diff options
Diffstat (limited to 'app/lib')
-rw-r--r-- | app/lib/activitypub/activity/create.rb | 2 | ||||
-rw-r--r-- | app/lib/extractor.rb | 3 | ||||
-rw-r--r-- | app/lib/language_detector.rb | 31 |
3 files changed, 28 insertions, 8 deletions
diff --git a/app/lib/activitypub/activity/create.rb b/app/lib/activitypub/activity/create.rb index 376684c00..66e4f7c5e 100644 --- a/app/lib/activitypub/activity/create.rb +++ b/app/lib/activitypub/activity/create.rb @@ -173,7 +173,7 @@ class ActivityPub::Activity::Create < ActivityPub::Activity end def language_from_content - return nil unless language_map? + return LanguageDetector.instance.detect(text_from_content, @account) unless language_map? @object['contentMap'].keys.first end diff --git a/app/lib/extractor.rb b/app/lib/extractor.rb index 957364293..738ec89a0 100644 --- a/app/lib/extractor.rb +++ b/app/lib/extractor.rb @@ -5,7 +5,8 @@ module Extractor module_function - def extract_mentions_or_lists_with_indices(text) # :yields: username, list_slug, start, end + # :yields: username, list_slug, start, end + def extract_mentions_or_lists_with_indices(text) return [] unless text =~ Twitter::Regex[:at_signs] possible_entries = [] diff --git a/app/lib/language_detector.rb b/app/lib/language_detector.rb index a42460e10..c6f52f0c7 100644 --- a/app/lib/language_detector.rb +++ b/app/lib/language_detector.rb @@ -38,12 +38,31 @@ class LanguageDetector end def simplify_text(text) - text.dup.tap do |new_text| - new_text.gsub!(FetchLinkCardService::URL_PATTERN, '') - new_text.gsub!(Account::MENTION_RE, '') - new_text.gsub!(Tag::HASHTAG_RE, '') - new_text.gsub!(/\s+/, ' ') - end + new_text = remove_html(text) + new_text.gsub!(FetchLinkCardService::URL_PATTERN, '') + new_text.gsub!(Account::MENTION_RE, '') + new_text.gsub!(Tag::HASHTAG_RE, '') + new_text.gsub!(/:#{CustomEmoji::SHORTCODE_RE_FRAGMENT}:/, '') + new_text.gsub!(/\s+/, ' ') + new_text + end + + def new_scrubber + scrubber = Rails::Html::PermitScrubber.new + scrubber.tags = %w(br p) + scrubber + end + + def scrubber + @scrubber ||= new_scrubber + end + + def remove_html(text) + text = Loofah.fragment(text).scrub!(scrubber).to_s + text.gsub!('<br>', "\n") + text.gsub!('</p><p>', "\n\n") + text.gsub!(/(^<p>|<\/p>$)/, '') + text end def default_locale(account) |