Add migration learning platform: Onboarding, Translation, Parent Portal

Phase 1.1 — user_language_api.py: Stores native language preference
per user (TR/AR/UK/RU/PL/DE/EN). Onboarding page with flag-based
language selection for students and parents.

Phase 1.2 — translation_service.py: Batch-translates vocabulary words
into target languages via Ollama LLM. Stores in translations JSONB.
New endpoint POST /vocabulary/translate triggers translation.

Phase 2.1 — Parent Portal (/parent): Simplified UI in parent's native
language showing child's learning progress. Daily tips translated.

Phase 2.2 — Parent Quiz (/parent/quiz/[unitId]): Parents can quiz
their child on vocabulary WITHOUT speaking DE or EN. Shows word in
child's learning language + parent's native language as hint.
Answer hidden by default, revealed on tap.

All UI text translated into 7 languages (DE/EN/TR/AR/UK/RU/PL).
Arabic gets RTL layout support.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

backend-lehrer/main.py

@@ -119,6 +119,10 @@ app.include_router(progress_router, prefix="/api")
 from vocabulary_api import router as vocabulary_router
 app.include_router(vocabulary_router, prefix="/api")
 
+# --- 4d. User Language Preferences ---
+from user_language_api import router as user_language_router
+app.include_router(user_language_router, prefix="/api")
+
 from unit_api import router as unit_router
 app.include_router(unit_router)  # Already has /api/units prefix
 

backend-lehrer/translation_service.py

@@ -0,0 +1,179 @@
+"""
+Translation Service — Batch-translates vocabulary words into target languages.
+
+Uses Ollama (local LLM) to translate EN/DE word pairs into TR, AR, UK, RU, PL.
+Translations are cached in vocabulary_words.translations JSONB field.
+
+All processing happens locally — no external API calls, GDPR-compliant.
+"""
+
+import json
+import logging
+import os
+from typing import Any, Dict, List
+
+import httpx
+
+logger = logging.getLogger(__name__)
+
+OLLAMA_BASE_URL = os.getenv("OLLAMA_BASE_URL", "http://host.docker.internal:11434")
+TRANSLATION_MODEL = os.getenv("TRANSLATION_MODEL", "qwen3:30b-a3b")
+
+LANGUAGE_NAMES = {
+    "tr": "Turkish",
+    "ar": "Arabic",
+    "uk": "Ukrainian",
+    "ru": "Russian",
+    "pl": "Polish",
+    "fr": "French",
+    "es": "Spanish",
+}
+
+
+async def translate_words_batch(
+    words: List[Dict[str, str]],
+    target_language: str,
+    batch_size: int = 30,
+) -> List[Dict[str, str]]:
+    """
+    Translate a batch of EN/DE word pairs into a target language.
+
+    Args:
+        words: List of dicts with 'english' and 'german' keys
+        target_language: ISO 639-1 code (tr, ar, uk, ru, pl)
+        batch_size: Words per LLM request
+
+    Returns:
+        List of dicts with 'english', 'translation', 'example' keys
+    """
+    lang_name = LANGUAGE_NAMES.get(target_language, target_language)
+    all_translations = []
+
+    for i in range(0, len(words), batch_size):
+        batch = words[i:i + batch_size]
+        word_list = "\n".join(
+            f"{j+1}. {w['english']} = {w.get('german', '')}"
+            for j, w in enumerate(batch)
+        )
+
+        prompt = f"""Translate these English/German word pairs into {lang_name}.
+For each word, provide the translation and a short example sentence in {lang_name}.
+
+Words:
+{word_list}
+
+Reply ONLY with a JSON array, no explanation:
+[
+  {{"english": "word", "translation": "...", "example": "..."}},
+  ...
+]"""
+
+        try:
+            async with httpx.AsyncClient(timeout=120.0) as client:
+                resp = await client.post(
+                    f"{OLLAMA_BASE_URL}/api/generate",
+                    json={
+                        "model": TRANSLATION_MODEL,
+                        "prompt": prompt,
+                        "stream": False,
+                        "options": {"temperature": 0.2, "num_predict": 4096},
+                    },
+                )
+                resp.raise_for_status()
+                response_text = resp.json().get("response", "")
+
+                # Parse JSON from response
+                import re
+                match = re.search(r'\[[\s\S]*\]', response_text)
+                if match:
+                    batch_translations = json.loads(match.group())
+                    all_translations.extend(batch_translations)
+                    logger.info(
+                        f"Translated batch {i//batch_size + 1}: "
+                        f"{len(batch_translations)} words → {lang_name}"
+                    )
+                else:
+                    logger.warning(f"No JSON array in LLM response for {lang_name}")
+
+        except Exception as e:
+            logger.error(f"Translation batch failed ({lang_name}): {e}")
+
+    return all_translations
+
+
+async def translate_and_store(
+    word_ids: List[str],
+    target_language: str,
+) -> int:
+    """
+    Translate vocabulary words and store in the database.
+
+    Fetches words from DB, translates via LLM, stores in translations JSONB.
+    Skips words that already have a translation for the target language.
+
+    Returns count of newly translated words.
+    """
+    from vocabulary_db import get_pool
+
+    pool = await get_pool()
+    async with pool.acquire() as conn:
+        # Fetch words that need translation
+        rows = await conn.fetch(
+            """
+            SELECT id, english, german, translations
+            FROM vocabulary_words
+            WHERE id = ANY($1::uuid[])
+            """,
+            [__import__('uuid').UUID(wid) for wid in word_ids],
+        )
+
+    words_to_translate = []
+    word_map = {}
+    for row in rows:
+        translations = row["translations"] or {}
+        if isinstance(translations, str):
+            translations = json.loads(translations)
+        if target_language not in translations:
+            words_to_translate.append({
+                "english": row["english"],
+                "german": row["german"],
+            })
+            word_map[row["english"].lower()] = str(row["id"])
+
+    if not words_to_translate:
+        logger.info(f"All {len(rows)} words already translated to {target_language}")
+        return 0
+
+    # Translate
+    results = await translate_words_batch(words_to_translate, target_language)
+
+    # Store results
+    updated = 0
+    async with pool.acquire() as conn:
+        for result in results:
+            en = result.get("english", "").lower()
+            word_id = word_map.get(en)
+            if not word_id:
+                continue
+
+            translation = result.get("translation", "")
+            example = result.get("example", "")
+            if not translation:
+                continue
+
+            await conn.execute(
+                """
+                UPDATE vocabulary_words
+                SET translations = translations || $1::jsonb
+                WHERE id = $2
+                """,
+                json.dumps({target_language: {
+                    "text": translation,
+                    "example": example,
+                }}),
+                __import__('uuid').UUID(word_id),
+            )
+            updated += 1
+
+    logger.info(f"Stored {updated} translations for {target_language}")
+    return updated

backend-lehrer/user_language_api.py

@@ -0,0 +1,86 @@
+"""
+User Language Preferences API — Stores native language + learning level.
+
+Each user (student, parent, teacher) can set their native language.
+This drives: UI language, third-language display in flashcards,
+parent portal language, and translation generation.
+
+Supported languages: de, en, tr, ar, uk, ru, pl
+"""
+
+import logging
+import os
+from typing import Any, Dict, Optional
+
+from fastapi import APIRouter, HTTPException, Query
+from pydantic import BaseModel
+
+logger = logging.getLogger(__name__)
+
+router = APIRouter(prefix="/user", tags=["user-language"])
+
+# Supported native languages with metadata
+SUPPORTED_LANGUAGES = {
+    "de": {"name": "Deutsch", "name_native": "Deutsch", "flag": "de", "rtl": False},
+    "en": {"name": "English", "name_native": "English", "flag": "gb", "rtl": False},
+    "tr": {"name": "Tuerkisch", "name_native": "Turkce", "flag": "tr", "rtl": False},
+    "ar": {"name": "Arabisch", "name_native": "\u0627\u0644\u0639\u0631\u0628\u064a\u0629", "flag": "sy", "rtl": True},
+    "uk": {"name": "Ukrainisch", "name_native": "\u0423\u043a\u0440\u0430\u0457\u043d\u0441\u044c\u043a\u0430", "flag": "ua", "rtl": False},
+    "ru": {"name": "Russisch", "name_native": "\u0420\u0443\u0441\u0441\u043a\u0438\u0439", "flag": "ru", "rtl": False},
+    "pl": {"name": "Polnisch", "name_native": "Polski", "flag": "pl", "rtl": False},
+}
+
+# In-memory store (will be replaced with DB later)
+_preferences: Dict[str, Dict[str, Any]] = {}
+
+
+class LanguagePreference(BaseModel):
+    native_language: str  # ISO 639-1 code
+    role: str = "student"  # student, parent, teacher
+    learning_level: str = "A1"  # A1, A2, B1, B2, C1
+
+
+@router.get("/languages")
+def get_supported_languages():
+    """List all supported native languages with metadata."""
+    return {
+        "languages": [
+            {"code": code, **meta}
+            for code, meta in SUPPORTED_LANGUAGES.items()
+        ]
+    }
+
+
+@router.get("/language-preference")
+def get_language_preference(user_id: str = Query("default")):
+    """Get user's language preference."""
+    pref = _preferences.get(user_id)
+    if not pref:
+        return {"user_id": user_id, "native_language": "de", "role": "student", "learning_level": "A1", "is_default": True}
+    return {**pref, "is_default": False}
+
+
+@router.put("/language-preference")
+def set_language_preference(
+    pref: LanguagePreference,
+    user_id: str = Query("default"),
+):
+    """Set user's native language and learning level."""
+    if pref.native_language not in SUPPORTED_LANGUAGES:
+        raise HTTPException(
+            status_code=400,
+            detail=f"Sprache '{pref.native_language}' nicht unterstuetzt. "
+                   f"Verfuegbar: {', '.join(SUPPORTED_LANGUAGES.keys())}",
+        )
+
+    _preferences[user_id] = {
+        "user_id": user_id,
+        "native_language": pref.native_language,
+        "role": pref.role,
+        "learning_level": pref.learning_level,
+    }
+
+    lang_meta = SUPPORTED_LANGUAGES[pref.native_language]
+    logger.info(f"Language preference set: user={user_id} lang={pref.native_language} ({lang_meta['name']})")
+
+    return {**_preferences[user_id], "language_meta": lang_meta}

backend-lehrer/vocabulary_api.py

@@ -324,3 +324,29 @@ async def api_bulk_import(payload: BulkImportPayload):
     count = await insert_words_bulk(words)
     logger.info(f"Bulk imported {count} vocabulary words")
     return {"imported": count}
+
+
+# ---------------------------------------------------------------------------
+# Translation Generation
+# ---------------------------------------------------------------------------
+
+
+class TranslateRequest(BaseModel):
+    word_ids: List[str]
+    target_language: str
+
+
+@router.post("/translate")
+async def api_translate_words(payload: TranslateRequest):
+    """Generate translations for vocabulary words into a target language.
+
+    Uses local LLM (Ollama) for translation. Results are cached in the
+    vocabulary_words.translations JSONB field.
+    """
+    from translation_service import translate_and_store
+
+    if payload.target_language not in {"tr", "ar", "uk", "ru", "pl", "fr", "es"}:
+        raise HTTPException(status_code=400, detail=f"Sprache '{payload.target_language}' nicht unterstuetzt")
+
+    count = await translate_and_store(payload.word_ids, payload.target_language)
+    return {"translated": count, "target_language": payload.target_language}