hoàn thành tính năng mp3 player

backend/alembic/versions/0003_add_pdf_audios.py

@@ -0,0 +1,35 @@
+"""add pdf_audios table
+
+Revision ID: 0003_add_pdf_audios
+Revises: 0002_add_role_status
+Create Date: 2026-04-02
+"""
+from alembic import op
+import sqlalchemy as sa
+
+revision = "0003_add_pdf_audios"
+down_revision = "0002_add_role_status"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "pdf_audios",
+        sa.Column("id", sa.Integer(), primary_key=True, index=True),
+        sa.Column("pdf_id", sa.Integer(), sa.ForeignKey("pdfs.id", ondelete="CASCADE"), nullable=False, index=True),
+        sa.Column("file_path", sa.Text(), nullable=False),
+        sa.Column("track_name", sa.String(255), nullable=False),
+        sa.Column("page_number", sa.Integer(), nullable=True),
+        sa.Column("sort_order", sa.Integer(), nullable=False, server_default="1"),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.func.now(),
+            nullable=False,
+        ),
+    )
+
+
+def downgrade() -> None:
+    op.drop_table("pdf_audios")

backend/app/main.py

@@ -4,7 +4,7 @@ import os
 from fastapi import FastAPI
 from fastapi.middleware.cors import CORSMiddleware
 
-from .routers import admin, annotations, auth, backup, pdfs, ws
+from .routers import admin, annotations, audio, auth, backup, pdfs, ws
 from .database import SessionLocal
 from .models import User, UserRole, UserStatus
 from .security import hash_password
@@ -57,5 +57,6 @@ app.include_router(auth.router, prefix="/api")
 app.include_router(admin.router, prefix="/api")
 app.include_router(backup.router, prefix="/api")
 app.include_router(pdfs.router, prefix="/api")
+app.include_router(audio.router, prefix="/api")
 app.include_router(annotations.router, prefix="/api")
 app.include_router(ws.router)  # WebSocket — no /api prefix (ws:// path)

backend/app/models.py

@@ -10,6 +10,7 @@ from sqlalchemy import (
     ForeignKey,
     DateTime,
     UniqueConstraint,
+    Boolean,
 )
 from sqlalchemy.dialects.postgresql import JSONB
 from sqlalchemy.orm import relationship
@@ -57,6 +58,7 @@ class PDF(Base):
     # Relationships
     owner = relationship("User", back_populates="pdfs")
     annotations = relationship("Annotation", back_populates="pdf", cascade="all, delete-orphan")
+    audio_tracks = relationship("PDFAudio", back_populates="pdf", cascade="all, delete-orphan")
 
 
 class Annotation(Base):
@@ -81,3 +83,18 @@ class Annotation(Base):
     # Relationships
     pdf = relationship("PDF", back_populates="annotations")
     owner = relationship("User", back_populates="annotations")
+
+
+class PDFAudio(Base):
+    __tablename__ = "pdf_audios"
+
+    id = Column(Integer, primary_key=True, index=True)
+    pdf_id = Column(Integer, ForeignKey("pdfs.id", ondelete="CASCADE"), nullable=False, index=True)
+    file_path = Column(Text, nullable=False)
+    track_name = Column(String(255), nullable=False)
+    page_number = Column(Integer, nullable=True)   # optional: link to a specific page
+    sort_order = Column(Integer, nullable=False, default=1)
+    created_at = Column(DateTime(timezone=True), default=lambda: datetime.now(timezone.utc), nullable=False)
+
+    # Relationships
+    pdf = relationship("PDF", back_populates="audio_tracks")

backend/app/routers/audio.py

@@ -0,0 +1,185 @@
+"""
+Audio management endpoints.
+
+POST /api/pdfs/{pdf_id}/audio  — upload one or more MP3 files (admin/teacher only)
+GET  /api/pdfs/{pdf_id}/audio  — list all audio tracks for a PDF (any authenticated user)
+DELETE /api/pdfs/{pdf_id}/audio/{audio_id} — delete a single track (admin/teacher only)
+GET  /api/pdfs/{pdf_id}/audio/{audio_id}/file — stream the audio file
+"""
+import os
+import re
+import uuid
+from pathlib import Path
+
+from fastapi import APIRouter, Depends, Form, HTTPException, Query, UploadFile, status
+from fastapi.responses import FileResponse
+from sqlalchemy.orm import Session
+
+from ..database import get_db
+from ..dependencies import get_current_user, require_role
+from ..models import PDF, PDFAudio, UserRole
+from ..schemas import AudioOut, AudioUploadResult
+
+router = APIRouter(prefix="/pdfs", tags=["audio"])
+
+UPLOAD_DIR = Path(os.getenv("UPLOAD_DIR", "/uploads"))
+MAX_AUDIO_SIZE = 200 * 1024 * 1024  # 200 MB per file
+ALLOWED_AUDIO_TYPES = {"audio/mpeg", "audio/mp3", "audio/mp4", "audio/x-m4a", "audio/ogg", "application/octet-stream"}
+
+
+def _audio_dir(pdf_id: int) -> Path:
+    d = UPLOAD_DIR / "audio" / str(pdf_id)
+    d.mkdir(parents=True, exist_ok=True)
+    return d
+
+
+def _safe_track_name(raw: str) -> str:
+    """Strip control characters; truncate to 255 chars."""
+    cleaned = re.sub(r"[\x00-\x1f\x7f]", "", raw).strip()
+    return cleaned[:255] or "Track"
+
+
+# ── GET /api/pdfs/{pdf_id}/audio ──────────────────────────────────────────────
+
+@router.get("/{pdf_id}/audio", response_model=list[AudioOut])
+def list_audio(
+    pdf_id: int,
+    current_user=Depends(get_current_user),
+    db: Session = Depends(get_db),
+):
+    pdf = db.get(PDF, pdf_id)
+    if not pdf:
+        raise HTTPException(status_code=404, detail="PDF not found.")
+    tracks = (
+        db.query(PDFAudio)
+        .filter(PDFAudio.pdf_id == pdf_id)
+        .order_by(PDFAudio.sort_order, PDFAudio.id)
+        .all()
+    )
+    return tracks
+
+
+# ── POST /api/pdfs/{pdf_id}/audio ────────────────────────────────────────────
+
+@router.post(
+    "/{pdf_id}/audio",
+    response_model=list[AudioUploadResult],
+    status_code=status.HTTP_201_CREATED,
+)
+async def upload_audio(
+    pdf_id: int,
+    files: list[UploadFile],
+    track_names: list[str] = Form(default=[]),
+    page_numbers: list[int] = Form(default=[]),
+    sort_orders: list[int] = Form(default=[]),
+    current_user=Depends(require_role(UserRole.admin, UserRole.teacher)),
+    db: Session = Depends(get_db),
+):
+    """
+    Upload one or more audio files for a PDF.
+
+    Optional form fields (parallel arrays, indexed per file):
+      - track_names: display name for each track
+      - page_numbers: page the track belongs to (0 = whole PDF / unset)
+      - sort_orders: ordering integer (default 1)
+    """
+    pdf = db.get(PDF, pdf_id)
+    if not pdf:
+        raise HTTPException(status_code=404, detail="PDF not found.")
+
+    results: list[AudioUploadResult] = []
+
+    for idx, file in enumerate(files):
+        fname = file.filename or f"audio_{idx + 1}"
+
+        content_type = (file.content_type or "").lower()
+        if content_type not in ALLOWED_AUDIO_TYPES:
+            # Accept based on extension if browser sends wrong MIME
+            ext = Path(fname).suffix.lower()
+            if ext not in {".mp3", ".m4a", ".ogg", ".wav", ".aac"}:
+                results.append(AudioUploadResult(filename=fname, success=False, error="Unsupported audio format."))
+                continue
+
+        body = await file.read()
+        if len(body) > MAX_AUDIO_SIZE:
+            results.append(AudioUploadResult(filename=fname, success=False, error="File exceeds 200 MB limit."))
+            continue
+
+        safe_name = f"{uuid.uuid4()}{Path(fname).suffix or '.mp3'}"
+        dest = _audio_dir(pdf_id) / safe_name
+        dest.write_bytes(body)
+
+        # Derive metadata from parallel arrays (fall back to sensible defaults)
+        raw_track_name = track_names[idx] if idx < len(track_names) else Path(fname).stem
+        track_name = _safe_track_name(raw_track_name) or Path(fname).stem or f"Track {idx + 1}"
+        page_num = page_numbers[idx] if idx < len(page_numbers) else None
+        if page_num == 0:
+            page_num = None
+        sort_ord = sort_orders[idx] if idx < len(sort_orders) else (idx + 1)
+
+        # Determine next sort order if not supplied
+        if not sort_orders:
+            existing_count = db.query(PDFAudio).filter(PDFAudio.pdf_id == pdf_id).count()
+            sort_ord = existing_count + idx + 1
+
+        audio = PDFAudio(
+            pdf_id=pdf_id,
+            file_path=str(dest),
+            track_name=track_name,
+            page_number=page_num,
+            sort_order=sort_ord,
+        )
+        db.add(audio)
+        db.commit()
+        db.refresh(audio)
+
+        results.append(AudioUploadResult(filename=fname, success=True, audio=AudioOut.model_validate(audio)))
+
+    return results
+
+
+# ── DELETE /api/pdfs/{pdf_id}/audio/{audio_id} ───────────────────────────────
+
+@router.delete("/{pdf_id}/audio/{audio_id}", status_code=status.HTTP_204_NO_CONTENT)
+def delete_audio(
+    pdf_id: int,
+    audio_id: int,
+    current_user=Depends(require_role(UserRole.admin, UserRole.teacher)),
+    db: Session = Depends(get_db),
+):
+    track = db.query(PDFAudio).filter(PDFAudio.id == audio_id, PDFAudio.pdf_id == pdf_id).first()
+    if not track:
+        raise HTTPException(status_code=404, detail="Audio track not found.")
+    file_path = Path(track.file_path)
+    db.delete(track)
+    db.commit()
+    if file_path.exists():
+        file_path.unlink(missing_ok=True)
+
+
+# ── GET /api/pdfs/{pdf_id}/audio/{audio_id}/file ─────────────────────────────
+
+@router.get("/{pdf_id}/audio/{audio_id}/file")
+def serve_audio(
+    pdf_id: int,
+    audio_id: int,
+    current_user=Depends(get_current_user),
+    db: Session = Depends(get_db),
+):
+    track = db.query(PDFAudio).filter(PDFAudio.id == audio_id, PDFAudio.pdf_id == pdf_id).first()
+    if not track:
+        raise HTTPException(status_code=404, detail="Audio track not found.")
+    file_path = Path(track.file_path)
+    if not file_path.exists():
+        raise HTTPException(status_code=404, detail="Audio file not found on disk.")
+
+    # Determine media type from extension
+    ext = file_path.suffix.lower()
+    media_type_map = {".mp3": "audio/mpeg", ".m4a": "audio/mp4", ".ogg": "audio/ogg", ".wav": "audio/wav", ".aac": "audio/aac"}
+    media_type = media_type_map.get(ext, "audio/mpeg")
+
+    return FileResponse(
+        path=str(file_path),
+        media_type=media_type,
+        filename=f"{track.track_name}{ext}",
+    )

backend/app/schemas.py

@@ -98,6 +98,26 @@ class AnnotationIn(BaseModel):
     canvas_data: dict
 
 
+# ── Audio ─────────────────────────────────────────────────────────────────────
+
+class AudioOut(BaseModel):
+    id: int
+    pdf_id: int
+    track_name: str
+    page_number: int | None
+    sort_order: int
+    created_at: datetime
+
+    model_config = {"from_attributes": True}
+
+
+class AudioUploadResult(BaseModel):
+    filename: str
+    success: bool
+    audio: AudioOut | None = None
+    error: str | None = None
+
+
 class AnnotationOut(BaseModel):
     id: int
     pdf_id: int