admin/snakkimo-cmt
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: prevent pronouns/function words from being detected as vocabulary

Browse Source 
- fuzzyMatch threshold 0.6 → 0.85 (man/Mann 0.75 now fails)
- min token length 2 → 4 (excludes man, der, die, das, ich, wir...)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
admin
2026-05-28 21:53:59 +02:00
parent c595cbb448
commit 4d618bd5b3
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
src/pages/ContentCreation.jsx
View File

@@ -77,7 +77,7 @@ async function resolvePlaceholders(text, allObjects) {
// ─── Fuzzy word matching ────────────────────────────────────────────────────── // ─── Fuzzy word matching ──────────────────────────────────────────────────────
function fuzzyMatch(token, wordTitle, threshold = 0.6) { function fuzzyMatch(token, wordTitle, threshold = 0.85) {
const t = token.toLowerCase(); const t = token.toLowerCase();
const w = wordTitle.toLowerCase(); const w = wordTitle.toLowerCase();
if (!t.startsWith(w) && !w.startsWith(t)) return false; if (!t.startsWith(w) && !w.startsWith(t)) return false;
@@ -234,7 +234,7 @@ function PairForm({ objectId, allObjects, onPairSaved }) {
if (!allText.trim()) { setWordMap({}); return; } if (!allText.trim()) { setWordMap({}); return; }
const t = setTimeout(async () => { const t = setTimeout(async () => {
try { try {
const tokens = [...new Set(allText.split(/[\s.,!?;:()\[\]"']+/).filter(w => w.length >= 2))]; const tokens = [...new Set(allText.split(/[\s.,!?;:()\[\]"']+/).filter(w => w.length >= 4))];
if (!tokens.length) return; if (!tokens.length) return;
const results = await Promise.allSettled( const results = await Promise.allSettled(
tokens.map(async w => { tokens.map(async w => {
@@ -532,7 +532,7 @@ function EditPairForm({ pair, allObjects, onSaved, onCancel, onDeleted }) {
if (!allText.trim()) { setWordMap({}); return; } if (!allText.trim()) { setWordMap({}); return; }
const t = setTimeout(async () => { const t = setTimeout(async () => {
try { try {
const tokens = [...new Set(allText.split(/[\s.,!?;:()\[\]"']+/).filter(w => w.length >= 2))]; const tokens = [...new Set(allText.split(/[\s.,!?;:()\[\]"']+/).filter(w => w.length >= 4))];
if (!tokens.length) return; if (!tokens.length) return;
const results = await Promise.allSettled( const results = await Promise.allSettled(
tokens.map(async w => { tokens.map(async w => {