From d827fc49a11b6f84bba3d006b54a70a6a05757fd Mon Sep 17 00:00:00 2001 From: a01200356 Date: Tue, 29 Dec 2015 21:11:49 -0600 Subject: [PATCH] Remove unnecessary code in wolframalpha_noapi engine The answer is scraped from a js function, so parsing the html tree doesn't achieve anything here. --- searx/engines/wolframalpha_noapi.py | 49 +++++++++++------------------ 1 file changed, 18 insertions(+), 31 deletions(-) diff --git a/searx/engines/wolframalpha_noapi.py b/searx/engines/wolframalpha_noapi.py index 1ce2aa1f..29600ca1 100644 --- a/searx/engines/wolframalpha_noapi.py +++ b/searx/engines/wolframalpha_noapi.py @@ -10,8 +10,6 @@ import re import json from urllib import urlencode -from lxml import html -from searx.engines.xpath import extract_text # search-url url = 'http://www.wolframalpha.com/' @@ -25,42 +23,31 @@ def request(query, params): return params -# tries to find answer under the pattern given -def extract_answer(script_list, pattern): - answer = None +# get response from search-request +def response(resp): + results = [] + + # the answer is inside a js function + # answer can be located in different 'pods', although by default it should be in pod_0200 + possible_locations = ['pod_0200\.push(.*)\n', + 'pod_0100\.push(.*)\n'] # get line that matches the pattern - for script in script_list: + for pattern in possible_locations: try: - line = re.search(pattern, script.text_content()).group(1) + line = re.search(pattern, resp.text).group(1) + break except AttributeError: continue - # extract answer from json - answer = line[line.find('{') : line.rfind('}')+1] - answer = json.loads(answer.encode('unicode-escape')) - answer = answer['stringified'].decode('unicode-escape') + if not line: + return results - return answer + # extract answer from json + answer = line[line.find('{') : line.rfind('}')+1] + answer = json.loads(answer.encode('unicode-escape')) + answer = answer['stringified'].decode('unicode-escape') - -# get response from search-request -def response(resp): - - dom = html.fromstring(resp.text) - - # the answer is inside a js script - scripts = dom.xpath('//script') - - results = [] - - # answer can be located in different 'pods', although by default it should be in pod_0200 - answer = extract_answer(scripts, 'pod_0200\.push(.*)\n') - if not answer: - answer = extract_answer(scripts, 'pod_0100\.push(.*)\n') - if answer: - results.append({'answer': answer}) - else: - results.append({'answer': answer}) + results.append({'answer': answer}) return results