diff options
Diffstat (limited to 'searx/utils.py')
| -rw-r--r-- | searx/utils.py | 4 |
1 files changed, 3 insertions, 1 deletions
diff --git a/searx/utils.py b/searx/utils.py index c47d3be17..c0afc94cb 100644 --- a/searx/utils.py +++ b/searx/utils.py @@ -115,10 +115,12 @@ class HTMLTextExtractor(HTMLParser): self.result.append(name) def get_text(self): - return u''.join(self.result) + return u''.join(self.result).strip() def html_to_text(html): + html = html.replace('\n', ' ') + html = ' '.join(html.split()) s = HTMLTextExtractor() s.feed(html) return s.get_text() |