ZDDC/zddc/internal/apps/fetch.go

package apps

import (
	"context"
	"fmt"
	"io"
	"log/slog"
	"net/http"
	"sync"
	"time"
)

// Fetcher pulls URL sources once, caches the body forever, and serves
// from cache on subsequent calls. Path sources don't go through here —
// the handler reads the file directly.
//
// Concurrent calls for the same URL dedupe via singleflight. There is no
// background refresh, no conditional GET, no SHA-256 verification.
type Fetcher struct {
	Cache  *Cache
	Client *http.Client
	Logger *slog.Logger

	sf            singleflightGroup
	embeddedFails sync.Map // url → struct{} (rate-limit "fell back to embedded" warnings)
}

// NewFetcher returns a Fetcher with sensible defaults: 10s timeout, no
// redirects (ops must point at the final URL).
func NewFetcher(cache *Cache, logger *slog.Logger) *Fetcher {
	if logger == nil {
		logger = slog.Default()
	}
	return &Fetcher{
		Cache:  cache,
		Logger: logger,
		Client: &http.Client{
			Timeout: 10 * time.Second,
			CheckRedirect: func(*http.Request, []*http.Request) error {
				return http.ErrUseLastResponse
			},
		},
	}
}

// Fetch returns the body for url. If the cache already has it, returns
// the cached bytes immediately. Otherwise fetches, caches, and returns.
// All concurrent requests for the same URL share one outbound fetch.
func (f *Fetcher) Fetch(ctx context.Context, urlStr string) ([]byte, error) {
	if f.Cache != nil {
		if body, err := f.Cache.Read(urlStr); err == nil {
			return body, nil
		}
	}
	val, err := f.sf.Do(urlStr, func() (any, error) {
		return f.fetchOnce(ctx, urlStr)
	})
	if err != nil {
		return nil, err
	}
	return val.([]byte), nil
}

func (f *Fetcher) fetchOnce(ctx context.Context, urlStr string) ([]byte, error) {
	req, err := http.NewRequestWithContext(ctx, http.MethodGet, urlStr, nil)
	if err != nil {
		return nil, err
	}
	resp, err := f.Client.Do(req)
	if err != nil {
		return nil, err
	}
	defer resp.Body.Close()
	if resp.StatusCode < 200 || resp.StatusCode >= 300 {
		return nil, fmt.Errorf("upstream %s returned HTTP %d", urlStr, resp.StatusCode)
	}
	const maxBytes = 25 * 1024 * 1024
	body, err := io.ReadAll(io.LimitReader(resp.Body, maxBytes+1))
	if err != nil {
		return nil, err
	}
	if int64(len(body)) > maxBytes {
		return nil, fmt.Errorf("response from %s exceeds %d bytes", urlStr, maxBytes)
	}
	if f.Cache != nil {
		if err := f.Cache.Write(urlStr, body); err != nil {
			f.Logger.Warn("cache write failed; serving from response anyway",
				"url", urlStr, "err", err)
		}
	}
	return body, nil
}

// LogEmbeddedFallback emits a one-time warning when the embedded fallback
// is used for a particular source URL. Rate-limited per URL.
func (f *Fetcher) LogEmbeddedFallback(app, urlStr string, reason error) {
	if _, loaded := f.embeddedFails.LoadOrStore(urlStr, struct{}{}); loaded {
		return
	}
	f.Logger.Warn("serving embedded fallback for app HTML",
		"app", app, "url", urlStr, "reason", reason)
}