fix(TokenClassification): Display characters between tokens words (#1418

) This PR fixes the display of characters between duplicate words or when the suffix matches the following word in the Token Classifier Closes #1414 Closes #1383 (cherry picked from commit 97ffc64)
argilla-io · Apr 25, 2022 · a08cd7b · a08cd7b
1 parent 35efceb
commit a08cd7b
Showing 1 changed file with 3 additions and 10 deletions.
diff --git a/frontend/components/token-classifier/results/RecordTokenClassification.vue b/frontend/components/token-classifier/results/RecordTokenClassification.vue
@@ -65,17 +65,10 @@ export default {
           const start = recordHasEmoji
             ? indexOf(this.record.text, token, startPosition)
             : this.record.text.indexOf(token, startPosition);
-          const nextStart = recordHasEmoji
-            ? indexOf(
-                this.record.text,
-                this.record.tokens[index + 1],
-                startPosition
-              )
-            : this.record.text.indexOf(
-                this.record.tokens[index + 1],
-                startPosition
-              );
           const end = start + (recordHasEmoji ? length(token) : token.length);
+          const nextStart = recordHasEmoji
+            ? indexOf(this.record.text, this.record.tokens[index + 1], end)
+            : this.record.text.indexOf(this.record.tokens[index + 1], end);
           const charsBetweenTokens = this.record.text.slice(end, nextStart);
           let highlighted = false;
           for (let highlight of searchKeywordsSpans) {