feat(VocabService): improve SRS pair validation and introduce new text analysis methods

- Refactored SRS pair validation logic by introducing the `_isTrainableSrsPair` method to enhance the criteria for acceptable learning and reference pairs.
- Added `_isInstructionLikeText` method to filter out instructional texts from SRS pairs, improving the quality of vocabulary training.
- Updated various methods to utilize the new validation logic, ensuring consistent handling of vocabulary entries across the service.
- Enhanced the `_courseHasDueSrsItems` and `_extractTrainerVocabsFromLessonDidactics` methods to improve data retrieval and filtering based on the new criteria.

backend/services/vocabService.js

@@ -42,7 +42,7 @@ export default class VocabService {
       .map((entry) => {
         const learning = String(entry?.learning || '').trim();
         const reference = String(entry?.reference || '').trim();
-        if (!learning || !reference || this._normalizeSrsText(learning) === this._normalizeSrsText(reference)) {
+        if (!this._isTrainableSrsPair({ learning, reference })) {
           return null;
         }
         const direction = String(entry?.direction || 'BOTH').toUpperCase();
@@ -61,6 +61,30 @@ export default class VocabService {
       .filter(Boolean);
   }
 
+  _isInstructionLikeText(value) {
+    const text = String(value || '').trim();
+    if (!text) {
+      return false;
+    }
+
+    const wordCount = text.split(/\s+/).filter(Boolean).length;
+    if (wordCount < 3) {
+      return false;
+    }
+
+    return /^(sage|sag|frage|frag|bitte|stelle|sprich|erzähle|beschreibe|bilde|wähle|ordne|übersetze|nenne|nenn|beginne|verwende|reagiere|kombiniere|spiele|löse|beantworte|ergänze|formuliere)\b/i.test(text);
+  }
+
+  _isTrainableSrsPair(entry) {
+    const learning = String(entry?.learning || '').trim();
+    const reference = String(entry?.reference || '').trim();
+    if (!learning || !reference || this._normalizeSrsText(learning) === this._normalizeSrsText(reference)) {
+      return false;
+    }
+
+    return !this._isInstructionLikeText(learning) && !this._isInstructionLikeText(reference);
+  }
+
   _calculateSrsSchedule(item, { correct, rating = null } = {}) {
     const now = new Date();
     const previousStage = Math.max(0, Number(item?.stage) || 0);
@@ -459,16 +483,17 @@ export default class VocabService {
   }
 
   async _courseHasDueSrsItems(userId, courseId) {
-    const due = await VocabSrsItem.count({
+    const dueItems = await VocabSrsItem.findAll({
       where: {
         userId,
         courseId: Number(courseId),
         nextDueAt: {
           [Op.lte]: new Date()
         }
-      }
+      },
+      attributes: ['learning', 'reference']
     });
-    return due > 0;
+    return dueItems.some((item) => this._isTrainableSrsPair(item));
   }
 
   async _getUserByHashedId(hashedUserId) {
@@ -713,7 +738,7 @@ export default class VocabService {
           const question = String(qData.question || qData.text || '');
 
           let match = question.match(/Wie sagt man ['"]([^'"]+)['"]/i);
-          if (match && match[1] && correctAnswer && match[1].trim() !== String(correctAnswer).trim()) {
+          if (match && this._isTrainableSrsPair({ learning: match[1], reference: String(correctAnswer) })) {
             vocabMap.set(`${match[1]}-${correctAnswer}`, {
               learning: match[1],
               reference: String(correctAnswer)
@@ -722,7 +747,7 @@ export default class VocabService {
           }
 
           match = question.match(/Was bedeutet ['"]([^'"]+)['"]/i);
-          if (match && match[1] && correctAnswer && match[1].trim() !== String(correctAnswer).trim()) {
+          if (match && this._isTrainableSrsPair({ learning: String(correctAnswer), reference: match[1] })) {
             vocabMap.set(`${correctAnswer}-${match[1]}`, {
               learning: String(correctAnswer),
               reference: match[1]
@@ -745,7 +770,7 @@ export default class VocabService {
           answers.forEach((answer, index) => {
             const nativeWord = nativeWords[index];
             const normalizedAnswer = String(answer || '').trim();
-            if (!nativeWord || !normalizedAnswer || nativeWord === normalizedAnswer) {
+            if (!this._isTrainableSrsPair({ learning: nativeWord, reference: normalizedAnswer })) {
               return;
             }
             vocabMap.set(`${nativeWord}-${normalizedAnswer}`, {
@@ -764,31 +789,13 @@ export default class VocabService {
 
   _extractTrainerVocabsFromLessonDidactics(lesson) {
     const vocabMap = new Map();
-    const speakingPrompts = Array.isArray(lesson?.speakingPrompts) ? lesson.speakingPrompts : [];
-    const practicalTasks = Array.isArray(lesson?.practicalTasks) ? lesson.practicalTasks : [];
     const corePatterns = Array.isArray(lesson?.corePatterns) ? lesson.corePatterns : [];
 
     corePatterns.forEach((entry) => {
       const pattern = this._normalizeCorePatternEntry(entry);
       const reference = String(pattern?.target || '').trim();
       const learning = String(pattern?.gloss || '').trim();
-      if (!learning || !reference || learning === reference) return;
-      vocabMap.set(`${learning}-${reference}`, { learning, reference });
-    });
-
-    speakingPrompts.forEach((prompt, index) => {
-      const learning = String(prompt?.prompt || prompt?.title || '').trim();
-      const refEntry = corePatterns[index] ?? corePatterns[0];
-      const reference = String(prompt?.cue || this._corePatternTarget(refEntry) || '').trim();
-      if (!learning || !reference || learning === reference) return;
-      vocabMap.set(`${learning}-${reference}`, { learning, reference });
-    });
-
-    practicalTasks.forEach((task, index) => {
-      const learning = String(task?.text || task?.title || '').trim();
-      const refEntry = corePatterns[index] ?? corePatterns[0];
-      const reference = String(this._corePatternTarget(refEntry) || '').trim();
-      if (!learning || !reference || learning === reference) return;
+      if (!this._isTrainableSrsPair({ learning, reference })) return;
       vocabMap.set(`${learning}-${reference}`, { learning, reference });
     });
 
@@ -1808,16 +1815,17 @@ export default class VocabService {
         [Op.lte]: now
       }
     };
-    const totalDueCount = await VocabSrsItem.count({ where: dueWhere });
-    const rows = await VocabSrsItem.findAll({
+    const dueRows = await VocabSrsItem.findAll({
       where: dueWhere,
       order: [
         ['nextDueAt', 'ASC'],
         ['wrongCount', 'DESC'],
         ['stage', 'ASC']
-      ],
-      limit
+      ]
     });
+    const validDueRows = dueRows.filter((item) => this._isTrainableSrsPair(item));
+    const rows = validDueRows.slice(0, limit);
+    const totalDueCount = validDueRows.length;
 
     return {
       courseId: course.id,