Translation: a hashtag in a script the reader can't read gets its translation in brackets after it — "#総合部絵チャ (…)" — the tag itself kept as the link it is. A post that is only such hashtags now counts as foreign (it was judged by its prose alone, of which it has none, so nothing ever offered to translate it), and ⋯ → Translate on a post with no words says so instead of "already in your language". The registrar's language detection reads the glossed tags too; lang-guess's scriptOf takes a minimum (a two-character tag has a script too).

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01Vps2H27rhbG9ZqseGfMrc3

1 files changed, 3 insertions, 3 deletions

diff --git a/src/lang-guess.js b/src/lang-guess.js
index f688cac..1d60c59 100644
--- a/src/lang-guess.js
+++ b/src/lang-guess.js
@@ -15,7 +15,7 @@ const SCRIPTS = [
   ['han', /[一-鿿㐀-䶿]/g],
   ['hangul', /[가-힯ᄀ-ᇿ]/g],
 ];
-const LANG_SCRIPT = {
+export const LANG_SCRIPT = {
   en: 'latin', es: 'latin', fr: 'latin', de: 'latin', pt: 'latin', id: 'latin', sw: 'latin', tr: 'latin', vi: 'latin',
   ru: 'cyrillic', ar: 'arabic', ur: 'arabic', hi: 'devanagari', mr: 'devanagari', bn: 'bengali', ta: 'tamil', te: 'telugu',
   ja: 'kana', ko: 'hangul', zh: 'han',
@@ -37,14 +37,14 @@ const STOPSETS = Object.fromEntries(Object.entries(STOP).map(([k, v]) => [k, new
 export const proseOf = (text) => String(text || '')
   .replace(/https?:\/\/\S+/g, ' ').replace(/nostr:[a-z0-9]+/gi, ' ').replace(/#[\p{L}\p{N}_]+/gu, ' ');
 
-export function scriptOf(text) {
+export function scriptOf(text, min = 3) {
   let best = null, bestN = 0, total = 0;
   for (const [name, re] of SCRIPTS) {
     const n = (text.match(re) || []).length;
     total += n;
     if (n > bestN) { best = name; bestN = n; }
   }
-  return total < 3 ? null : best;
+  return total < min ? null : best;
 }
 
 // A language code from the app's list, or null when it cannot tell.