Fix hashtag matching by replacing negative lookbehind with positive lookbehind (#37684)

This commit is contained in:
Claire
2026-02-17 11:27:36 +01:00
committed by GitHub
parent 0a6412faf9
commit 4a6d17ad7b
2 changed files with 5 additions and 1 deletions

View File

@@ -41,7 +41,7 @@ class Tag < ApplicationRecord
HASHTAG_LAST_SEQUENCE = '([[:word:]_]*[[:alpha:]][[:word:]_]*)'
HASHTAG_NAME_PAT = "#{HASHTAG_FIRST_SEQUENCE}|#{HASHTAG_LAST_SEQUENCE}".freeze
HASHTAG_RE = %r{(?<![=/)\p{Alnum}])[#](#{HASHTAG_NAME_PAT})}
HASHTAG_RE = /(?<=^|\s)[#](#{HASHTAG_NAME_PAT})/
HASHTAG_NAME_RE = /\A(#{HASHTAG_NAME_PAT})\z/i
HASHTAG_INVALID_CHARS_RE = /[^[:alnum:]\u0E47-\u0E4E#{HASHTAG_SEPARATORS}]/

View File

@@ -85,6 +85,10 @@ RSpec.describe Tag do
expect(subject.match('https://en.wikipedia.org/wiki/Ghostbusters_(song)?foo=#Lawsuit')).to be_nil
end
it 'does not match URLs with hashtag-like anchors after a dot' do
expect(subject.match('https://en.wikipedia.org/wiki/Google_LLC_v._Oracle_America,_Inc.#Decision')).to be_nil
end
it 'matches #' do
expect(subject.match('this is #').to_s).to eq '#'
end