feat(reader): autonomous Wikipedia import and reading with media, tables and settings

app/main.py

@@ -30,7 +30,11 @@ from app.auth import (
 from app.dictionary import lookup_term, USER_AGENT, close_http_client
 from app.export import export_csv, export_json, export_anki_tsv, export_markdown
 from app.srs import calculate_sm2, generate_quiz_options, build_activity_heatmap, calculate_cefr_level
-from app.reader import parse_epub, parse_text, parse_pdf, count_words, extract_rfc_number, fetch_and_parse_rfc, parse_rfc_text
+from app.reader import (
+    parse_epub, parse_text, parse_pdf, count_words, extract_rfc_number,
+    fetch_and_parse_rfc, parse_rfc_text, fetch_and_parse_wikipedia,
+    extract_wikipedia_title_and_lang
+)
 from app.ai import translate_sentence_ai, get_ollama_status
 from app.phrasal import detect_phrasal_verb
 from app.cefr import (
@@ -187,6 +191,14 @@ class RfcFetchRequest(BaseModel):
     force: Optional[bool] = False
 
 
+class WikipediaFetchRequest(BaseModel):
+    query: Optional[str] = ""
+    lang: Optional[str] = "simple"
+    time_budget: Optional[str] = "10-15"
+    topic: Optional[str] = "random"
+    force: Optional[bool] = False
+
+
 class SentenceTranslateRequest(BaseModel):
     sentence: str = Field(..., min_length=1)
     target_word: Optional[str] = ""
@@ -1194,7 +1206,7 @@ async def api_get_books(user: dict = Depends(require_user)):
             """
             SELECT id, user_id, title, author, source_type,
                    json_array_length(chapters_json) AS chapters_count,
-                   total_words, current_chapter, current_position,
+                   total_words, current_chapter, current_position, cover_url,
                    created_at, updated_at
             FROM books
             WHERE user_id = ?
@@ -1309,7 +1321,7 @@ async def api_create_book(
             """
             SELECT id, user_id, title, author, source_type, chapters_json,
                    json_array_length(chapters_json) AS chapters_count,
-                   current_chapter, current_position, total_words,
+                   current_chapter, current_position, total_words, cover_url,
                    created_at, updated_at
             FROM books WHERE id = ?
             """,
@@ -1347,7 +1359,7 @@ async def api_fetch_rfc(
                 """
                 SELECT id, user_id, title, author, source_type, chapters_json,
                        json_array_length(chapters_json) AS chapters_count,
-                       current_chapter, current_position, total_words,
+                       current_chapter, current_position, total_words, cover_url,
                        created_at, updated_at
                 FROM books
                 WHERE user_id = ? AND source_type = 'rfc' AND (title LIKE ? OR title LIKE ?)
@@ -1388,7 +1400,114 @@ async def api_fetch_rfc(
             """
             SELECT id, user_id, title, author, source_type, chapters_json,
                    json_array_length(chapters_json) AS chapters_count,
-                   current_chapter, current_position, total_words,
+                   current_chapter, current_position, total_words, cover_url,
+                   created_at, updated_at
+            FROM books WHERE id = ?
+            """,
+            (new_id,),
+        ).fetchone()
+
+        res = dict(row)
+        res["chapters"] = chapters
+        res["already_existed"] = False
+        return res
+
+
+@app.post("/api/books/fetch-wikipedia")
+async def api_fetch_wikipedia(
+    req: WikipediaFetchRequest,
+    user: dict = Depends(require_user),
+):
+    """
+    Fetches a Wikipedia article (by title/URL or random topic & duration),
+    parses sections into readable chapters with media and tables,
+    and stores autonomously in the user's SQLite bookshelf (source_type = 'wikipedia').
+    """
+    raw_query = (req.query or "").strip()
+    target_lang = (req.lang or "simple").strip()
+    time_budget = (req.time_budget or "10-15").strip()
+    topic = (req.topic or "random").strip()
+
+    # If specific title or URL provided and force is False, check for existing book
+    if raw_query and not req.force:
+        title_cand, lang_cand = extract_wikipedia_title_and_lang(raw_query, default_lang=target_lang)
+        search_title = title_cand.replace("_", " ")
+        with get_db() as conn:
+            existing = conn.execute(
+                """
+                SELECT id, user_id, title, author, source_type, chapters_json,
+                       json_array_length(chapters_json) AS chapters_count,
+                       current_chapter, current_position, total_words, cover_url,
+                       created_at, updated_at
+                FROM books
+                WHERE user_id = ? AND source_type = 'wikipedia' AND lower(title) = lower(?)
+                ORDER BY id DESC LIMIT 1
+                """,
+                (user["id"], search_title),
+            ).fetchone()
+            if existing:
+                res = dict(existing)
+                try:
+                    res["chapters"] = json.loads(res["chapters_json"])
+                except Exception:
+                    res["chapters"] = []
+                res["already_existed"] = True
+                return res
+
+    try:
+        title, author, chapters, cover_url, total_words, wiki_url = fetch_and_parse_wikipedia(
+            title_or_url=raw_query,
+            lang=target_lang,
+            time_budget=time_budget,
+            topic=topic,
+        )
+    except ValueError as ve:
+        raise HTTPException(status_code=400, detail=str(ve))
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Ошибка загрузки статьи Wikipedia: {e}")
+
+    # Check again by resolved title if not forcing
+    if not req.force:
+        with get_db() as conn:
+            existing = conn.execute(
+                """
+                SELECT id, user_id, title, author, source_type, chapters_json,
+                       json_array_length(chapters_json) AS chapters_count,
+                       current_chapter, current_position, total_words, cover_url,
+                       created_at, updated_at
+                FROM books
+                WHERE user_id = ? AND source_type = 'wikipedia' AND lower(title) = lower(?)
+                ORDER BY id DESC LIMIT 1
+                """,
+                (user["id"], title),
+            ).fetchone()
+            if existing:
+                res = dict(existing)
+                try:
+                    res["chapters"] = json.loads(res["chapters_json"])
+                except Exception:
+                    res["chapters"] = []
+                res["already_existed"] = True
+                return res
+
+    chapters_json = json.dumps(chapters, ensure_ascii=False)
+
+    with get_db() as conn:
+        cursor = conn.execute(
+            """
+            INSERT INTO books (
+                user_id, title, author, source_type, chapters_json,
+                current_chapter, current_position, total_words, cover_url
+            ) VALUES (?, ?, ?, 'wikipedia', ?, 0, 0.0, ?, ?)
+            """,
+            (user["id"], title, author, chapters_json, total_words, cover_url),
+        )
+        new_id = cursor.lastrowid
+        row = conn.execute(
+            """
+            SELECT id, user_id, title, author, source_type, chapters_json,
+                   json_array_length(chapters_json) AS chapters_count,
+                   current_chapter, current_position, total_words, cover_url,
                    created_at, updated_at
             FROM books WHERE id = ?
             """,
@@ -1411,7 +1530,7 @@ async def api_get_book(book_id: int, user: dict = Depends(require_user)):
             """
             SELECT id, user_id, title, author, source_type, chapters_json,
                    json_array_length(chapters_json) AS chapters_count,
-                   current_chapter, current_position, total_words,
+                   current_chapter, current_position, total_words, cover_url,
                    created_at, updated_at
             FROM books
             WHERE id = ? AND user_id = ?
@@ -1488,7 +1607,7 @@ async def api_update_book_progress(
             """
             SELECT id, user_id, title, author, source_type, chapters_json,
                    json_array_length(chapters_json) AS chapters_count,
-                   current_chapter, current_position, total_words,
+                   current_chapter, current_position, total_words, cover_url,
                    created_at, updated_at
             FROM books
             WHERE id = ?