mirror of
https://github.com/mastodon/mastodon.git
synced 2024-08-20 21:08:15 -07:00
Improve RTL detection (#3682)
- Use plaintext - Strip out URLs - Strip out mentions - Strip out hashtags - Strip out whitespace from "overall" count - Consistent between JS and Ruby
This commit is contained in:
parent
4919b89ab8
commit
8015fd7600
5 changed files with 26 additions and 5 deletions
|
@ -47,11 +47,16 @@ module StreamEntriesHelper
|
|||
end
|
||||
end
|
||||
|
||||
def rtl_status?(status)
|
||||
status.local? ? rtl?(status.text) : rtl?(strip_tags(status.text))
|
||||
end
|
||||
|
||||
def rtl?(text)
|
||||
text = simplified_text(text)
|
||||
rtl_characters = /[\p{Hebrew}|\p{Arabic}|\p{Syriac}|\p{Thaana}|\p{Nko}]+/m.match(text)
|
||||
|
||||
if rtl_characters.present?
|
||||
total_size = text.strip.size.to_f
|
||||
total_size = text.size.to_f
|
||||
rtl_size(rtl_characters.to_a) / total_size > 0.3
|
||||
else
|
||||
false
|
||||
|
@ -60,6 +65,18 @@ module StreamEntriesHelper
|
|||
|
||||
private
|
||||
|
||||
def simplified_text(text)
|
||||
text.dup.tap do |new_text|
|
||||
URI.extract(new_text).each do |url|
|
||||
new_text.gsub!(url, '')
|
||||
end
|
||||
|
||||
new_text.gsub!(Account::MENTION_RE, '')
|
||||
new_text.gsub!(Tag::HASHTAG_RE, '')
|
||||
new_text.gsub!(/\s+/, '')
|
||||
end
|
||||
end
|
||||
|
||||
def rtl_size(characters)
|
||||
characters.reduce(0) { |acc, elem| acc + elem.size }.to_f
|
||||
end
|
||||
|
|
|
@ -100,7 +100,7 @@ class StatusContent extends React.PureComponent {
|
|||
const spoilerContent = { __html: emojify(escapeTextContentForBrowser(status.get('spoiler_text', ''))) };
|
||||
const directionStyle = { direction: 'ltr' };
|
||||
|
||||
if (isRtl(status.get('content'))) {
|
||||
if (isRtl(status.get('search_index'))) {
|
||||
directionStyle.direction = 'rtl';
|
||||
}
|
||||
|
||||
|
|
|
@ -17,11 +17,15 @@ export function isRtl(text) {
|
|||
return false;
|
||||
}
|
||||
|
||||
text = text.replace(/(?:^|[^\/\w])@([a-z0-9_]+(@[a-z0-9\.\-]+)?)/ig, '');
|
||||
text = text.replace(/(?:^|[^\/\w])#([\S]+)/ig, '');
|
||||
text = text.replace(/\s+/g, '');
|
||||
|
||||
const matches = text.match(rtlChars);
|
||||
|
||||
if (!matches) {
|
||||
return false;
|
||||
}
|
||||
|
||||
return matches.length / text.trim().length > 0.3;
|
||||
return matches.length / text.length > 0.3;
|
||||
};
|
||||
|
|
|
@ -12,7 +12,7 @@
|
|||
%p{ style: 'margin-bottom: 0' }<
|
||||
%span.p-summary> #{status.spoiler_text}
|
||||
%a.status__content__spoiler-link{ href: '#' }= t('statuses.show_more')
|
||||
.e-content{ lang: status.language, style: "display: #{status.spoiler_text? ? 'none' : 'block'}; direction: #{rtl?(status.content) ? 'rtl' : 'ltr'}" }= Formatter.instance.format(status)
|
||||
.e-content{ lang: status.language, style: "display: #{status.spoiler_text? ? 'none' : 'block'}; direction: #{rtl_status?(status) ? 'rtl' : 'ltr'}" }= Formatter.instance.format(status)
|
||||
|
||||
- unless status.media_attachments.empty?
|
||||
- if status.media_attachments.first.video?
|
||||
|
|
|
@ -18,7 +18,7 @@
|
|||
%p{ style: 'margin-bottom: 0' }<
|
||||
%span.p-summary> #{status.spoiler_text}
|
||||
%a.status__content__spoiler-link{ href: '#' }= t('statuses.show_more')
|
||||
.e-content{ lang: status.language, style: "display: #{status.spoiler_text? ? 'none' : 'block'}; direction: #{rtl?(status.content) ? 'rtl' : 'ltr'}" }= Formatter.instance.format(status)
|
||||
.e-content{ lang: status.language, style: "display: #{status.spoiler_text? ? 'none' : 'block'}; direction: #{rtl_status?(status) ? 'rtl' : 'ltr'}" }= Formatter.instance.format(status)
|
||||
|
||||
- unless status.media_attachments.empty?
|
||||
.status__attachments
|
||||
|
|
Loading…
Reference in a new issue