Pete/internal/web/thumbs.go

package web

import (
	"bytes"
	"crypto/sha256"
	"encoding/hex"
	"fmt"
	"image"
	_ "image/gif"
	_ "image/jpeg"
	"image/png"
	"io"
	"log/slog"
	"net/http"
	"net/url"
	"os"
	"os/exec"
	"path/filepath"
	"strings"
	"sync"
	"time"

	"golang.org/x/image/draw"
	_ "golang.org/x/image/webp"

	"pete/internal/storage"
)

const (
	thumbWidth        = 800
	thumbMaxBytes     = 12 << 20 // 12 MiB cap on source download
	thumbFetchTimeout = 12 * time.Second
	thumbCacheDir     = "./data/img-cache"
	thumbQuality      = "45"
	thumbSpeed        = "8"
)

var (
	thumbClient = &http.Client{Timeout: thumbFetchTimeout}

	thumbMu       sync.Mutex
	thumbInFlight = map[string]*sync.WaitGroup{}
)

func thumbKey(u string) string {
	sum := sha256.Sum256([]byte(u))
	return hex.EncodeToString(sum[:16])
}

// thumbURL maps a source image URL to the /img proxy path. Empty input → "".
func thumbURL(src string) string {
	if src == "" {
		return ""
	}
	return "/img/" + thumbKey(src) + ".avif?u=" + url.QueryEscape(src)
}

func (s *Server) handleImg(w http.ResponseWriter, r *http.Request) {
	name := r.PathValue("name")
	hash, ok := strings.CutSuffix(name, ".avif")
	if !ok {
		http.NotFound(w, r)
		return
	}
	src := r.URL.Query().Get("u")
	if src == "" || thumbKey(src) != hash {
		http.NotFound(w, r)
		return
	}
	if !(strings.HasPrefix(src, "http://") || strings.HasPrefix(src, "https://")) {
		http.NotFound(w, r)
		return
	}
	if !storage.IsKnownImageURL(src) {
		http.NotFound(w, r)
		return
	}

	if err := os.MkdirAll(thumbCacheDir, 0o755); err != nil {
		http.Error(w, "cache dir", http.StatusInternalServerError)
		return
	}
	cachePath := filepath.Join(thumbCacheDir, hash+".avif")
	if fi, err := os.Stat(cachePath); err == nil && fi.Size() > 0 {
		serveAvif(w, r, cachePath)
		return
	}

	// Single-flight: collapse concurrent requests for the same image.
	thumbMu.Lock()
	wg, busy := thumbInFlight[hash]
	if !busy {
		wg = &sync.WaitGroup{}
		wg.Add(1)
		thumbInFlight[hash] = wg
	}
	thumbMu.Unlock()

	if busy {
		wg.Wait()
		if fi, err := os.Stat(cachePath); err == nil && fi.Size() > 0 {
			serveAvif(w, r, cachePath)
			return
		}
		http.Redirect(w, r, src, http.StatusFound)
		return
	}
	defer func() {
		thumbMu.Lock()
		delete(thumbInFlight, hash)
		thumbMu.Unlock()
		wg.Done()
	}()

	if err := buildAvifThumb(src, cachePath); err != nil {
		slog.Warn("thumb build failed", "url", src, "err", err)
		http.Redirect(w, r, src, http.StatusFound)
		return
	}
	serveAvif(w, r, cachePath)
}

func serveAvif(w http.ResponseWriter, r *http.Request, path string) {
	w.Header().Set("Content-Type", "image/avif")
	w.Header().Set("Cache-Control", "public, max-age=2592000, immutable")
	http.ServeFile(w, r, path)
}

// buildAvifThumb downloads src, decodes + resizes in Go, then shells out to
// avifenc to produce a compact AVIF at outPath.
func buildAvifThumb(src, outPath string) error {
	req, err := http.NewRequest(http.MethodGet, src, nil)
	if err != nil {
		return err
	}
	req.Header.Set("User-Agent", "Pete-Thumb/1.0 (+https://pete.example)")
	resp, err := thumbClient.Do(req)
	if err != nil {
		return err
	}
	defer resp.Body.Close()
	if resp.StatusCode != http.StatusOK {
		return fmt.Errorf("upstream %d", resp.StatusCode)
	}
	body, err := io.ReadAll(io.LimitReader(resp.Body, thumbMaxBytes+1))
	if err != nil {
		return err
	}
	if len(body) > thumbMaxBytes {
		return fmt.Errorf("image too big (>%d bytes)", thumbMaxBytes)
	}

	if isAvif(body) {
		// If the source AVIF is already within our width budget, skip the
		// decode/resize/re-encode roundtrip and cache the original bytes.
		w, _, infoErr := avifInfo(body)
		if infoErr == nil && w > 0 && w <= thumbWidth {
			return writeFile(outPath, body)
		}
		body, err = avifToPNG(body)
		if err != nil {
			return fmt.Errorf("avifdec: %w", err)
		}
	}
	src2, _, err := image.Decode(bytes.NewReader(body))
	if err != nil {
		return fmt.Errorf("decode: %w", err)
	}
	b := src2.Bounds()
	srcW, srcH := b.Dx(), b.Dy()
	if srcW <= 0 || srcH <= 0 {
		return fmt.Errorf("empty image")
	}
	dstW, dstH := srcW, srcH
	if srcW > thumbWidth {
		dstW = thumbWidth
		dstH = int(float64(srcH) * float64(thumbWidth) / float64(srcW))
		if dstH < 1 {
			dstH = 1
		}
	}
	var resized image.Image = src2
	if dstW != srcW || dstH != srcH {
		dst := image.NewRGBA(image.Rect(0, 0, dstW, dstH))
		draw.CatmullRom.Scale(dst, dst.Bounds(), src2, b, draw.Over, nil)
		resized = dst
	}

	tmp, err := os.CreateTemp("", "pete-thumb-*.png")
	if err != nil {
		return err
	}
	tmpPath := tmp.Name()
	defer os.Remove(tmpPath)
	if err := png.Encode(tmp, resized); err != nil {
		tmp.Close()
		return err
	}
	if err := tmp.Close(); err != nil {
		return err
	}

	// Encode to a sibling temp file then atomically rename so partial outputs
	// never appear in the cache.
	stagedOut := outPath + ".tmp"
	cmd := exec.Command("avifenc",
		"-q", thumbQuality,
		"-s", thumbSpeed,
		tmpPath, stagedOut,
	)
	if out, err := cmd.CombinedOutput(); err != nil {
		os.Remove(stagedOut)
		return fmt.Errorf("avifenc: %w: %s", err, strings.TrimSpace(string(out)))
	}
	return os.Rename(stagedOut, outPath)
}

// isAvif reports whether body looks like an AVIF file via the ISO-BMFF "ftyp"
// brand at bytes 4..12 ("ftypavif"/"ftypavis").
func isAvif(body []byte) bool {
	if len(body) < 12 {
		return false
	}
	if string(body[4:8]) != "ftyp" {
		return false
	}
	brand := string(body[8:12])
	return brand == "avif" || brand == "avis"
}

// avifInfo returns the width and height of an AVIF by shelling out to
// `avifdec --info`. It still decodes the image internally but doesn't write
// pixels to disk, so it's cheaper than full avifToPNG.
func avifInfo(body []byte) (int, int, error) {
	in, err := os.CreateTemp("", "pete-avif-info-*.avif")
	if err != nil {
		return 0, 0, err
	}
	inPath := in.Name()
	defer os.Remove(inPath)
	if _, err := in.Write(body); err != nil {
		in.Close()
		return 0, 0, err
	}
	if err := in.Close(); err != nil {
		return 0, 0, err
	}
	out, err := exec.Command("avifdec", "--info", inPath).CombinedOutput()
	if err != nil {
		return 0, 0, fmt.Errorf("%w: %s", err, strings.TrimSpace(string(out)))
	}
	for _, line := range strings.Split(string(out), "\n") {
		line = strings.TrimSpace(line)
		rest, ok := strings.CutPrefix(line, "* Resolution")
		if !ok {
			continue
		}
		_, dims, ok := strings.Cut(rest, ":")
		if !ok {
			continue
		}
		var w, h int
		if _, err := fmt.Sscanf(strings.TrimSpace(dims), "%dx%d", &w, &h); err == nil {
			return w, h, nil
		}
	}
	return 0, 0, fmt.Errorf("no resolution in avifdec output")
}

// writeFile atomically writes body to path via a sibling temp + rename.
func writeFile(path string, body []byte) error {
	staged := path + ".tmp"
	if err := os.WriteFile(staged, body, 0o644); err != nil {
		os.Remove(staged)
		return err
	}
	return os.Rename(staged, path)
}

// avifToPNG decodes AVIF bytes to PNG bytes by shelling out to avifdec.
func avifToPNG(body []byte) ([]byte, error) {
	in, err := os.CreateTemp("", "pete-avif-in-*.avif")
	if err != nil {
		return nil, err
	}
	inPath := in.Name()
	defer os.Remove(inPath)
	if _, err := in.Write(body); err != nil {
		in.Close()
		return nil, err
	}
	if err := in.Close(); err != nil {
		return nil, err
	}
	outPath := inPath + ".png"
	defer os.Remove(outPath)
	cmd := exec.Command("avifdec", inPath, outPath)
	if out, err := cmd.CombinedOutput(); err != nil {
		return nil, fmt.Errorf("%w: %s", err, strings.TrimSpace(string(out)))
	}
	return os.ReadFile(outPath)
}