Accounts Phase 1 foundation: schema + WAL, auth core, email sender

Groundwork for self-hosted accounts (magic link + Google later), no third parties. - db: account tables (users, identities, login_tokens, sessions, saved_articles, user_history, user_prefs); identities link multiple sign-in methods to one user by verified email. connect() now enables WAL + busy_timeout so the API can write account data alongside the host ingestion cycle. - auth.py: users/identities (find-or-create + link), single-use magic-link tokens, opaque sessions — all secrets stored only as SHA-256 hashes. - email_send.py: minimal STARTTLS SMTP sender + the magic-link email. Secrets (SMTP, Google, session) live in the API container's env_file, not git. API endpoints + sign-in UI come next. 105 tests pass. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-03 01:02:24 +00:00
parent acbc06a9e5
commit 6a514aa56b
4 changed files with 375 additions and 1 deletions
@@ -0,0 +1,154 @@
+"""Account auth core: users, identities, magic-link tokens, sessions.
+
+Stdlib-only helpers over a sqlite3 connection. The caller (the API) supplies the
+connection and commits. Self-hosted, minimal-PII. Secrets are random and stored
+only as SHA-256 hashes — the raw token is shown exactly once (in the magic link
+or returned to the client as a session token).
+
+Identity model: a user can have several `identities` (email magic-link, google,
+later apple). They resolve to ONE user by verified email, so signing in with
+Google and with a magic link to the same address lands on the same account.
+"""
+
+from __future__ import annotations
+
+import hashlib
+import secrets
+import sqlite3
+from datetime import datetime, timedelta, timezone
+
+LOGIN_TOKEN_TTL = timedelta(minutes=15)
+SESSION_TTL = timedelta(days=60)
+
+
+def _now() -> datetime:
+    return datetime.now(timezone.utc)
+
+
+def _iso(dt: datetime) -> str:
+    return dt.replace(microsecond=0).isoformat()
+
+
+def _parse(value: str) -> datetime:
+    dt = datetime.fromisoformat(value)
+    return dt if dt.tzinfo else dt.replace(tzinfo=timezone.utc)
+
+
+def _hash(token: str) -> str:
+    return hashlib.sha256(token.encode("utf-8")).hexdigest()
+
+
+def normalize_email(email: str) -> str:
+    return (email or "").strip().lower()
+
+
+# ---- users & identities ------------------------------------------------------
+
+def get_user(conn: sqlite3.Connection, user_id: int) -> sqlite3.Row | None:
+    return conn.execute(
+        "SELECT id, email, display_name, created_at FROM users WHERE id = ?", (user_id,)
+    ).fetchone()
+
+
+def find_or_create_user(
+    conn: sqlite3.Connection,
+    email: str,
+    provider: str,
+    provider_subject: str,
+    display_name: str | None = None,
+) -> int:
+    """Resolve (or create) the user for a verified sign-in, linking the identity.
+
+    Order: an existing identity wins; else an existing user with the same verified
+    email is linked to the new identity; else a fresh user is created.
+    """
+    email = normalize_email(email)
+    existing = conn.execute(
+        "SELECT user_id FROM identities WHERE provider = ? AND provider_subject = ?",
+        (provider, provider_subject),
+    ).fetchone()
+    if existing:
+        return existing["user_id"]
+
+    user = conn.execute("SELECT id FROM users WHERE email = ?", (email,)).fetchone()
+    if user:
+        user_id = user["id"]
+        if display_name:
+            conn.execute(
+                "UPDATE users SET display_name = COALESCE(display_name, ?), "
+                "updated_at = CURRENT_TIMESTAMP WHERE id = ?",
+                (display_name, user_id),
+            )
+    else:
+        user_id = conn.execute(
+            "INSERT INTO users (email, display_name) VALUES (?, ?)", (email, display_name)
+        ).lastrowid
+
+    conn.execute(
+        "INSERT OR IGNORE INTO identities (user_id, provider, provider_subject) VALUES (?, ?, ?)",
+        (user_id, provider, provider_subject),
+    )
+    return user_id
+
+
+# ---- magic-link tokens -------------------------------------------------------
+
+def create_login_token(conn: sqlite3.Connection, email: str) -> str:
+    """Create a single-use sign-in token; return the raw token (goes in the link)."""
+    raw = secrets.token_urlsafe(32)
+    conn.execute(
+        "INSERT INTO login_tokens (email, token_hash, expires_at) VALUES (?, ?, ?)",
+        (normalize_email(email), _hash(raw), _iso(_now() + LOGIN_TOKEN_TTL)),
+    )
+    return raw
+
+
+def consume_login_token(conn: sqlite3.Connection, raw: str) -> str | None:
+    """Return the email for a valid, unused, unexpired token, marking it used.
+
+    Returns None (and changes nothing) if the token is unknown, already used, or
+    expired. Single-use even under concurrent taps (the consumed_at guard).
+    """
+    if not raw:
+        return None
+    row = conn.execute(
+        "SELECT id, email, expires_at, consumed_at FROM login_tokens WHERE token_hash = ?",
+        (_hash(raw),),
+    ).fetchone()
+    if not row or row["consumed_at"] is not None or _parse(row["expires_at"]) < _now():
+        return None
+    consumed = conn.execute(
+        "UPDATE login_tokens SET consumed_at = ? WHERE id = ? AND consumed_at IS NULL",
+        (_iso(_now()), row["id"]),
+    ).rowcount
+    return row["email"] if consumed else None
+
+
+# ---- sessions ----------------------------------------------------------------
+
+def create_session(conn: sqlite3.Connection, user_id: int, user_agent: str | None = None) -> str:
+    """Create a session; return the raw token (cookie value / bearer token)."""
+    raw = secrets.token_urlsafe(32)
+    conn.execute(
+        "INSERT INTO sessions (user_id, token_hash, expires_at, user_agent) VALUES (?, ?, ?, ?)",
+        (user_id, _hash(raw), _iso(_now() + SESSION_TTL), (user_agent or "")[:300]),
+    )
+    return raw
+
+
+def resolve_session(conn: sqlite3.Connection, raw: str | None) -> sqlite3.Row | None:
+    """Return the user for a valid session token (and refresh last_seen), else None."""
+    if not raw:
+        return None
+    row = conn.execute(
+        "SELECT id, user_id, expires_at FROM sessions WHERE token_hash = ?", (_hash(raw),)
+    ).fetchone()
+    if not row or _parse(row["expires_at"]) < _now():
+        return None
+    conn.execute("UPDATE sessions SET last_seen_at = ? WHERE id = ?", (_iso(_now()), row["id"]))
+    return get_user(conn, row["user_id"])
+
+
+def revoke_session(conn: sqlite3.Connection, raw: str | None) -> None:
+    if raw:
+        conn.execute("DELETE FROM sessions WHERE token_hash = ?", (_hash(raw),))
@@ -127,15 +127,90 @@ CREATE TABLE IF NOT EXISTS daily_brief_items (
    PRIMARY KEY (brief_id, article_id),
    UNIQUE (brief_id, rank)
 );
+
+-- ---- Accounts ----------------------------------------------------------------
+-- Self-hosted, minimal-PII. The host ingestion owns the content tables above;
+-- the API owns these (writes happen via the API, so the DB runs in WAL mode).
+
+CREATE TABLE IF NOT EXISTS users (
+    id INTEGER PRIMARY KEY AUTOINCREMENT,
+    email TEXT NOT NULL UNIQUE,
+    display_name TEXT,
+    created_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    updated_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP
+);
+
+-- One row per sign-in method linked to a user; lets Google + magic-link
+-- (same verified email) resolve to a single account.
+CREATE TABLE IF NOT EXISTS identities (
+    id INTEGER PRIMARY KEY AUTOINCREMENT,
+    user_id INTEGER NOT NULL REFERENCES users(id) ON DELETE CASCADE,
+    provider TEXT NOT NULL,           -- 'email' | 'google' | 'apple'
+    provider_subject TEXT NOT NULL,   -- email address, or the provider's stable user id
+    created_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    UNIQUE (provider, provider_subject)
+);
+CREATE INDEX IF NOT EXISTS idx_identities_user ON identities(user_id);
+
+-- Single-use, short-lived magic-link tokens (stored hashed).
+CREATE TABLE IF NOT EXISTS login_tokens (
+    id INTEGER PRIMARY KEY AUTOINCREMENT,
+    email TEXT NOT NULL,
+    token_hash TEXT NOT NULL UNIQUE,
+    created_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    expires_at TEXT NOT NULL,
+    consumed_at TEXT
+);
+CREATE INDEX IF NOT EXISTS idx_login_tokens_email ON login_tokens(email);
+
+-- Active sessions (opaque token stored hashed); validated for cookie or bearer.
+CREATE TABLE IF NOT EXISTS sessions (
+    id INTEGER PRIMARY KEY AUTOINCREMENT,
+    user_id INTEGER NOT NULL REFERENCES users(id) ON DELETE CASCADE,
+    token_hash TEXT NOT NULL UNIQUE,
+    created_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    last_seen_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    expires_at TEXT NOT NULL,
+    user_agent TEXT
+);
+CREATE INDEX IF NOT EXISTS idx_sessions_user ON sessions(user_id);
+
+CREATE TABLE IF NOT EXISTS saved_articles (
+    user_id INTEGER NOT NULL REFERENCES users(id) ON DELETE CASCADE,
+    article_id INTEGER NOT NULL REFERENCES articles(id) ON DELETE CASCADE,
+    saved_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    PRIMARY KEY (user_id, article_id)
+);
+
+CREATE TABLE IF NOT EXISTS user_history (
+    user_id INTEGER NOT NULL REFERENCES users(id) ON DELETE CASCADE,
+    article_id INTEGER NOT NULL REFERENCES articles(id) ON DELETE CASCADE,
+    event TEXT NOT NULL DEFAULT 'seen',   -- 'seen' | 'dismissed'
+    at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    PRIMARY KEY (user_id, article_id, event)
+);
+
+CREATE TABLE IF NOT EXISTS user_prefs (
+    user_id INTEGER PRIMARY KEY REFERENCES users(id) ON DELETE CASCADE,
+    prefs_json TEXT NOT NULL DEFAULT '{}',
+    updated_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP
+);
 """


 def connect(db_path: Path | str) -> sqlite3.Connection:
    path = Path(db_path)
    path.parent.mkdir(parents=True, exist_ok=True)
-    conn = sqlite3.connect(path)
+    conn = sqlite3.connect(path, check_same_thread=False)
    conn.row_factory = sqlite3.Row
    conn.execute("PRAGMA foreign_keys = ON")
+    # WAL lets the API write account data while the ingestion cycle writes content
+    # concurrently (readers never block the writer). busy_timeout rides out the
+    # brief moments the single writer lock is held. Both are no-ops if already set.
+    conn.execute("PRAGMA busy_timeout = 5000")
+    if str(path) != ":memory:":
+        conn.execute("PRAGMA journal_mode = WAL")
+        conn.execute("PRAGMA synchronous = NORMAL")
    return conn


@@ -0,0 +1,74 @@
+"""Minimal transactional email over SMTP (magic-link sign-in).
+
+Config comes from the environment (GOODNEWS_SMTP_*), supplied to the API
+container via its env_file. STARTTLS submission; plain-text with a simple HTML
+alternative. No third-party email API — just the configured relay.
+"""
+
+from __future__ import annotations
+
+import os
+import smtplib
+import ssl
+from email.message import EmailMessage
+from html import escape
+
+
+def smtp_configured() -> bool:
+    return bool(os.environ.get("GOODNEWS_SMTP_HOST"))
+
+
+def _cfg() -> dict:
+    return {
+        "host": os.environ.get("GOODNEWS_SMTP_HOST", ""),
+        "port": int(os.environ.get("GOODNEWS_SMTP_PORT", "587")),
+        "user": os.environ.get("GOODNEWS_SMTP_USER", ""),
+        "password": os.environ.get("GOODNEWS_SMTP_PASSWORD", ""),
+        "sender": os.environ.get("GOODNEWS_SMTP_FROM", "Upbeat Bytes <hello@upbeatbytes.com>"),
+    }
+
+
+def send_email(to: str, subject: str, text: str, html: str | None = None) -> None:
+    """Send one message. Raises on failure (caller decides how loud to be)."""
+    cfg = _cfg()
+    if not cfg["host"]:
+        raise RuntimeError("SMTP not configured (set GOODNEWS_SMTP_HOST)")
+    msg = EmailMessage()
+    msg["From"] = cfg["sender"]
+    msg["To"] = to
+    msg["Subject"] = subject
+    msg.set_content(text)
+    if html:
+        msg.add_alternative(html, subtype="html")
+    context = ssl.create_default_context()
+    with smtplib.SMTP(cfg["host"], cfg["port"], timeout=20) as server:
+        server.ehlo()
+        server.starttls(context=context)
+        server.ehlo()
+        if cfg["user"]:
+            server.login(cfg["user"], cfg["password"])
+        server.send_message(msg)
+
+
+def send_magic_link(to: str, link: str) -> None:
+    """Send a calm, single-purpose sign-in email."""
+    subject = "Your Upbeat Bytes sign-in link"
+    text = (
+        "Welcome back to Upbeat Bytes.\n\n"
+        f"Tap to sign in:\n{link}\n\n"
+        "This link works once and expires in 15 minutes.\n"
+        "If you didn't request it, you can safely ignore this email."
+    )
+    safe = escape(link, quote=True)
+    html = (
+        '<div style="font-family:-apple-system,Segoe UI,Roboto,Helvetica,Arial,sans-serif;'
+        'color:#16263a;line-height:1.6">'
+        "<p>Welcome back to <strong>Upbeat Bytes</strong>.</p>"
+        f'<p><a href="{safe}" style="display:inline-block;background:#0083ad;color:#fff;'
+        'text-decoration:none;padding:11px 20px;border-radius:999px;font-weight:600">'
+        "Sign in</a></p>"
+        '<p style="color:#5d6b78;font-size:14px">This link works once and expires in 15 minutes. '
+        "If you didn't request it, you can safely ignore this email.</p>"
+        "</div>"
+    )
+    send_email(to, subject, text, html)
@@ -0,0 +1,71 @@
+import sqlite3
+from datetime import timedelta
+
+from goodnews import auth
+from goodnews.db import connect, init_db
+
+
+def _db():
+    c = connect(":memory:")
+    init_db(c)
+    # an article to save/history against later flows
+    c.execute("INSERT INTO sources (id,name,feed_url,trust_score) VALUES (1,'S','http://s/f',5)")
+    c.execute("INSERT INTO articles (id,source_id,canonical_url,title,url_hash) "
+              "VALUES (1,1,'http://s/1','t1','h1')")
+    c.commit()
+    return c
+
+
+def test_normalize_email():
+    assert auth.normalize_email("  Foo@Bar.COM ") == "foo@bar.com"
+
+
+def test_find_or_create_links_by_email_and_dedupes_identity():
+    c = _db()
+    uid = auth.find_or_create_user(c, "a@b.com", "email", "a@b.com")
+    # same identity again → same user, no duplicate
+    assert auth.find_or_create_user(c, "a@b.com", "email", "a@b.com") == uid
+    # a different provider with the SAME verified email links to the same user
+    uid2 = auth.find_or_create_user(c, "A@B.com", "google", "google-sub-123", display_name="A")
+    assert uid2 == uid
+    assert c.execute("SELECT COUNT(*) FROM users").fetchone()[0] == 1
+    assert c.execute("SELECT COUNT(*) FROM identities WHERE user_id=?", (uid,)).fetchone()[0] == 2
+    assert auth.get_user(c, uid)["display_name"] == "A"
+
+
+def test_magic_link_token_single_use():
+    c = _db()
+    raw = auth.create_login_token(c, "a@b.com")
+    assert auth.consume_login_token(c, raw) == "a@b.com"   # first use works
+    assert auth.consume_login_token(c, raw) is None         # reuse rejected
+    assert auth.consume_login_token(c, "nonsense") is None  # unknown rejected
+
+
+def test_magic_link_token_expiry():
+    c = _db()
+    raw = auth.create_login_token(c, "a@b.com")
+    # backdate expiry into the past
+    c.execute("UPDATE login_tokens SET expires_at = ?",
+              (auth._iso(auth._now() - timedelta(minutes=1)),))
+    assert auth.consume_login_token(c, raw) is None
+
+
+def test_session_lifecycle():
+    c = _db()
+    uid = auth.find_or_create_user(c, "a@b.com", "email", "a@b.com")
+    tok = auth.create_session(c, uid, user_agent="pytest")
+    user = auth.resolve_session(c, tok)
+    assert user and user["id"] == uid and user["email"] == "a@b.com"
+    assert auth.resolve_session(c, None) is None
+    assert auth.resolve_session(c, "bogus") is None
+    auth.revoke_session(c, tok)
+    assert auth.resolve_session(c, tok) is None
+
+
+def test_session_expiry():
+    c = _db()
+    uid = auth.find_or_create_user(c, "a@b.com", "email", "a@b.com")
+    tok = auth.create_session(c, uid)
+    c.execute("UPDATE sessions SET expires_at = ?",
+              (auth._iso(auth._now() - timedelta(seconds=1)),))
+    assert auth.resolve_session(c, tok) is None