diff options
| -rw-r--r-- | AUTHORS.rst | 2 | ||||
| -rw-r--r-- | searx/engines/wikipedia.py | 9 | ||||
| -rw-r--r-- | searx/webutils.py | 5 | ||||
| -rw-r--r-- | tests/unit/test_webutils.py | 22 |
4 files changed, 34 insertions, 4 deletions
diff --git a/AUTHORS.rst b/AUTHORS.rst index 036ae0fe1..b44a10b55 100644 --- a/AUTHORS.rst +++ b/AUTHORS.rst @@ -154,6 +154,6 @@ generally made searx better: - @mrwormo - Xiaoyu WEI @xywei - @joshu9h - +- Daniel Hones diff --git a/searx/engines/wikipedia.py b/searx/engines/wikipedia.py index eff301145..c8e589e64 100644 --- a/searx/engines/wikipedia.py +++ b/searx/engines/wikipedia.py @@ -22,6 +22,7 @@ about = { # search-url search_url = 'https://{language}.wikipedia.org/api/rest_v1/page/summary/{title}' supported_languages_url = 'https://meta.wikimedia.org/wiki/List_of_Wikipedias' +language_variants = {"zh": ("zh-cn", "zh-hk", "zh-mo", "zh-my", "zh-sg", "zh-tw")} # set language in base_url @@ -37,8 +38,12 @@ def request(query, params): if query.islower(): query = query.title() + language = url_lang(params['language']) params['url'] = search_url.format(title=quote(query), - language=url_lang(params['language'])) + language=language) + + if params['language'].lower() in language_variants.get(language, []): + params['headers']['Accept-Language'] = params['language'].lower() params['headers']['User-Agent'] = searx_useragent() params['raise_for_httperror'] = False @@ -60,7 +65,7 @@ def response(resp): if api_result.get('type') != 'standard': return [] - title = api_result['title'] + title = api_result['displaytitle'] wikipedia_link = api_result['content_urls']['desktop']['page'] results.append({'url': wikipedia_link, 'title': title}) diff --git a/searx/webutils.py b/searx/webutils.py index 8be8fcecd..2464a097f 100644 --- a/searx/webutils.py +++ b/searx/webutils.py @@ -119,7 +119,10 @@ def highlight_content(content, query): else: regex_parts = [] for chunk in query.split(): - if len(chunk) == 1: + chunk = chunk.replace('"', '') + if len(chunk) == 0: + continue + elif len(chunk) == 1: regex_parts.append('\\W+{0}\\W+'.format(re.escape(chunk))) else: regex_parts.append('{0}'.format(re.escape(chunk))) diff --git a/tests/unit/test_webutils.py b/tests/unit/test_webutils.py index aa464688b..023374b04 100644 --- a/tests/unit/test_webutils.py +++ b/tests/unit/test_webutils.py @@ -34,6 +34,28 @@ class TestWebUtils(SearxTestCase): query = 'a test' self.assertEqual(webutils.highlight_content(content, query), content) + data = ( + ('" test "', + 'a test string', + 'a <span class="highlight">test</span> string'), + ('"a"', + 'this is a test string', + 'this is<span class="highlight"> a </span>test string'), + ('a test', + 'this is a test string that matches entire query', + 'this is <span class="highlight">a test</span> string that matches entire query'), + ('this a test', + 'this is a string to test.', + ('<span class="highlight">this</span> is<span class="highlight"> a </span>' + 'string to <span class="highlight">test</span>.')), + ('match this "exact phrase"', + 'this string contains the exact phrase we want to match', + ('<span class="highlight">this</span> string contains the <span class="highlight">exact</span>' + ' <span class="highlight">phrase</span> we want to <span class="highlight">match</span>')) + ) + for query, content, expected in data: + self.assertEqual(webutils.highlight_content(content, query), expected) + class TestUnicodeWriter(SearxTestCase): |