Files
majordomo/provider/llamaswap/mediautil_test.go
T
steveandClaude Fable 5 e98493bcfb
CI / Tidy (pull_request) Successful in 9m23s
CI / Build & Test (pull_request) Successful in 10m24s
Adversarial Review (Gadfly) / review (pull_request) Successful in 15m37s
feat: media expansion surfaces — edit mask, upscale, background removal, interpolation, diarization, meshgen (ADR-0020)
- imagegen.EditRequest.Mask -> sd-server img2img inpainting (white=repaint)
- imagegen.Upscaler + BackgroundRemover, videogen.Interpolator,
  audio.DiarizationModel: new optional provider-minted surfaces
- NEW meshgen leaf package (image->3D, glb/stl/obj)
- provider/llamaswap: all five via the /upstream/<model>/<path> passthrough
  (upstreamPath helper, shared one-file multipart builder); binary success
  bodies validated (non-image, non-video, JSON-mesh rejection); diarization
  pins output=json (vtt/srt drop speaker labels)

Co-Authored-By: Claude Fable 5 <[email protected]>
2026-07-12 23:52:34 -04:00

254 lines
8.2 KiB
Go

package llamaswap
import (
"context"
"encoding/base64"
"errors"
"net/http"
"net/http/httptest"
"strings"
"testing"
"gitea.stevedudenhoeffer.com/steve/majordomo/imagegen"
"gitea.stevedudenhoeffer.com/steve/majordomo/llm"
"gitea.stevedudenhoeffer.com/steve/majordomo/videogen"
)
func pngFixture(t *testing.T) []byte {
t.Helper()
raw, err := base64.StdEncoding.DecodeString(onePixelPNG)
if err != nil {
t.Fatalf("decode fixture: %v", err)
}
return raw
}
func TestUpscale(t *testing.T) {
png := pngFixture(t)
var gotPath, gotScale, gotFilename string
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
gotPath = r.URL.Path
if err := r.ParseMultipartForm(1 << 20); err != nil {
t.Errorf("parse multipart: %v", err)
}
gotScale = r.FormValue("scale")
if f, hdr, err := r.FormFile("file"); err == nil {
gotFilename = hdr.Filename
f.Close()
} else {
t.Errorf("file part: %v", err)
}
w.Header().Set("Content-Type", "image/png")
_, _ = w.Write(png)
}))
defer srv.Close()
p := New(WithBaseURL(srv.URL), WithHTTPClient(srv.Client()))
up, err := p.UpscaleModel("mediautils")
if err != nil {
t.Fatalf("UpscaleModel: %v", err)
}
res, err := up.Upscale(context.Background(),
imagegen.UpscaleRequest{Image: imagegen.Image{MIME: "image/png", Data: png}},
imagegen.WithUpscaleScale(2))
if err != nil {
t.Fatalf("Upscale: %v", err)
}
if gotPath != "/upstream/mediautils/v1/upscale" {
t.Errorf("path = %q", gotPath)
}
if gotScale != "2" || gotFilename != "image.png" {
t.Errorf("scale/filename = %q/%q", gotScale, gotFilename)
}
if len(res.Images) != 1 || res.Images[0].MIME != "image/png" {
t.Fatalf("images = %+v", res.Images)
}
}
func TestUpscaleRejectsBadArgs(t *testing.T) {
p := New(WithBaseURL("http://unused"))
up, _ := p.UpscaleModel("mediautils")
if _, err := up.Upscale(context.Background(), imagegen.UpscaleRequest{}); !errors.Is(err, llm.ErrUnsupported) {
t.Errorf("no image: err = %v, want ErrUnsupported", err)
}
if _, err := up.Upscale(context.Background(),
imagegen.UpscaleRequest{Image: imagegen.Image{Data: []byte{1}}, Scale: 3}); !errors.Is(err, llm.ErrUnsupported) {
t.Errorf("scale 3: err = %v, want ErrUnsupported", err)
}
}
func TestUpscaleRejectsNonImageResponse(t *testing.T) {
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
w.Header().Set("Content-Type", "text/html")
_, _ = w.Write([]byte("<html>proxy error page</html>"))
}))
defer srv.Close()
p := New(WithBaseURL(srv.URL), WithHTTPClient(srv.Client()))
up, _ := p.UpscaleModel("mediautils")
_, err := up.Upscale(context.Background(),
imagegen.UpscaleRequest{Image: imagegen.Image{Data: pngFixture(t)}})
var apiErr *llm.APIError
if !errors.As(err, &apiErr) {
t.Fatalf("err = %v, want APIError for non-image body", err)
}
}
func TestRemoveBackground(t *testing.T) {
png := pngFixture(t)
var gotPath, gotNet, gotOM string
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
gotPath = r.URL.Path
if err := r.ParseMultipartForm(1 << 20); err != nil {
t.Errorf("parse multipart: %v", err)
}
gotNet = r.FormValue("model")
gotOM = r.FormValue("om")
w.Header().Set("Content-Type", "image/png")
_, _ = w.Write(png)
}))
defer srv.Close()
p := New(WithBaseURL(srv.URL), WithHTTPClient(srv.Client()))
br, err := p.BackgroundRemovalModel("rembg")
if err != nil {
t.Fatalf("BackgroundRemovalModel: %v", err)
}
res, err := br.RemoveBackground(context.Background(),
imagegen.BackgroundRemovalRequest{Image: imagegen.Image{Data: png}},
imagegen.WithBackgroundNet("birefnet-general"), imagegen.WithOnlyMask())
if err != nil {
t.Fatalf("RemoveBackground: %v", err)
}
if gotPath != "/upstream/rembg/api/remove" {
t.Errorf("path = %q", gotPath)
}
if gotNet != "birefnet-general" || gotOM != "true" {
t.Errorf("net/om = %q/%q", gotNet, gotOM)
}
if len(res.Images) != 1 {
t.Fatalf("images = %+v", res.Images)
}
}
func TestRemoveBackgroundOmitsUnsetFields(t *testing.T) {
png := pngFixture(t)
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
if err := r.ParseMultipartForm(1 << 20); err != nil {
t.Errorf("parse multipart: %v", err)
}
if _, ok := r.MultipartForm.Value["model"]; ok {
t.Error("model field sent for default net; want omitted")
}
if _, ok := r.MultipartForm.Value["om"]; ok {
t.Error("om field sent for cutout mode; want omitted")
}
w.Header().Set("Content-Type", "image/png")
_, _ = w.Write(png)
}))
defer srv.Close()
p := New(WithBaseURL(srv.URL), WithHTTPClient(srv.Client()))
br, _ := p.BackgroundRemovalModel("rembg")
if _, err := br.RemoveBackground(context.Background(),
imagegen.BackgroundRemovalRequest{Image: imagegen.Image{Data: png}}); err != nil {
t.Fatalf("RemoveBackground: %v", err)
}
}
// mp4Fixture is a minimal ftyp box so http.DetectContentType sniffs video/mp4.
func mp4Fixture() []byte {
return []byte{0, 0, 0, 24, 'f', 't', 'y', 'p', 'i', 's', 'o', 'm',
0, 0, 2, 0, 'i', 's', 'o', 'm', 'i', 's', 'o', '2'}
}
func TestInterpolate(t *testing.T) {
var gotPath, gotMulti, gotMode, gotTargetFPS string
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
gotPath = r.URL.Path
if err := r.ParseMultipartForm(1 << 20); err != nil {
t.Errorf("parse multipart: %v", err)
}
gotMulti = r.FormValue("multi")
gotMode = r.FormValue("mode")
gotTargetFPS = r.FormValue("target_fps")
w.Header().Set("Content-Type", "video/mp4")
_, _ = w.Write(mp4Fixture())
}))
defer srv.Close()
p := New(WithBaseURL(srv.URL), WithHTTPClient(srv.Client()))
ip, err := p.InterpolatorModel("mediautils")
if err != nil {
t.Fatalf("InterpolatorModel: %v", err)
}
res, err := ip.Interpolate(context.Background(),
videogen.InterpolateRequest{Video: videogen.Video{Data: mp4Fixture(), MIME: "video/mp4"}},
videogen.WithInterpolateMulti(2), videogen.WithTargetFPS(60))
if err != nil {
t.Fatalf("Interpolate: %v", err)
}
if gotPath != "/upstream/mediautils/v1/interpolate" {
t.Errorf("path = %q", gotPath)
}
if gotMulti != "2" || gotMode != "" || gotTargetFPS != "60" {
t.Errorf("multi/mode/target_fps = %q/%q/%q", gotMulti, gotMode, gotTargetFPS)
}
if res.Video.MIME != "video/mp4" || len(res.Video.Data) == 0 {
t.Fatalf("video = %q/%d bytes", res.Video.MIME, len(res.Video.Data))
}
}
func TestInterpolateSlowMoSetsModeAndDropsTargetFPS(t *testing.T) {
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
if err := r.ParseMultipartForm(1 << 20); err != nil {
t.Errorf("parse multipart: %v", err)
}
if got := r.FormValue("mode"); got != "slowmo" {
t.Errorf("mode = %q, want slowmo", got)
}
if _, ok := r.MultipartForm.Value["target_fps"]; ok {
t.Error("target_fps sent in slowmo mode; want omitted")
}
w.Header().Set("Content-Type", "video/mp4")
_, _ = w.Write(mp4Fixture())
}))
defer srv.Close()
p := New(WithBaseURL(srv.URL), WithHTTPClient(srv.Client()))
ip, _ := p.InterpolatorModel("mediautils")
_, err := ip.Interpolate(context.Background(),
videogen.InterpolateRequest{Video: videogen.Video{Data: mp4Fixture()}, TargetFPS: 60},
videogen.WithSlowMo())
if err != nil {
t.Fatalf("Interpolate: %v", err)
}
}
func TestInterpolateRejectsBadArgs(t *testing.T) {
p := New(WithBaseURL("http://unused"))
ip, _ := p.InterpolatorModel("mediautils")
if _, err := ip.Interpolate(context.Background(), videogen.InterpolateRequest{}); !errors.Is(err, llm.ErrUnsupported) {
t.Errorf("no video: err = %v, want ErrUnsupported", err)
}
if _, err := ip.Interpolate(context.Background(),
videogen.InterpolateRequest{Video: videogen.Video{Data: []byte{1}}, Multi: 3}); !errors.Is(err, llm.ErrUnsupported) {
t.Errorf("multi 3: err = %v, want ErrUnsupported", err)
}
}
func TestUpstreamPathRejectsSeparators(t *testing.T) {
for _, bad := range []string{"", "a/b", "a?b", "a#b"} {
if _, err := upstreamPath(bad, "/x"); err == nil {
t.Errorf("upstreamPath(%q) succeeded; want error", bad)
}
}
got, err := upstreamPath("rembg", "api/remove")
if err != nil || got != "/upstream/rembg/api/remove" {
t.Errorf("upstreamPath = %q, %v", got, err)
}
if !strings.HasPrefix(got, "/upstream/") {
t.Errorf("path prefix wrong: %q", got)
}
}