logo

searx

Unnamed repository; edit this file 'description' to name the repository.
commit: 0540ea9ee21a2465b417249af4abed9b50917603
parent: e56d7a58c7e6680f3fbcf6dc3c691a08c04ee674
Author: asciimoo <asciimoo@gmail.com>
Date:   Fri,  7 Feb 2014 01:19:07 +0100

[enh][mod] code refactor ++ search engine filtering

Diffstat:

searx/search.py | 94+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
searx/tests/test_webapp.py | 8++++----
searx/webapp.py | 131+++++++++++++++++++++----------------------------------------------------------
3 files changed, 133 insertions(+), 100 deletions(-)

diff --git a/searx/search.py b/searx/search.py @@ -0,0 +1,94 @@ +from searx.engines import ( + categories, engines, engine_shortcuts +) +from searx.languages import language_codes + + +class Search(object): + + """Search information container""" + + def __init__(self, request): + super(Search, self).__init__() + self.query = None + self.engines = [] + self.categories = [] + query_engines = [] + self.paging = False + self.pageno = 1 + self.lang = 'all' + if request.cookies.get('blocked_engines'): + self.blocked_engines = request.cookies['blocked_engines'].split(',') + else: + self.blocked_engines = [] + self.results = [] + self.suggestions = [] + self.request_data = {} + + if request.cookies.get('language')\ + and request.cookies['language'] in (x[0] for x in language_codes): + self.lang = request.cookies['language'] + + if request.method == 'POST': + self.request_data = request.form + else: + self.request_data = request.args + + # TODO better exceptions + if not self.request_data.get('q'): + raise Exception('noquery') + + self.query = self.request_data['q'] + + pageno_param = self.request_data.get('pageno', '1') + if not pageno_param.isdigit() or int(pageno_param) < 1: + raise Exception('wrong pagenumber') + + self.pageno = int(pageno_param) + + query_parts = self.query.split() + if query_parts[0].startswith('!'): + prefix = query_parts[0][1:].replace('_', ' ') + if prefix in engine_shortcuts\ + and not engine_shortcuts[prefix] in self.blocked_engines: + self.engines.append({'category': 'none', + 'name': engine_shortcuts[prefix]}) + elif prefix in engines\ + and not prefix in self.blocked_engines: + self.engines.append({'category': 'none', + 'name': prefix}) + elif prefix in categories: + self.engines.extend({'category': prefix, + 'name': engine.name} + for engine in categories[prefix] + if not engine in self.blocked_engines) + + if len(query_engines): + self.query = self.query.replace(query_parts[0], '', 1).strip() + + self.categories = [] + + if len(self.engines): + self.categories = list(set(engine['category'] + for engine in self.engines)) + else: + for pd_name, pd in self.request_data.items(): + if pd_name.startswith('category_'): + category = pd_name[9:] + if not category in categories: + continue + self.categories.append(category) + if not len(self.categories): + cookie_categories = request.cookies.get('categories', '') + cookie_categories = cookie_categories.split(',') + for ccateg in cookie_categories: + if ccateg in categories: + self.categories.append(ccateg) + if not len(self.categories): + self.categories = ['general'] + + for categ in self.categories: + self.engines.extend({'category': categ, + 'name': x.name} + for x in categories[categ] + if not x.name in self.blocked_engines) diff --git a/searx/tests/test_webapp.py b/searx/tests/test_webapp.py @@ -39,7 +39,7 @@ class ViewsTestCase(SearxTestCase): self.assertEqual(result.status_code, 200) self.assertIn('<div class="title"><h1>searx</h1></div>', result.data) - @patch('searx.webapp.search') + @patch('searx.webapp.do_search') def test_index_html(self, search): search.return_value = ( self.test_results, @@ -55,7 +55,7 @@ class ViewsTestCase(SearxTestCase): result.data ) - @patch('searx.webapp.search') + @patch('searx.webapp.do_search') def test_index_json(self, search): search.return_value = ( self.test_results, @@ -71,7 +71,7 @@ class ViewsTestCase(SearxTestCase): self.assertEqual( result_dict['results'][0]['url'], 'http://first.test.xyz') - @patch('searx.webapp.search') + @patch('searx.webapp.do_search') def test_index_csv(self, search): search.return_value = ( self.test_results, @@ -86,7 +86,7 @@ class ViewsTestCase(SearxTestCase): result.data ) - @patch('searx.webapp.search') + @patch('searx.webapp.do_search') def test_index_rss(self, search): search.return_value = ( self.test_results, diff --git a/searx/webapp.py b/searx/webapp.py @@ -28,10 +28,12 @@ from flask import ( from flask.ext.babel import Babel from searx import settings, searx_dir from searx.engines import ( - search, categories, engines, get_engines_stats, engine_shortcuts + search as do_search, categories, engines, get_engines_stats, + engine_shortcuts ) from searx.utils import UnicodeWriter, highlight_content, html_to_text from searx.languages import language_codes +from searx.search import Search app = Flask( @@ -94,95 +96,32 @@ def render(template_name, **kwargs): return render_template(template_name, **kwargs) -def parse_query(query): - query_engines = [] - query_parts = query.split() - - if query_parts[0].startswith('!'): - prefix = query_parts[0][1:].replace('_', ' ') - if prefix in engine_shortcuts: - query_engines.append({'category': 'none', - 'name': engine_shortcuts[prefix]}) - elif prefix in engines: - query_engines.append({'category': 'none', - 'name': prefix}) - elif prefix in categories: - query_engines.extend({'category': prefix, - 'name': engine.name} - for engine in categories[prefix]) - - if len(query_engines): - query = query.replace(query_parts[0], '', 1).strip() - return query, query_engines - - @app.route('/', methods=['GET', 'POST']) def index(): """Render index page. Supported outputs: html, json, csv, rss. """ - paging = False - lang = 'all' - - if request.cookies.get('language')\ - and request.cookies['language'] in (x[0] for x in language_codes): - lang = request.cookies['language'] - - if request.method == 'POST': - request_data = request.form - else: - request_data = request.args - if not request_data.get('q'): - return render('index.html') - pageno_param = request_data.get('pageno', '1') - if not pageno_param.isdigit() or int(pageno_param) < 1: + try: + search = Search(request) + except: return render('index.html') - pageno = int(pageno_param) - - selected_categories = [] - - query, selected_engines = parse_query(request_data['q'].encode('utf-8')) - - if len(selected_engines): - selected_categories = list(set(engine['category'] - for engine in selected_engines)) - else: - for pd_name, pd in request_data.items(): - if pd_name.startswith('category_'): - category = pd_name[9:] - if not category in categories: - continue - selected_categories.append(category) - if not len(selected_categories): - cookie_categories = request.cookies.get('categories', '') - cookie_categories = cookie_categories.split(',') - for ccateg in cookie_categories: - if ccateg in categories: - selected_categories.append(ccateg) - if not len(selected_categories): - selected_categories = ['general'] - - for categ in selected_categories: - selected_engines.extend({'category': categ, - 'name': x.name} - for x in categories[categ]) - - results, suggestions = search(query, - request, - selected_engines, - pageno, - lang) - - for result in results: - if not paging and engines[result['engine']].paging: - paging = True - if request_data.get('format', 'html') == 'html': + # TODO moar refactor - do_search integration into Search class + search.results, search.suggestions = do_search(search.query, + request, + search.engines, + search.pageno, + search.lang) + + for result in search.results: + if not search.paging and engines[result['engine']].paging: + search.paging = True + if search.request_data.get('format', 'html') == 'html': if 'content' in result: - result['content'] = highlight_content(result['content'], query) - result['title'] = highlight_content(result['title'], query) + result['content'] = highlight_content(result['content'], search.query) + result['title'] = highlight_content(result['title'], search.query) else: if 'content' in result: result['content'] = html_to_text(result['content']).strip() @@ -199,40 +138,40 @@ def index(): if engine in favicons: result['favicon'] = engine - if request_data.get('format') == 'json': - return Response(json.dumps({'query': query, 'results': results}), + if search.request_data.get('format') == 'json': + return Response(json.dumps({'query': search.query, 'results': search.results}), mimetype='application/json') - elif request_data.get('format') == 'csv': + elif search.request_data.get('format') == 'csv': csv = UnicodeWriter(cStringIO.StringIO()) keys = ('title', 'url', 'content', 'host', 'engine', 'score') - if len(results): + if len(search.results): csv.writerow(keys) - for row in results: + for row in search.results: row['host'] = row['parsed_url'].netloc csv.writerow([row.get(key, '') for key in keys]) csv.stream.seek(0) response = Response(csv.stream.read(), mimetype='application/csv') - content_disp = 'attachment;Filename=searx_-_{0}.csv'.format(query) + content_disp = 'attachment;Filename=searx_-_{0}.csv'.format(search.query) response.headers.add('Content-Disposition', content_disp) return response - elif request_data.get('format') == 'rss': + elif search.request_data.get('format') == 'rss': response_rss = render( 'opensearch_response_rss.xml', - results=results, - q=request_data['q'], - number_of_results=len(results), + results=search.results, + q=search.request_data['q'], + number_of_results=len(search.results), base_url=get_base_url() ) return Response(response_rss, mimetype='text/xml') return render( 'results.html', - results=results, - q=request_data['q'], - selected_categories=selected_categories, - paging=paging, - pageno=pageno, - suggestions=suggestions + results=search.results, + q=search.request_data['q'], + selected_categories=search.categories, + paging=search.paging, + pageno=search.pageno, + suggestions=search.suggestions )