Files
tapir/internal/runner/runner.go
T
mathias 598ba5d34f
CI / Lint / Test / Vet (push) Successful in 10s
CI / Build & Import (push) Successful in 9s
CI / Mirror to GitHub (push) Failing after 3s
feat(runner): TAPIR_FETCH_DELAY to throttle transcript fetches
Diagnosis of a live run: caption tracks resolve fine (player + watch-page scrape),
but the timedtext baseUrl fetch returns 429 under back-to-back volume — YouTube
rate-limits the unauthenticated caption-download endpoint per IP. A per-video
delay spaces the fetches. (Follow-up: treat 429 distinctly from genuine
no-caption instead of silently degrading to SourceNone; consider Whisper if the
endpoint stays hostile at any sustainable rate.)
2026-06-03 12:20:12 +02:00

170 lines
5.1 KiB
Go

// Package runner wires the engine to the durable store for the `tapir run`
// command. It owns the cross-restart dedup the engine core deliberately does
// not: the engine's in-memory processed map is process-lifetime only, so this
// loads the store's SeenVideoIDs and skips videos already summarized in a prior
// run. It also assigns each video its durable store id (UpsertVideo) before
// processing, so the summary's video_id equals the dedup key.
//
// It depends on small local interfaces (VideoStore, Processor), not concrete
// types, so the loop is tested with fakes — no live YouTube, gateway, or PG.
package runner
import (
"context"
"errors"
"fmt"
"log/slog"
"os"
"time"
"gitea.d-ma.be/mathias/tapir/internal/domain"
"gitea.d-ma.be/mathias/tapir/internal/ports"
"gitea.d-ma.be/mathias/tapir/internal/usecase"
)
// VideoStore is the durable persistence the run loop needs: assign a stable id +
// metadata, and read the already-summarized set. *store.Store satisfies it.
type VideoStore interface {
UpsertVideo(ctx context.Context, v domain.Video) (string, error)
SeenVideoIDs(ctx context.Context, userID string) (map[string]bool, error)
}
// Processor runs the core use case for a single video. *usecase.Engine
// satisfies it.
type Processor interface {
ProcessNewVideo(ctx context.Context, v domain.Video) (usecase.ProcessResult, error)
}
// Runner walks a user's subscriptions, persists each candidate video, skips the
// ones already summarized (durably), and processes the rest through the engine.
type Runner struct {
src ports.VideoSource
store VideoStore
engine Processor
userID string
log *slog.Logger
}
// New builds a Runner. A nil logger falls back to slog.Default.
func New(src ports.VideoSource, store VideoStore, engine Processor, userID string, log *slog.Logger) *Runner {
if log == nil {
log = slog.Default()
}
return &Runner{src: src, store: store, engine: engine, userID: userID, log: log}
}
// Stats summarizes one RunOnce pass.
type Stats struct {
Candidates int
Summarized int
SkippedSeen int
SkippedNoText int
Errors int
}
// RunOnce performs a single pass over the user's subscriptions. Per-item errors
// are logged and collected (one bad video or channel does not abort the pass)
// and returned joined alongside the Stats gathered.
func (r *Runner) RunOnce(ctx context.Context) (Stats, error) {
var (
stats Stats
errs []error
)
// Throttle transcript fetches: unauthenticated caption scraping gets
// soft-throttled by YouTube under heavy back-to-back volume (captionTracks
// silently stripped from the player response). A small per-video delay keeps
// a full pass under the radar. TAPIR_FETCH_DELAY (Go duration), 0 = off.
fetchDelay, _ := time.ParseDuration(os.Getenv("TAPIR_FETCH_DELAY"))
seen, err := r.store.SeenVideoIDs(ctx, r.userID)
if err != nil {
return stats, fmt.Errorf("runner: load seen videos: %w", err)
}
subs, err := r.src.ListSubscriptions(ctx, r.userID)
if err != nil {
return stats, fmt.Errorf("runner: list subscriptions: %w", err)
}
for _, sub := range subs {
vids, err := r.src.NewVideos(ctx, sub)
if err != nil {
errs = append(errs, fmt.Errorf("new videos for %q: %w", sub.ChannelTitle, err))
stats.Errors++
continue
}
for _, v := range vids {
stats.Candidates++
v.UserID = r.userID // keep the dedup/FK key consistent with config
id, err := r.store.UpsertVideo(ctx, v)
if err != nil {
errs = append(errs, fmt.Errorf("upsert video %q: %w", v.ProviderVideoID, err))
stats.Errors++
continue
}
v.ID = id
if seen[id] {
stats.SkippedSeen++
continue
}
seen[id] = true // also guard against the same video within this pass
if fetchDelay > 0 {
time.Sleep(fetchDelay)
}
res, err := r.engine.ProcessNewVideo(ctx, v)
if err != nil {
errs = append(errs, fmt.Errorf("process %q: %w", v.ProviderVideoID, err))
stats.Errors++
continue
}
switch {
case res.Skipped:
stats.SkippedNoText++
r.log.Info("skipped video (no transcript)", "video", v.ProviderVideoID, "title", v.Title)
case res.Summary != nil:
stats.Summarized++
r.log.Info("summarized video", "video", v.ProviderVideoID, "title", v.Title,
"provider", res.Summary.AIProvider, "model", res.Summary.AIModel)
}
}
}
return stats, errors.Join(errs...)
}
// Loop runs RunOnce immediately, then on every interval tick until ctx is
// cancelled. A zero or negative interval means a single pass (no loop). Per-pass
// errors are logged, not fatal, so a transient failure doesn't kill the watcher.
func (r *Runner) Loop(ctx context.Context, interval time.Duration) error {
runPass := func() {
stats, err := r.RunOnce(ctx)
r.log.Info("run pass complete",
"candidates", stats.Candidates, "summarized", stats.Summarized,
"skipped_seen", stats.SkippedSeen, "skipped_no_text", stats.SkippedNoText,
"errors", stats.Errors)
if err != nil {
r.log.Warn("run pass had errors", "err", err)
}
}
runPass()
if interval <= 0 {
return nil
}
ticker := time.NewTicker(interval)
defer ticker.Stop()
for {
select {
case <-ctx.Done():
return ctx.Err()
case <-ticker.C:
runPass()
}
}
}