feat(auth): interactive YouTube OAuth code flow

`tapir auth` mints a refresh token for the single Stage-0 user: bind a local redirect listener, print the consent URL (offline access + forced consent so Google returns a refresh token), validate the state param, exchange the code, and persist the refresh token through the SecretStore port. Written fresh on x/oauth2 (ADR-006). Token is never logged or returned. Tests cover exchange, missing-refresh-token rejection, and the full listener flow with httptest. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-02 20:59:16 +02:00
parent c93b433aaf
commit 460f4bb1de
2 changed files with 364 additions and 0 deletions
@@ -0,0 +1,176 @@
+// Package auth implements the interactive OAuth 2.0 authorization-code flow used
+// by `tapir auth` to mint a YouTube refresh token for the single Stage-0 user.
+// It is written fresh on golang.org/x/oauth2 (ADR-006): the ingestion repo's
+// oauth package is inbound MCP server auth and unrelated to this outbound
+// provider flow.
+//
+// The minted refresh token is persisted through the SecretStore port (the file
+// store at Stage 0) and is never logged or returned to the caller. Only the
+// opaque token ref crosses package boundaries afterward.
+package auth
+
+import (
+	"context"
+	"crypto/rand"
+	"encoding/hex"
+	"fmt"
+	"io"
+	"net"
+	"net/http"
+	"net/url"
+
+	"golang.org/x/oauth2"
+)
+
+// GoogleEndpoint is Google's OAuth2 endpoint, inlined to avoid the heavy
+// golang.org/x/oauth2/google dependency for two URLs (mirrors the youtube
+// adapter's choice).
+var GoogleEndpoint = oauth2.Endpoint{
+	AuthURL:  "https://accounts.google.com/o/oauth2/auth",
+	TokenURL: "https://oauth2.googleapis.com/token",
+}
+
+// DefaultScopes request read access to subscriptions/search and the force-ssl
+// scope the Data API captions endpoints require.
+var DefaultScopes = []string{
+	"https://www.googleapis.com/auth/youtube.readonly",
+	"https://www.googleapis.com/auth/youtube.force-ssl",
+}
+
+// TokenWriter persists a secret value under an opaque ref. *secrets.FileStore
+// satisfies it; tests use a fake. (The read side is ports.SecretStore.)
+type TokenWriter interface {
+	Put(ref, value string) error
+}
+
+// Config wires the flow. Endpoint defaults to GoogleEndpoint when zero, so tests
+// can point it at an httptest token server.
+type Config struct {
+	ClientID     string
+	ClientSecret string
+	RedirectURL  string // e.g. "http://localhost:8080/callback"
+	Scopes       []string
+	TokenRef     string // SecretStore ref to persist the refresh token under
+	Endpoint     oauth2.Endpoint
+}
+
+func oauthConfig(c Config) *oauth2.Config {
+	ep := c.Endpoint
+	if ep == (oauth2.Endpoint{}) {
+		ep = GoogleEndpoint
+	}
+	scopes := c.Scopes
+	if len(scopes) == 0 {
+		scopes = DefaultScopes
+	}
+	return &oauth2.Config{
+		ClientID:     c.ClientID,
+		ClientSecret: c.ClientSecret,
+		RedirectURL:  c.RedirectURL,
+		Scopes:       scopes,
+		Endpoint:     ep,
+	}
+}
+
+// Exchange swaps an authorization code for a token and persists the refresh
+// token through the writer. It errors if the provider returned no refresh token
+// (e.g. consent was not forced with offline access), since without one the
+// token is useless for the unattended run loop. The token is never logged.
+func Exchange(ctx context.Context, c Config, secrets TokenWriter, code string) error {
+	conf := oauthConfig(c)
+	tok, err := conf.Exchange(ctx, code)
+	if err != nil {
+		return fmt.Errorf("auth: exchange code: %w", err)
+	}
+	if tok.RefreshToken == "" {
+		return fmt.Errorf("auth: provider returned no refresh token (re-consent with offline access)")
+	}
+	if err := secrets.Put(c.TokenRef, tok.RefreshToken); err != nil {
+		return fmt.Errorf("auth: persist refresh token: %w", err)
+	}
+	return nil
+}
+
+// Run performs the full interactive flow: it binds a local listener on the
+// redirect URL's host, prints the consent URL to out, waits for the provider's
+// redirect (validating the state parameter), then exchanges the captured code
+// and persists the refresh token. It blocks until the callback arrives, ctx is
+// cancelled, or the listener fails.
+func Run(ctx context.Context, c Config, secrets TokenWriter, out io.Writer) error {
+	conf := oauthConfig(c)
+
+	u, err := url.Parse(c.RedirectURL)
+	if err != nil {
+		return fmt.Errorf("auth: parse redirect url %q: %w", c.RedirectURL, err)
+	}
+	ln, err := net.Listen("tcp", u.Host)
+	if err != nil {
+		return fmt.Errorf("auth: bind redirect listener on %q: %w", u.Host, err)
+	}
+
+	state, err := randomState()
+	if err != nil {
+		return err
+	}
+
+	type result struct {
+		code string
+		err  error
+	}
+	resCh := make(chan result, 1)
+
+	mux := http.NewServeMux()
+	mux.HandleFunc(u.Path, func(w http.ResponseWriter, r *http.Request) {
+		q := r.URL.Query()
+		if e := q.Get("error"); e != "" {
+			http.Error(w, "authorization failed: "+e, http.StatusBadRequest)
+			resCh <- result{err: fmt.Errorf("auth: provider returned error %q", e)}
+			return
+		}
+		if q.Get("state") != state {
+			http.Error(w, "state mismatch", http.StatusBadRequest)
+			resCh <- result{err: fmt.Errorf("auth: state mismatch (possible CSRF)")}
+			return
+		}
+		code := q.Get("code")
+		if code == "" {
+			http.Error(w, "missing code", http.StatusBadRequest)
+			resCh <- result{err: fmt.Errorf("auth: redirect carried no code")}
+			return
+		}
+		_, _ = io.WriteString(w, "Tapir: authorization received. You can close this tab.")
+		resCh <- result{code: code}
+	})
+
+	srv := &http.Server{Handler: mux}
+	go func() { _ = srv.Serve(ln) }()
+	defer func() { _ = srv.Shutdown(context.Background()) }()
+
+	// AccessTypeOffline + ApprovalForce make Google return a refresh token even
+	// on re-authorization; without them a repeat consent yields only an access
+	// token and Exchange would reject it.
+	authURL := conf.AuthCodeURL(state, oauth2.AccessTypeOffline, oauth2.ApprovalForce)
+	fmt.Fprintf(out, "Open this URL to authorize Tapir, then return here:\n\n%s\n\n", authURL)
+
+	select {
+	case <-ctx.Done():
+		return ctx.Err()
+	case res := <-resCh:
+		if res.err != nil {
+			return res.err
+		}
+		if err := Exchange(ctx, c, secrets, res.code); err != nil {
+			return err
+		}
+		fmt.Fprintln(out, "Refresh token stored. You can now run `tapir run`.")
+		return nil
+	}
+}
+
+func randomState() (string, error) {
+	b := make([]byte, 16)
+	if _, err := rand.Read(b); err != nil {
+		return "", fmt.Errorf("auth: generate state: %w", err)
+	}
+	return hex.EncodeToString(b), nil
+}
@@ -0,0 +1,188 @@
+package auth_test
+
+import (
+	"context"
+	"net"
+	"net/http"
+	"net/http/httptest"
+	"regexp"
+	"sync"
+	"testing"
+	"time"
+
+	"golang.org/x/oauth2"
+
+	"gitea.d-ma.be/mathias/tapir/internal/auth"
+)
+
+// fakeWriter is a TokenWriter capturing the persisted (ref, value).
+type fakeWriter struct {
+	mu       sync.Mutex
+	ref, val string
+	calls    int
+}
+
+func (w *fakeWriter) Put(ref, value string) error {
+	w.mu.Lock()
+	defer w.mu.Unlock()
+	w.ref, w.val, w.calls = ref, value, w.calls+1
+	return nil
+}
+
+// tokenServer fakes Google's token endpoint, returning the given JSON body for
+// any POST. No live Google contact.
+func tokenServer(t *testing.T, body string) *httptest.Server {
+	t.Helper()
+	return httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(body))
+	}))
+}
+
+func cfg(srvURL string) auth.Config {
+	return auth.Config{
+		ClientID:     "cid",
+		ClientSecret: "csecret",
+		RedirectURL:  "http://localhost:18099/callback",
+		TokenRef:     "youtube/refresh_token",
+		Endpoint:     oauth2.Endpoint{AuthURL: srvURL + "/auth", TokenURL: srvURL + "/token"},
+	}
+}
+
+func TestExchange_PersistsRefreshToken(t *testing.T) {
+	srv := tokenServer(t, `{"access_token":"at","refresh_token":"rt-secret","token_type":"Bearer","expires_in":3600}`)
+	defer srv.Close()
+
+	w := &fakeWriter{}
+	if err := auth.Exchange(context.Background(), cfg(srv.URL), w, "the-code"); err != nil {
+		t.Fatalf("Exchange: %v", err)
+	}
+	if w.ref != "youtube/refresh_token" {
+		t.Errorf("persisted ref = %q", w.ref)
+	}
+	if w.val != "rt-secret" {
+		t.Errorf("persisted token = %q, want rt-secret", w.val)
+	}
+}
+
+func TestExchange_RejectsMissingRefreshToken(t *testing.T) {
+	srv := tokenServer(t, `{"access_token":"at","token_type":"Bearer","expires_in":3600}`)
+	defer srv.Close()
+
+	w := &fakeWriter{}
+	err := auth.Exchange(context.Background(), cfg(srv.URL), w, "the-code")
+	if err == nil {
+		t.Fatal("want error when no refresh token returned")
+	}
+	if w.calls != 0 {
+		t.Errorf("nothing should be persisted on failure; Put called %d times", w.calls)
+	}
+}
+
+// stateRe pulls the CSRF state out of the printed consent URL.
+var stateRe = regexp.MustCompile(`[?&]state=([a-f0-9]+)`)
+
+// urlWriter forwards each Write to a channel so the test can read the consent
+// URL Run prints before it blocks on the redirect.
+type urlWriter struct{ ch chan string }
+
+func (w urlWriter) Write(p []byte) (int, error) {
+	w.ch <- string(p)
+	return len(p), nil
+}
+
+func TestRun_FullFlow(t *testing.T) {
+	srv := tokenServer(t, `{"access_token":"at","refresh_token":"rt-secret","token_type":"Bearer","expires_in":3600}`)
+	defer srv.Close()
+
+	// Ensure the fixed redirect port is free before binding.
+	if ln, err := net.Listen("tcp", "localhost:18099"); err == nil {
+		_ = ln.Close()
+	} else {
+		t.Skipf("redirect port 18099 unavailable: %v", err)
+	}
+
+	w := &fakeWriter{}
+	out := urlWriter{ch: make(chan string, 4)}
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	runErr := make(chan error, 1)
+	go func() { runErr <- auth.Run(ctx, cfg(srv.URL), w, out) }()
+
+	// First message carries the consent URL with the state param.
+	var state string
+	select {
+	case msg := <-out.ch:
+		m := stateRe.FindStringSubmatch(msg)
+		if m == nil {
+			t.Fatalf("no state in consent message: %q", msg)
+		}
+		state = m[1]
+	case <-time.After(3 * time.Second):
+		t.Fatal("timed out waiting for consent URL")
+	}
+
+	// Simulate the browser hitting the local redirect with code + state.
+	resp, err := http.Get("http://localhost:18099/callback?state=" + state + "&code=the-code")
+	if err != nil {
+		t.Fatalf("callback GET: %v", err)
+	}
+	_ = resp.Body.Close()
+
+	select {
+	case err := <-runErr:
+		if err != nil {
+			t.Fatalf("Run: %v", err)
+		}
+	case <-time.After(3 * time.Second):
+		t.Fatal("Run did not complete after callback")
+	}
+
+	if w.val != "rt-secret" {
+		t.Errorf("persisted token = %q, want rt-secret", w.val)
+	}
+}
+
+func TestRun_RejectsStateMismatch(t *testing.T) {
+	srv := tokenServer(t, `{"access_token":"at","refresh_token":"rt","token_type":"Bearer","expires_in":3600}`)
+	defer srv.Close()
+
+	if ln, err := net.Listen("tcp", "localhost:18099"); err == nil {
+		_ = ln.Close()
+	} else {
+		t.Skipf("redirect port 18099 unavailable: %v", err)
+	}
+
+	w := &fakeWriter{}
+	out := urlWriter{ch: make(chan string, 4)}
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	runErr := make(chan error, 1)
+	go func() { runErr <- auth.Run(ctx, cfg(srv.URL), w, out) }()
+
+	select {
+	case <-out.ch: // drain consent URL
+	case <-time.After(3 * time.Second):
+		t.Fatal("timed out waiting for consent URL")
+	}
+
+	resp, err := http.Get("http://localhost:18099/callback?state=WRONG&code=the-code")
+	if err != nil {
+		t.Fatalf("callback GET: %v", err)
+	}
+	_ = resp.Body.Close()
+
+	select {
+	case err := <-runErr:
+		if err == nil {
+			t.Fatal("want error on state mismatch")
+		}
+	case <-time.After(3 * time.Second):
+		t.Fatal("Run did not return after bad callback")
+	}
+	if w.calls != 0 {
+		t.Errorf("no token should be persisted on state mismatch; Put called %d times", w.calls)
+	}
+}