storage: JSON files -> SQLite (users/history/searches/events)

- one-time import from the legacy JSON (left in place as snapshot);
  ghost rows (int vs str ep_id) deduped on import, unrepresentable after
- history POST is now an upsert: the (user, slug_id, ep_id) PK is the
  merge key, so the unguarded read-modify-write (lost updates) and the
  corruption-swallow (JSONDecodeError -> [] -> fresh overwrite) are gone
- searches: NOCASE PK upsert + cap 8
- events table + position/search logging (watch-time trajectory),
  90-day retention pruned at boot
- account delete cascades all four tables
- 22/22 tests green; old JSON files untouched
This commit is contained in:
jozamudi committed 2026-08-24 16:02:27 +00:00
1 parent 2c0ee8eae3
commit 202bfa41c3
5 files changed
+222 -105

No files matched your search

+41 -38
View File
@@ -1,4 +1,4 @@
"""users.py — multi-user accounts for ani-web (PBKDF2, single json file).
"""users.py — multi-user accounts for ani-web (PBKDF2, SQLite via db.py).
CLI: .venv/bin/python -m server.users add <username> (password via prompt)
.venv/bin/python -m server.users rm <username>
@@ -10,11 +10,12 @@ import json
import os
import re
import secrets
import sqlite3
import time
from pathlib import Path
DATA = Path(os.environ.get("ANIWEB_DATA", Path(__file__).parent / "data"))
USERS = DATA / "users.json"
from . import db
DATA = db.DATA
SECRET = DATA / "secret.key"
ITER = 210_000 # OWASP 2023 minimum for PBKDF2-SHA256
NAME_RE = re.compile(r"^[a-z0-9_-]{3,32}$")
@@ -32,9 +33,10 @@ def get_secret() -> bytes:
def migrate_legacy():
"""one-time: single-password era -> users.json + per-user history files."""
"""one-time: single-password era -> users.json + per-user history files,
which db.init() (called after this) absorbs into the tables."""
cred = DATA / "credentials.json"
if not USERS.exists() and cred.exists():
if not (DATA / "users.json").exists() and cred.exists():
d = json.loads(cred.read_text())
save({"admin": {"salt": d["salt"], "hash": d["hash"], "created": time.time()}})
cred.unlink()
@@ -45,26 +47,16 @@ def migrate_legacy():
def load() -> dict:
try:
return json.loads(USERS.read_text())
except (FileNotFoundError, json.JSONDecodeError):
return {}
def atomic_write(path, text):
"""tmp + os.replace: a crash mid-write leaves the previous file intact
(in-place writes corrupted the file -> load() returned {} -> accounts
or history silently vanished)."""
tmp = path.parent / (path.name + ".tmp")
fd = os.open(tmp, os.O_WRONLY | os.O_CREAT | os.O_TRUNC, 0o600)
with os.fdopen(fd, "w") as f:
f.write(text)
os.replace(tmp, path)
c = db.connect()
rows = c.execute("SELECT name, salt, hash, iter, created FROM users").fetchall()
c.close()
return {n: {"salt": s, "hash": h, "iter": i, "created": cr} for n, s, h, i, cr in rows}
def save(users: dict):
"""file-based save, only for migrate_legacy (runs before db.init)."""
DATA.mkdir(exist_ok=True)
atomic_write(USERS, json.dumps(users, indent=1))
(DATA / "users.json").write_text(json.dumps(users, indent=1))
def add(username: str, password: str):
@@ -73,37 +65,48 @@ def add(username: str, password: str):
raise ValueError("username: 3-32 chars, only a-z 0-9 _ -")
if len(password) < 8:
raise ValueError("password: min 8 chars")
users = load()
if u in users:
raise ValueError("username already taken")
salt = secrets.token_bytes(16)
users[u] = {"salt": salt.hex(),
"hash": hashlib.pbkdf2_hmac("sha256", password.encode(), salt, ITER).hex(),
"created": time.time()}
save(users)
c = db.connect()
try:
c.execute("INSERT INTO users VALUES (?,?,?,?,?)",
(u, salt.hex(),
hashlib.pbkdf2_hmac("sha256", password.encode(), salt, ITER).hex(),
ITER, time.time()))
c.commit()
except sqlite3.IntegrityError:
raise ValueError("username already taken")
finally:
c.close()
def verify(username: str, password: str) -> bool:
d = load().get(username.lower())
if not d:
c = db.connect()
row = c.execute("SELECT salt, hash, iter FROM users WHERE name=?",
(username.lower(),)).fetchone()
c.close()
if not row:
return False
h = hashlib.pbkdf2_hmac("sha256", password.encode(),
bytes.fromhex(d["salt"]), d.get("iter", ITER))
return secrets.compare_digest(h.hex(), d["hash"])
bytes.fromhex(row[0]), row[2] or ITER)
return secrets.compare_digest(h.hex(), row[1])
def remove(username: str):
u = username.lower()
users = load()
users.pop(u, None)
save(users)
(DATA / "history" / f"{u}.json").unlink(missing_ok=True)
(DATA / "searches" / f"{u}.json").unlink(missing_ok=True)
c = db.connect()
c.execute("DELETE FROM users WHERE name=?", (u,))
c.execute("DELETE FROM history WHERE user=?", (u,))
c.execute("DELETE FROM searches WHERE user=?", (u,))
c.execute("DELETE FROM events WHERE user=?", (u,))
c.commit()
c.close()
if __name__ == "__main__":
import sys
cmd = sys.argv[1] if len(sys.argv) > 1 else ""
migrate_legacy() # file-based, runs before db.init absorbs its output
db.init()
if cmd == "add" and len(sys.argv) > 2:
name = sys.argv[2]
pw = getpass.getpass(f"Password for {name} (min 8 chars): ")