Add a utility to just run the thumbnailer (#633)

.gitignore

@@ -10,6 +10,7 @@
 /ipfs
 /dev/conduit-db
 /dev/psql
+/dev/thumb_*
 /vcpkg
 /libheif
 

CHANGELOG.md

@@ -10,6 +10,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 ### Added
 
 * Allow guests to access uploaded media, as per [MSC4189](https://github.com/matrix-org/matrix-spec-proposals/pull/4189).
+* The thumbnailer can now be run independently with the `thumbnailer` binary. See `thumbnailer -help` for details.
 
 ### Changed
 

Dockerfile

@@ -61,6 +61,7 @@ COPY --from=builder \
  /opt/bin/s3_consistency_check \
  /opt/bin/combine_signing_keys \
  /opt/bin/generate_signing_key \
+ /opt/bin/thumbnailer \
  /usr/local/bin/
 
 COPY ./config.sample.yaml /etc/media-repo.yaml.sample

cmd/utilities/thumbnailer/main.go

@@ -0,0 +1,92 @@
+package main
+
+import (
+	"flag"
+	"io"
+	"os"
+
+	"github.com/t2bot/matrix-media-repo/common/config"
+	"github.com/t2bot/matrix-media-repo/common/logging"
+	"github.com/t2bot/matrix-media-repo/common/rcontext"
+	"github.com/t2bot/matrix-media-repo/thumbnailing"
+	"github.com/t2bot/matrix-media-repo/util"
+)
+
+func main() {
+	configPath := flag.String("config", "media-repo.yaml", "The path to the configuration")
+	inFile := flag.String("i", "", "The input file to thumbnail")
+	outFile := flag.String("o", "", "The output file to write the thumbnail to. Note that MMR chooses the output file format and won't inspect the extension type supplied here. Typically, output format is PNG.")
+	targetWidth := flag.Int("w", 512, "The target width of the thumbnail")
+	targetHeight := flag.Int("h", 512, "The target height of the thumbnail")
+	targetMethod := flag.String("m", "scale", "The method to use for resizing. Can be 'scale' or 'crop'")
+	targetAnimated := flag.Bool("a", false, "Whether the thumbnail should be animated (if supported)")
+	forceMime := flag.String("f", "", "Force the mime type of the input file, ignoring the detected mime type")
+	flag.Parse()
+
+	if inFile == nil || *inFile == "" {
+		panic("No input file specified")
+	}
+	if outFile == nil || *outFile == "" {
+		panic("No output file specified")
+	}
+
+	// Override config path with config for Docker users
+	configEnv := os.Getenv("REPO_CONFIG")
+	if configEnv != "" {
+		configPath = &configEnv
+	}
+
+	config.Runtime.IsImportProcess = true // prevents us from creating media by accident
+	config.Path = *configPath
+
+	var err error
+	err = logging.Setup(
+		config.Get().General.LogDirectory,
+		config.Get().General.LogColors,
+		config.Get().General.JsonLogs,
+		config.Get().General.LogLevel,
+	)
+	if err != nil {
+		panic(err)
+	}
+	ctx := rcontext.Initial()
+
+	ctx.Log.WithField("width", *targetWidth).WithField("height", *targetHeight).WithField("method", *targetMethod).WithField("animated", *targetAnimated).Info("Thumbnailing options:")
+
+	// Read source image
+	f, err := os.Open(*inFile)
+	if err != nil {
+		panic(err)
+	}
+	defer f.Close()
+
+	mime, err := util.DetectMimeType(f)
+	if err != nil {
+		panic(err)
+	}
+	ctx.Log.WithField("mime", mime).Info("Detected mime type")
+
+	if forceMime != nil && *forceMime != "" {
+		mime = *forceMime
+		ctx.Log.WithField("mime", mime).Warn("Forcing mime type")
+	}
+
+	ctx.Log.Info("Generating thumbnail")
+	t, err := thumbnailing.GenerateThumbnail(f, mime, *targetWidth, *targetHeight, *targetMethod, *targetAnimated, ctx)
+	if err != nil {
+		panic(err)
+	}
+
+	ctx.Log.WithField("animated", t.Animated).WithField("contentType", t.ContentType).Info("Writing generated thumbnail")
+
+	f2, err := os.Create(*outFile)
+	if err != nil {
+		panic(err)
+	}
+	defer f2.Close()
+	if _, err = io.Copy(f2, t.Reader); err != nil {
+		panic(err)
+	}
+
+	ctx.Log.Info("Done!")
+}

go.mod

@@ -47,6 +47,7 @@ require (
 	github.com/didip/tollbooth/v7 v7.0.2
 	github.com/docker/go-connections v0.5.0
 	github.com/go-redsync/redsync/v4 v4.13.0
+	github.com/h2non/filetype v1.1.3
 	github.com/julienschmidt/httprouter v1.3.0
 	github.com/minio/minio-go/v7 v7.0.82
 	github.com/panjf2000/ants/v2 v2.10.0

go.sum

@@ -170,6 +170,8 @@ github.com/gopherjs/gopherjs v0.0.0-20190430165422-3e4dfb77656c h1:7lF+Vz0LqiRid
 github.com/gopherjs/gopherjs v0.0.0-20190430165422-3e4dfb77656c/go.mod h1:wJfORRmW1u3UXTncJ5qlYoELFm8eSnnEO6hX4iZ3EWY=
 github.com/grpc-ecosystem/grpc-gateway/v2 v2.16.0 h1:YBftPWNWd4WwGqtY2yeZL2ef8rHAxPBD8KFhJpmcqms=
 github.com/grpc-ecosystem/grpc-gateway/v2 v2.16.0/go.mod h1:YN5jB8ie0yfIUg6VvR9Kz84aCaG7AsGZnLjhHbUqwPg=
+github.com/h2non/filetype v1.1.3 h1:FKkx9QbD7HR/zjK1Ia5XiBsq9zdLi5Kf3zGyFTAFkGg=
+github.com/h2non/filetype v1.1.3/go.mod h1:319b3zT68BvV+WRj7cwy856M2ehB3HqNOt6sy1HndBY=
 github.com/hajimehoshi/go-mp3 v0.3.0/go.mod h1:qMJj/CSDxx6CGHiZeCgbiq2DSUkbK0UbtXShQcnfyMM=
 github.com/hajimehoshi/go-mp3 v0.3.4 h1:NUP7pBYH8OguP4diaTZ9wJbUbk3tC0KlfzsEpWmYj68=
 github.com/hajimehoshi/go-mp3 v0.3.4/go.mod h1:fRtZraRFcWb0pu7ok0LqyFhCUrPeMsGRSVop0eemFmo=

util/mime_detect.go

@@ -0,0 +1,47 @@
+package util
+
+import (
+	"io"
+	"net/http"
+	"strings"
+
+	"github.com/h2non/filetype"
+)
+
+func DetectMimeType(r io.ReadSeeker) (string, error) {
+	buf := make([]byte, 512)
+
+	current, err := r.Seek(0, io.SeekCurrent)
+	if err != nil {
+		return "", err
+	}
+	restore := func() error {
+		if _, err2 := r.Seek(current, io.SeekStart); err2 != nil {
+			return err2
+		}
+		return nil
+	}
+
+	if _, err := r.Seek(0, io.SeekStart); err != nil {
+		return "", err
+	}
+	if _, err := r.Read(buf); err != nil {
+		return "", err
+	}
+
+	kind, err := filetype.Match(buf)
+	if err != nil || kind == filetype.Unknown {
+		// Try against http library upon error
+		contentType := http.DetectContentType(buf)
+		contentType = strings.Split(contentType, ";")[0]
+
+		// http should return an octet-stream anyway, but just in case:
+		if contentType == "" {
+			contentType = "application/octet-stream"
+		}
+
+		return contentType, restore()
+	}
+
+	return kind.MIME.Value, restore()
+}