Change language detection to include hashtags as words (#11341)

This commit is contained in:
Eugen Rochko 2019-07-18 03:02:15 +02:00 committed by GitHub
parent 3a6fe657ba
commit 5bfe1e1f05
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 4 additions and 4 deletions

View File

@ -69,7 +69,7 @@ class LanguageDetector
new_text = remove_html(text) new_text = remove_html(text)
new_text.gsub!(FetchLinkCardService::URL_PATTERN, '') new_text.gsub!(FetchLinkCardService::URL_PATTERN, '')
new_text.gsub!(Account::MENTION_RE, '') new_text.gsub!(Account::MENTION_RE, '')
new_text.gsub!(Tag::HASHTAG_RE, '') new_text.gsub!(Tag::HASHTAG_RE) { |string| string.gsub(/[#_]/, '#' => '', '_' => ' ').gsub(/[a-z][A-Z]|[a-zA-Z][\d]/) { |s| s.insert(1, ' ') }.downcase }
new_text.gsub!(/:#{CustomEmoji::SHORTCODE_RE_FRAGMENT}:/, '') new_text.gsub!(/:#{CustomEmoji::SHORTCODE_RE_FRAGMENT}:/, '')
new_text.gsub!(/\s+/, ' ') new_text.gsub!(/\s+/, ' ')
new_text new_text

View File

@ -32,11 +32,11 @@ describe LanguageDetector do
expect(result).to eq 'Our website is and also' expect(result).to eq 'Our website is and also'
end end
it 'strips #hashtags from strings before detection' do it 'converts #hashtags back to normal text before detection' do
string = 'Hey look at all the #animals and #fish' string = 'Hey look at all the #animals and #FishAndChips'
result = described_class.instance.send(:prepare_text, string) result = described_class.instance.send(:prepare_text, string)
expect(result).to eq 'Hey look at all the and' expect(result).to eq 'Hey look at all the animals and fish and chips'
end end
end end