From f5cea4d2ea314a78d44ef013c6dabda32ad96df1 Mon Sep 17 00:00:00 2001 From: Sasha Sorokin Date: Fri, 26 Jun 2020 03:45:01 +0700 Subject: [PATCH] Fix Thai being skipped from language detection (#13989) Thai does not separate words by spaces, so I figured out it should be in 'reliable characters regexp' that denotes languages that do the same. Related #13891. --- app/lib/language_detector.rb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/app/lib/language_detector.rb b/app/lib/language_detector.rb index 05a06726d..2cc8ac615 100644 --- a/app/lib/language_detector.rb +++ b/app/lib/language_detector.rb @@ -4,7 +4,7 @@ class LanguageDetector include Singleton WORDS_THRESHOLD = 4 - RELIABLE_CHARACTERS_RE = /[\p{Hebrew}\p{Arabic}\p{Syriac}\p{Thaana}\p{Nko}\p{Han}\p{Katakana}\p{Hiragana}\p{Hangul}]+/m + RELIABLE_CHARACTERS_RE = /[\p{Hebrew}\p{Arabic}\p{Syriac}\p{Thaana}\p{Nko}\p{Han}\p{Katakana}\p{Hiragana}\p{Hangul}\p{Thai}]+/m def initialize @identifier = CLD3::NNetLanguageIdentifier.new(1, 2048)