add auth system and split backend into two-image CI/CD build
Some checks failed
Build and Push / build (push) Failing after 17s

Backend:
- password hashing via hashlib.scrypt
- stateless HMAC-SHA256 tokens (7-day expiry)
- POST /api/auth/login, /api/auth/register (admin), /api/auth/reset-password
- admin user created from NAT20_ADMIN_USERNAME/PASSWORD on first startup
- users table, campaign_shares table, created_by on campaigns
- require_user dependency on all routes except auth
- campaign sharing: GET/POST/DELETE /api/campaigns/{id}/shares

Frontend:
- AuthContext: user/token state, login/logout, global fetch Auth header
- Login page, Users page (admin user management)
- route protection, sidebar user info/sign out

Docker/CI:
- split backend/Dockerfile into thin app-only image
- backend/Dockerfile.deps builds the heavy WhisperX/PyTorch base
- CI builds deps only when requirements.txt changes
- docker compose pull now fetches ~100KB app layer instead of 3.5GB
This commit is contained in:
KansaiGaijin
2026-07-30 12:05:56 +12:00
parent 990276366d
commit 5081ba00f6
16 changed files with 800 additions and 114 deletions

100
backend/app/auth.py Normal file
View File

@@ -0,0 +1,100 @@
"""
Authentication module — zero-dependency (stdlib only).
Password hashing via hashlib.scrypt, stateless tokens via HMAC-SHA256.
"""
import base64
import hashlib
import hmac
import json
import os
import time
from pathlib import Path
from typing import Optional
from . import database as db
from .config import ADMIN_PASSWORD, ADMIN_USERNAME, JWT_SECRET
JWT_ALGORITHM = "HS256"
JWT_EXPIRY = 86400 * 7 # 7 days
def hash_password(password: str) -> str:
salt = os.urandom(16)
key = hashlib.scrypt(password.encode(), salt=salt, n=16384, r=8, p=1, dklen=32)
return base64.b64encode(salt + key).decode()
def verify_password(password: str, stored: str) -> bool:
try:
data = base64.b64decode(stored)
salt, key = data[:16], data[16:]
key2 = hashlib.scrypt(password.encode(), salt=salt, n=16384, r=8, p=1, dklen=32)
return hmac.compare_digest(key, key2)
except Exception:
return False
def create_token(user_id: str, username: str, is_admin: bool) -> str:
header = {"alg": "HS256", "typ": "JWT"}
payload = {
"sub": user_id,
"username": username,
"admin": is_admin,
"iat": time.time(),
"exp": time.time() + JWT_EXPIRY,
}
h = base64.urlsafe_b64encode(json.dumps(header, separators=(",", ":")).encode()).rstrip(b"=").decode()
p = base64.urlsafe_b64encode(json.dumps(payload, separators=(",", ":")).encode()).rstrip(b"=").decode()
sig = hmac.new(JWT_SECRET.encode(), f"{h}.{p}".encode(), hashlib.sha256).digest()
s = base64.urlsafe_b64encode(sig).rstrip(b"=").decode()
return f"{h}.{p}.{s}"
def decode_token(token: str) -> Optional[dict]:
try:
parts = token.split(".")
if len(parts) != 3:
return None
h, p, s = parts
sig = base64.urlsafe_b64decode(s + "==")
expected = hmac.new(JWT_SECRET.encode(), f"{h}.{p}".encode(), hashlib.sha256).digest()
if not hmac.compare_digest(sig, expected):
return None
payload = json.loads(base64.urlsafe_b64decode(p + "=="))
if payload.get("exp", 0) < time.time():
return None
return payload
except Exception:
return None
def authenticate(username: str, password: str) -> Optional[dict]:
user = db.get_user_by_username(username)
if not user or not verify_password(password, user["password_hash"]):
return None
token = create_token(user["id"], user["username"], bool(user["is_admin"]))
return {
"token": token,
"user": {"id": user["id"], "username": user["username"], "is_admin": bool(user["is_admin"])},
}
def get_user_from_token(token: str) -> Optional[dict]:
payload = decode_token(token)
if not payload:
return None
user = db.get_user(payload["sub"])
if not user:
return None
return {"id": user["id"], "username": user["username"], "is_admin": bool(user["is_admin"])}
def ensure_admin_exists():
"""Called on startup — creates the admin user if no users exist."""
conn = db.get_conn()
row = conn.execute("SELECT COUNT(*) as cnt FROM users").fetchone()
if row["cnt"] > 0:
return
pw = ADMIN_PASSWORD or os.urandom(8).hex()
db.create_user(ADMIN_USERNAME, hash_password(pw), is_admin=True)
print(f"[auth] Created admin user '{ADMIN_USERNAME}' (password: {pw})")

View File

@@ -28,6 +28,11 @@ def campaign_transcript_dir(campaign_id: str) -> Path:
def campaign_notes_dir(campaign_id: str) -> Path:
return campaign_dir(campaign_id) / "notes"
# Auth config
JWT_SECRET = os.environ.get("NAT20_JWT_SECRET", os.urandom(32).hex())
ADMIN_USERNAME = os.environ.get("NAT20_ADMIN_USERNAME", "admin")
ADMIN_PASSWORD = os.environ.get("NAT20_ADMIN_PASSWORD", "admin")
for d in (UPLOAD_DIR, AUDIO_DIR, TRANSCRIPT_DIR, NOTES_DIR):
d.mkdir(parents=True, exist_ok=True)

View File

@@ -87,14 +87,36 @@ def init_db():
updated_at REAL NOT NULL
)
""")
conn.execute("""
CREATE TABLE IF NOT EXISTS users (
id TEXT PRIMARY KEY,
username TEXT UNIQUE NOT NULL,
password_hash TEXT NOT NULL,
is_admin INTEGER NOT NULL DEFAULT 0,
created_at REAL NOT NULL
)
""")
conn.execute("""
CREATE TABLE IF NOT EXISTS campaign_shares (
campaign_id TEXT NOT NULL REFERENCES campaigns(id) ON DELETE CASCADE,
user_id TEXT NOT NULL REFERENCES users(id) ON DELETE CASCADE,
role TEXT NOT NULL DEFAULT 'editor',
PRIMARY KEY (campaign_id, user_id)
)
""")
conn.execute("""
CREATE TABLE IF NOT EXISTS campaigns (
id TEXT PRIMARY KEY,
name TEXT NOT NULL,
description TEXT NOT NULL DEFAULT '',
created_at REAL NOT NULL
created_at REAL NOT NULL,
created_by TEXT REFERENCES users(id) ON DELETE SET NULL
)
""")
try:
conn.execute("ALTER TABLE campaigns ADD COLUMN created_by TEXT REFERENCES users(id) ON DELETE SET NULL")
except Exception:
pass
conn.execute("""
CREATE TABLE IF NOT EXISTS campaign_settings (
campaign_id TEXT NOT NULL REFERENCES campaigns(id) ON DELETE CASCADE,
@@ -110,7 +132,7 @@ def init_db():
default = conn.execute("SELECT id FROM campaigns WHERE id = 'default'").fetchone()
if not default:
conn.execute(
"INSERT INTO campaigns (id, name, description, created_at) VALUES (?, ?, '', ?)",
"INSERT INTO campaigns (id, name, description, created_at, created_by) VALUES (?, ?, '', ?, NULL)",
("default", "Default Campaign", now()),
)
conn.execute("UPDATE sessions SET campaign_id = 'default' WHERE campaign_id IS NULL")
@@ -207,12 +229,12 @@ def get_job(job_id: str) -> dict | None:
row = get_conn().execute("SELECT * FROM jobs WHERE id = ?", (job_id,)).fetchone()
return dict(row) if row else None
def create_campaign(name: str, description: str = "") -> dict:
def create_campaign(name: str, description: str = "", created_by: str | None = None) -> dict:
campaign_id = new_id()
with tx() as conn:
conn.execute(
"INSERT INTO campaigns (id, name, description, created_at) VALUES (?, ?, ?, ?)",
(campaign_id, name, description, now()),
"INSERT INTO campaigns (id, name, description, created_at, created_by) VALUES (?, ?, ?, ?, ?)",
(campaign_id, name, description, now(), created_by),
)
# Inherit CAMPAIGN_SETTINGS from the "default" campaign as a starting point
parent = conn.execute(
@@ -255,6 +277,89 @@ def update_campaign(campaign_id: str, name: str | None = None, description: str
conn.execute(f"UPDATE campaigns SET {cols} WHERE id = ?", (*fields.values(), campaign_id))
return get_campaign(campaign_id)
# ── Users ──────────────────────────────────────────────────────────────────
def get_user(user_id: str) -> dict | None:
row = get_conn().execute("SELECT * FROM users WHERE id = ?", (user_id,)).fetchone()
return dict(row) if row else None
def get_user_by_username(username: str) -> dict | None:
row = get_conn().execute("SELECT * FROM users WHERE username = ?", (username,)).fetchone()
return dict(row) if row else None
def create_user(username: str, password_hash: str, is_admin: bool = False) -> dict:
uid = new_id()
with tx() as conn:
conn.execute(
"INSERT INTO users (id, username, password_hash, is_admin, created_at) VALUES (?, ?, ?, ?, ?)",
(uid, username, password_hash, 1 if is_admin else 0, now()),
)
return get_user(uid)
def update_user_password(user_id: str, password_hash: str):
with tx() as conn:
conn.execute("UPDATE users SET password_hash = ? WHERE id = ?", (password_hash, user_id))
def list_users() -> list[dict]:
rows = get_conn().execute(
"SELECT id, username, is_admin, created_at FROM users ORDER BY created_at"
).fetchall()
return [dict(r) for r in rows]
# ── Campaign sharing ──────────────────────────────────────────────────────
def share_campaign(campaign_id: str, user_id: str, role: str = "editor"):
with tx() as conn:
conn.execute(
"INSERT OR REPLACE INTO campaign_shares (campaign_id, user_id, role) VALUES (?, ?, ?)",
(campaign_id, user_id, role),
)
def unshare_campaign(campaign_id: str, user_id: str):
with tx() as conn:
conn.execute(
"DELETE FROM campaign_shares WHERE campaign_id = ? AND user_id = ?",
(campaign_id, user_id),
)
def get_campaign_shares(campaign_id: str) -> list[dict]:
rows = get_conn().execute("""
SELECT u.id, u.username, cs.role
FROM campaign_shares cs
JOIN users u ON u.id = cs.user_id
WHERE cs.campaign_id = ?
""", (campaign_id,)).fetchall()
return [dict(r) for r in rows]
def get_user_campaigns(user_id: str) -> list[dict]:
rows = get_conn().execute("""
SELECT c.*, cs.role as access_role,
(SELECT COUNT(*) FROM sessions s WHERE s.campaign_id = c.id) as session_count
FROM campaigns c
LEFT JOIN campaign_shares cs ON cs.campaign_id = c.id
WHERE c.created_by = ? OR cs.user_id = ?
ORDER BY c.created_at DESC
""", (user_id, user_id)).fetchall()
return [dict(r) for r in rows]
def can_access_campaign(user_id: str, campaign_id: str) -> bool:
row = get_conn().execute("""
SELECT 1 FROM campaigns WHERE id = ? AND created_by = ?
UNION
SELECT 1 FROM campaign_shares WHERE campaign_id = ? AND user_id = ?
""", (campaign_id, user_id, campaign_id, user_id)).fetchone()
return row is not None
def delete_campaign(campaign_id: str) -> bool:
with tx() as conn:
conn.execute("UPDATE sessions SET campaign_id = NULL WHERE campaign_id = ?", (campaign_id,))

View File

@@ -1,6 +1,6 @@
from pathlib import Path
from fastapi import FastAPI, Request
from fastapi import Depends, FastAPI, Request
from fastapi.middleware.cors import CORSMiddleware
from fastapi.responses import FileResponse, JSONResponse
from fastapi import HTTPException
@@ -8,6 +8,9 @@ from fastapi import HTTPException
from . import database as db, config
from .logging_config import setup_logging, get_logger
from .errors import PipelineError
from .auth import ensure_admin_exists
from .routers.auth import require_user
from .routers import auth as auth_router
from .routers import settings as settings_router
from .routers import sessions as sessions_router
from .routers import speakers as speakers_router
@@ -46,18 +49,20 @@ async def unhandled_error_handler(request: Request, exc: Exception):
@app.on_event("startup")
def on_startup():
db.init_db()
ensure_admin_exists()
log.info("Database initialized at %s", config.DB_PATH)
app.include_router(settings_router.router)
app.include_router(sessions_router.router)
app.include_router(speakers_router.router)
app.include_router(notes_router.router)
app.include_router(jobs_router.router)
app.include_router(models_router.router)
app.include_router(diagnostics_router.router)
app.include_router(files_router.router)
app.include_router(campaigns_router.router)
app.include_router(auth_router.router)
app.include_router(settings_router.router, dependencies=[Depends(require_user)])
app.include_router(sessions_router.router, dependencies=[Depends(require_user)])
app.include_router(speakers_router.router, dependencies=[Depends(require_user)])
app.include_router(notes_router.router, dependencies=[Depends(require_user)])
app.include_router(jobs_router.router, dependencies=[Depends(require_user)])
app.include_router(models_router.router, dependencies=[Depends(require_user)])
app.include_router(diagnostics_router.router, dependencies=[Depends(require_user)])
app.include_router(files_router.router, dependencies=[Depends(require_user)])
app.include_router(campaigns_router.router, dependencies=[Depends(require_user)])
MEDIA_TYPES = {
@@ -77,7 +82,7 @@ AUDIO_EXTS = frozenset(MEDIA_TYPES.keys())
@app.get("/api/sessions/{session_id}/audio")
def get_audio(session_id: str):
def get_audio(session_id: str, user: dict = Depends(require_user)):
row = db.get_conn().execute(
"SELECT video_path, audio_path FROM sessions WHERE id = ?", (session_id,)
).fetchone()

View File

@@ -0,0 +1,75 @@
from fastapi import APIRouter, Depends, HTTPException, Header
from typing import Optional
from .. import database as db
from ..auth import authenticate, get_user_from_token, hash_password, ensure_admin_exists
router = APIRouter(prefix="/api/auth", tags=["auth"])
def require_user(authorization: Optional[str] = Header(None)):
if not authorization or not authorization.startswith("Bearer "):
raise HTTPException(401, "Not authenticated")
user = get_user_from_token(authorization[7:])
if not user:
raise HTTPException(401, "Invalid or expired token")
return user
def require_admin(user: dict = Depends(require_user)):
if not user["is_admin"]:
raise HTTPException(403, "Admin access required")
return user
@router.post("/login")
def login(body: dict):
username = body.get("username", "").strip()
password = body.get("password", "")
if not username or not password:
raise HTTPException(400, "Username and password are required")
result = authenticate(username, password)
if not result:
raise HTTPException(401, "Invalid username or password")
return result
@router.post("/register")
def register(body: dict, admin: dict = Depends(require_admin)):
username = body.get("username", "").strip()
password = body.get("password", "")
if not username or not password:
raise HTTPException(400, "Username and password are required")
if len(password) < 4:
raise HTTPException(400, "Password must be at least 4 characters")
existing = db.get_user_by_username(username)
if existing:
raise HTTPException(409, "Username already exists")
user = db.create_user(username, hash_password(password))
return {"id": user["id"], "username": user["username"]}
@router.post("/reset-password")
def reset_password(body: dict, admin: dict = Depends(require_admin)):
user_id = body.get("user_id", "").strip()
new_password = body.get("new_password", "")
if not user_id or not new_password:
raise HTTPException(400, "user_id and new_password are required")
if len(new_password) < 4:
raise HTTPException(400, "Password must be at least 4 characters")
user = db.get_user(user_id)
if not user:
raise HTTPException(404, "User not found")
db.update_user_password(user_id, hash_password(new_password))
return {"ok": True}
@router.get("/me")
def me(user: dict = Depends(require_user)):
user["is_admin"] = bool(user["is_admin"])
return user
@router.get("/users")
def list_users(admin: dict = Depends(require_admin)):
return db.list_users()

View File

@@ -1,36 +1,36 @@
from fastapi import APIRouter, HTTPException
from fastapi import APIRouter, Depends, HTTPException
from .. import database as db
from ..config import CAMPAIGN_SETTINGS, merge_campaign_settings_with_env
from .auth import require_user, require_admin
router = APIRouter(prefix="/api/campaigns", tags=["campaigns"])
@router.get("")
def list_campaigns():
return db.list_campaigns()
def list_campaigns(user: dict = Depends(require_user)):
return db.get_user_campaigns(user["id"])
@router.post("")
def create_campaign(body: dict):
def create_campaign(body: dict, user: dict = Depends(require_user)):
name = body.get("name", "").strip()
if not name:
raise HTTPException(400, "Name is required")
desc = body.get("description", "").strip()
return db.create_campaign(name, desc)
return db.create_campaign(name, desc, user["id"])
@router.get("/{campaign_id}")
def get_campaign(campaign_id: str):
campaign = db.get_campaign(campaign_id)
if not campaign:
def get_campaign(campaign_id: str, user: dict = Depends(require_user)):
if not db.can_access_campaign(user["id"], campaign_id):
raise HTTPException(404, "Campaign not found")
return campaign
return db.get_campaign(campaign_id)
@router.put("/{campaign_id}")
def update_campaign(campaign_id: str, body: dict):
if not db.get_campaign(campaign_id):
def update_campaign(campaign_id: str, body: dict, user: dict = Depends(require_user)):
if not db.can_access_campaign(user["id"], campaign_id):
raise HTTPException(404, "Campaign not found")
updated = db.update_campaign(
campaign_id,
@@ -41,26 +41,80 @@ def update_campaign(campaign_id: str, body: dict):
@router.delete("/{campaign_id}")
def delete_campaign(campaign_id: str):
if not db.get_campaign(campaign_id):
def delete_campaign(campaign_id: str, user: dict = Depends(require_user)):
campaign = db.get_campaign(campaign_id)
if not campaign:
raise HTTPException(404, "Campaign not found")
if campaign.get("created_by") != user["id"] and not user.get("is_admin"):
raise HTTPException(403, "Only the campaign owner can delete it")
db.delete_campaign(campaign_id)
return {"ok": True}
# ── Campaign settings ────────────────────────────────────────────────────
@router.get("/{campaign_id}/settings")
def get_campaign_settings(campaign_id: str):
if not db.get_campaign(campaign_id):
def get_campaign_settings(campaign_id: str, user: dict = Depends(require_user)):
if not db.can_access_campaign(user["id"], campaign_id):
raise HTTPException(404, "Campaign not found")
raw = db.get_campaign_settings(campaign_id)
return merge_campaign_settings_with_env(raw)
@router.post("/{campaign_id}/settings")
def update_campaign_settings(campaign_id: str, body: dict):
if not db.get_campaign(campaign_id):
def update_campaign_settings(campaign_id: str, body: dict, user: dict = Depends(require_user)):
if not db.can_access_campaign(user["id"], campaign_id):
raise HTTPException(404, "Campaign not found")
clean = {k: str(v) for k, v in body.items() if k in CAMPAIGN_SETTINGS}
db.update_campaign_settings(campaign_id, clean)
raw = db.get_campaign_settings(campaign_id)
return merge_campaign_settings_with_env(raw)
# ── Sharing ───────────────────────────────────────────────────────────────
@router.get("/{campaign_id}/shares")
def list_shares(campaign_id: str, user: dict = Depends(require_user)):
campaign = db.get_campaign(campaign_id)
if not campaign:
raise HTTPException(404, "Campaign not found")
if campaign.get("created_by") != user["id"] and not user.get("is_admin"):
raise HTTPException(403, "Only the campaign owner can manage sharing")
return db.get_campaign_shares(campaign_id)
@router.post("/{campaign_id}/shares")
def add_share(campaign_id: str, body: dict, user: dict = Depends(require_user)):
campaign = db.get_campaign(campaign_id)
if not campaign:
raise HTTPException(404, "Campaign not found")
if campaign.get("created_by") != user["id"] and not user.get("is_admin"):
raise HTTPException(403, "Only the campaign owner can manage sharing")
target_username = body.get("username", "").strip()
if not target_username:
raise HTTPException(400, "username is required")
target = db.get_user_by_username(target_username)
if not target:
raise HTTPException(404, "User not found")
if target["id"] == campaign["created_by"]:
raise HTTPException(400, "Cannot share campaign with its owner")
role = body.get("role", "editor")
db.share_campaign(campaign_id, target["id"], role)
return {"ok": True}
@router.delete("/{campaign_id}/shares")
def remove_share(campaign_id: str, body: dict, user: dict = Depends(require_user)):
campaign = db.get_campaign(campaign_id)
if not campaign:
raise HTTPException(404, "Campaign not found")
if campaign.get("created_by") != user["id"] and not user.get("is_admin"):
raise HTTPException(403, "Only the campaign owner can manage sharing")
target_username = body.get("username", "").strip()
if not target_username:
raise HTTPException(400, "username is required")
target = db.get_user_by_username(target_username)
if not target:
raise HTTPException(404, "User not found")
db.unshare_campaign(campaign_id, target["id"])
return {"ok": True}