/engines load_default_categories = True for pd_name, pd in form.items(): if pd_name == 'categories': query_categories.extend( categ for categ in map(unicode.strip, pd.split(',')) if categ in categories) elif pd_name == 'engines': pd_engines = [{ 'category': engines[engine].categories[0], 'name': engine } for engine in map(unicode.strip, pd.split(',')) if engine in engines] if pd_engines: query_engines.extend(pd_engines) load_default_categories = False elif pd_name.startswith('category_'): category = pd_name[9:] # if category is not found in list, skip if category not in categories: continue if pd != 'off': # add category to list query_categories.append(category) elif category in query_categories: # remove category from list if property is set to 'off' query_categories.remove(category) if not load_default_categories: if not query_categories: query_categories = list( set(engine['category'] for engine in engines)) else: # if no category is specified for this search, # using user-defined default-configuration which # (is stored in cookie) if not query_categories: cookie_categories = preferences.get_value('categories') for ccateg in cookie_categories: if ccateg in categories: query_categories.append(ccateg) # if still no category is specified, using general # as default-category if not query_categories: query_categories = ['general'] # using all engines for that search, which are # declared under the specific categories for categ in query_categories: query_engines.extend({ 'category': categ, 'name': engine.name } for engine in categories[categ] if (engine.name, categ) not in disabled_engines) return SearchQuery(query, query_engines, query_categories, query_lang, query_safesearch, query_pageno, query_time_range)
def get_search_query_from_webapp(preferences, form): # no text for the query ? if not form.get('q'): raise SearxParameterException('q', '') # set blocked engines disabled_engines = preferences.engines.get_disabled() # parse query, if tags are set, which change # the serch engine or search-language raw_text_query = RawTextQuery(form['q'], disabled_engines) raw_text_query.parse_query() # set query query = raw_text_query.getSearchQuery() # get and check page number pageno_param = form.get('pageno', '1') if not pageno_param.isdigit() or int(pageno_param) < 1: raise SearxParameterException('pageno', pageno_param) query_pageno = int(pageno_param) # get language # set specific language if set on request, query or preferences # TODO support search with multible languages if len(raw_text_query.languages): query_lang = raw_text_query.languages[-1] elif 'language' in form: query_lang = form.get('language') else: query_lang = preferences.get_value('language') # check language if not VALID_LANGUAGE_CODE.match(query_lang): raise SearxParameterException('language', query_lang) # get safesearch if 'safesearch' in form: query_safesearch = form.get('safesearch') # first check safesearch if not query_safesearch.isdigit(): raise SearxParameterException('safesearch', query_safesearch) query_safesearch = int(query_safesearch) else: query_safesearch = preferences.get_value('safesearch') # safesearch : second check if query_safesearch < 0 or query_safesearch > 2: raise SearxParameterException('safesearch', query_safesearch) # get time_range query_time_range = form.get('time_range') # check time_range if query_time_range not in ('None', None, '', 'day', 'week', 'month', 'year'): raise SearxParameterException('time_range', query_time_range) # query_engines query_engines = raw_text_query.engines # query_categories query_categories = [] # if engines are calculated from query, # set categories by using that informations if query_engines and raw_text_query.specific: query_categories = list(set(engine['category'] for engine in query_engines)) # otherwise, using defined categories to # calculate which engines should be used else: # set categories/engines load_default_categories = True for pd_name, pd in form.items(): if pd_name == 'categories': query_categories.extend(categ for categ in map(unicode.strip, pd.split(',')) if categ in categories) elif pd_name == 'engines': pd_engines = [{'category': engines[engine].categories[0], 'name': engine} for engine in map(unicode.strip, pd.split(',')) if engine in engines] if pd_engines: query_engines.extend(pd_engines) load_default_categories = False elif pd_name.startswith('category_'): category = pd_name[9:] # if category is not found in list, skip if category not in categories: continue if pd != 'off': # add category to list query_categories.append(category) elif category in query_categories: # remove category from list if property is set to 'off' query_categories.remove(category) if not load_default_categories: if not query_categories: query_categories = list(set(engine['category'] for engine in query_engines)) else: # if no category is specified for this search, # using user-defined default-configuration which # (is stored in cookie) if not query_categories: cookie_categories = preferences.get_value('categories') for ccateg in cookie_categories: if ccateg in categories: query_categories.append(ccateg) # if still no category is specified, using general # as default-category if not query_categories: query_categories = ['general'] # using all engines for that search, which are # declared under the specific categories for categ in query_categories: query_engines.extend({'category': categ, 'name': engine.name} for engine in categories[categ] if (engine.name, categ) not in disabled_engines) return SearchQuery(query, query_engines, query_categories, query_lang, query_safesearch, query_pageno, query_time_range)