[feature] support processing of (many) more media types (#3090)

* initial work replacing our media decoding / encoding pipeline with ffprobe + ffmpeg

* specify the video codec to use when generating static image from emoji

* update go-storage library (fixes incompatibility after updating go-iotools)

* maintain image aspect ratio when generating a thumbnail for it

* update readme to show go-ffmpreg

* fix a bunch of media tests, move filesize checking to callers of media manager for more flexibility

* remove extra debug from error message

* fix up incorrect function signatures

* update PutFile to just use regular file copy, as changes are file is on separate partition

* fix remaining tests, remove some unneeded tests now we're working with ffmpeg/ffprobe

* update more tests, add more code comments

* add utilities to generate processed emoji / media outputs

* fix remaining tests

* add test for opus media file, add license header to utility cmds

* limit the number of concurrently available ffmpeg / ffprobe instances

* reduce number of instances

* further reduce number of instances

* fix envparsing test with configuration variables

* update docs and configuration with new media-{local,remote}-max-size variables

internal/media/ffmpeg/cache.go

@@ -0,0 +1,46 @@
+// GoToSocial
+// Copyright (C) GoToSocial Authors admin@gotosocial.org
+// SPDX-License-Identifier: AGPL-3.0-or-later
+//
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU Affero General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+//
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU Affero General Public License for more details.
+//
+// You should have received a copy of the GNU Affero General Public License
+// along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+package ffmpeg
+
+import (
+	"os"
+
+	"github.com/tetratelabs/wazero"
+)
+
+// shared WASM compilation cache.
+var cache wazero.CompilationCache
+
+func initCache() {
+	if cache != nil {
+		return
+	}
+
+	if dir := os.Getenv("WAZERO_COMPILATION_CACHE"); dir != "" {
+		var err error
+
+		// Use on-filesystem compilation cache given by env.
+		cache, err = wazero.NewCompilationCacheWithDir(dir)
+		if err != nil {
+			panic(err)
+		}
+	} else {
+		// Use in-memory compilation cache.
+		cache = wazero.NewCompilationCache()
+	}
+}

internal/media/ffmpeg/ffmpeg.go

@@ -0,0 +1,92 @@
+// GoToSocial
+// Copyright (C) GoToSocial Authors admin@gotosocial.org
+// SPDX-License-Identifier: AGPL-3.0-or-later
+//
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU Affero General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+//
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU Affero General Public License for more details.
+//
+// You should have received a copy of the GNU Affero General Public License
+// along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+package ffmpeg
+
+import (
+	"context"
+
+	ffmpeglib "codeberg.org/gruf/go-ffmpreg/embed/ffmpeg"
+	"codeberg.org/gruf/go-ffmpreg/util"
+	"codeberg.org/gruf/go-ffmpreg/wasm"
+
+	"github.com/tetratelabs/wazero"
+	"github.com/tetratelabs/wazero/api"
+	"github.com/tetratelabs/wazero/imports/wasi_snapshot_preview1"
+)
+
+// InitFfmpeg initializes the ffmpeg WebAssembly instance pool,
+// with given maximum limiting the number of concurrent instances.
+func InitFfmpeg(ctx context.Context, max int) error {
+	initCache() // ensure compilation cache initialized
+	return ffmpegPool.Init(ctx, max)
+}
+
+// Ffmpeg runs the given arguments with an instance of ffmpeg.
+func Ffmpeg(ctx context.Context, args wasm.Args) (uint32, error) {
+	return ffmpegPool.Run(ctx, args)
+}
+
+var ffmpegPool = wasmInstancePool{
+	inst: wasm.Instantiator{
+
+		// WASM module name.
+		Module: "ffmpeg",
+
+		// Per-instance WebAssembly runtime (with shared cache).
+		Runtime: func(ctx context.Context) wazero.Runtime {
+
+			// Prepare config with cache.
+			cfg := wazero.NewRuntimeConfig()
+			cfg = cfg.WithCoreFeatures(ffmpeglib.CoreFeatures)
+			cfg = cfg.WithCompilationCache(cache)
+
+			// Instantiate runtime with our config.
+			rt := wazero.NewRuntimeWithConfig(ctx, cfg)
+
+			// Prepare default "env" host module.
+			env := rt.NewHostModuleBuilder("env")
+			env = env.NewFunctionBuilder().
+				WithGoModuleFunction(
+					api.GoModuleFunc(util.Wasm_Tempnam),
+					[]api.ValueType{api.ValueTypeI32, api.ValueTypeI32},
+					[]api.ValueType{api.ValueTypeI32},
+				).
+				Export("tempnam")
+
+			// Instantiate "env" module in our runtime.
+			_, err := env.Instantiate(context.Background())
+			if err != nil {
+				panic(err)
+			}
+
+			// Instantiate the wasi snapshot preview 1 in runtime.
+			_, err = wasi_snapshot_preview1.Instantiate(ctx, rt)
+			if err != nil {
+				panic(err)
+			}
+
+			return rt
+		},
+
+		// Per-run module configuration.
+		Config: wazero.NewModuleConfig,
+
+		// Embedded WASM.
+		Source: ffmpeglib.B,
+	},
+}

internal/media/ffmpeg/ffprobe.go

@@ -0,0 +1,92 @@
+// GoToSocial
+// Copyright (C) GoToSocial Authors admin@gotosocial.org
+// SPDX-License-Identifier: AGPL-3.0-or-later
+//
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU Affero General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+//
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU Affero General Public License for more details.
+//
+// You should have received a copy of the GNU Affero General Public License
+// along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+package ffmpeg
+
+import (
+	"context"
+
+	ffprobelib "codeberg.org/gruf/go-ffmpreg/embed/ffprobe"
+	"codeberg.org/gruf/go-ffmpreg/util"
+	"codeberg.org/gruf/go-ffmpreg/wasm"
+
+	"github.com/tetratelabs/wazero"
+	"github.com/tetratelabs/wazero/api"
+	"github.com/tetratelabs/wazero/imports/wasi_snapshot_preview1"
+)
+
+// InitFfprobe initializes the ffprobe WebAssembly instance pool,
+// with given maximum limiting the number of concurrent instances.
+func InitFfprobe(ctx context.Context, max int) error {
+	initCache() // ensure compilation cache initialized
+	return ffprobePool.Init(ctx, max)
+}
+
+// Ffprobe runs the given arguments with an instance of ffprobe.
+func Ffprobe(ctx context.Context, args wasm.Args) (uint32, error) {
+	return ffprobePool.Run(ctx, args)
+}
+
+var ffprobePool = wasmInstancePool{
+	inst: wasm.Instantiator{
+
+		// WASM module name.
+		Module: "ffprobe",
+
+		// Per-instance WebAssembly runtime (with shared cache).
+		Runtime: func(ctx context.Context) wazero.Runtime {
+
+			// Prepare config with cache.
+			cfg := wazero.NewRuntimeConfig()
+			cfg = cfg.WithCoreFeatures(ffprobelib.CoreFeatures)
+			cfg = cfg.WithCompilationCache(cache)
+
+			// Instantiate runtime with our config.
+			rt := wazero.NewRuntimeWithConfig(ctx, cfg)
+
+			// Prepare default "env" host module.
+			env := rt.NewHostModuleBuilder("env")
+			env = env.NewFunctionBuilder().
+				WithGoModuleFunction(
+					api.GoModuleFunc(util.Wasm_Tempnam),
+					[]api.ValueType{api.ValueTypeI32, api.ValueTypeI32},
+					[]api.ValueType{api.ValueTypeI32},
+				).
+				Export("tempnam")
+
+			// Instantiate "env" module in our runtime.
+			_, err := env.Instantiate(context.Background())
+			if err != nil {
+				panic(err)
+			}
+
+			// Instantiate the wasi snapshot preview 1 in runtime.
+			_, err = wasi_snapshot_preview1.Instantiate(ctx, rt)
+			if err != nil {
+				panic(err)
+			}
+
+			return rt
+		},
+
+		// Per-run module configuration.
+		Config: wazero.NewModuleConfig,
+
+		// Embedded WASM.
+		Source: ffprobelib.B,
+	},
+}

internal/media/ffmpeg/pool.go

@@ -0,0 +1,75 @@
+// GoToSocial
+// Copyright (C) GoToSocial Authors admin@gotosocial.org
+// SPDX-License-Identifier: AGPL-3.0-or-later
+//
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU Affero General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+//
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU Affero General Public License for more details.
+//
+// You should have received a copy of the GNU Affero General Public License
+// along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+package ffmpeg
+
+import (
+	"context"
+
+	"codeberg.org/gruf/go-ffmpreg/wasm"
+)
+
+// wasmInstancePool wraps a wasm.Instantiator{} and a
+// channel of wasm.Instance{}s to provide a concurrency
+// safe pool of WebAssembly module instances capable of
+// compiling new instances on-the-fly, with a predetermined
+// maximum number of concurrent instances at any one time.
+type wasmInstancePool struct {
+	inst wasm.Instantiator
+	pool chan *wasm.Instance
+}
+
+func (p *wasmInstancePool) Init(ctx context.Context, sz int) error {
+	p.pool = make(chan *wasm.Instance, sz)
+	for i := 0; i < sz; i++ {
+		inst, err := p.inst.New(ctx)
+		if err != nil {
+			return err
+		}
+		p.pool <- inst
+	}
+	return nil
+}
+
+func (p *wasmInstancePool) Run(ctx context.Context, args wasm.Args) (uint32, error) {
+	var inst *wasm.Instance
+
+	select {
+	// Context canceled.
+	case <-ctx.Done():
+		return 0, ctx.Err()
+
+	// Acquire instance.
+	case inst = <-p.pool:
+
+		// Ensure instance is
+		// ready for running.
+		if inst.IsClosed() {
+			var err error
+			inst, err = p.inst.New(ctx)
+			if err != nil {
+				return 0, err
+			}
+		}
+	}
+
+	// Release instance to pool on end.
+	defer func() { p.pool <- inst }()
+
+	// Pass args to instance.
+	return inst.Run(ctx, args)
+}