Detect a fetched image's format from its bytes once (closes #319)

magic.DetectFormat returns an imagerequest.Format. ValidateMagicBytes returns the format it checked against the upstream Content-Type. The cache reads a cached source's format from its file's first bytes, as it reads a variant's. The proxy hands the format to Process. Process refuses a format that is not a source format, and input libvips would load with another loader, such as an SVG passing as AVIF. Format.MIMEType declares each MIME string once, and the fetcher, the cache's variant read and the Accept matching use it. magic.MIMEType, magic.ImageFormat, MIMEToImageFormat, magic's SVG detection, detectFormat, formatFromString, FormatToMIME and Result.InputFormat are gone. Model: opus-5-5
2026-10-26 21:21:39 +01:00
parent ac850823f4
commit b73bbde46e
37 changed files with 570 additions and 589 deletions
@@ -1,9 +1,10 @@
 package magic

 import (
-	"bytes"
 	"errors"
 	"testing"
+
+	"sneak.berlin/go/pixa/internal/imagerequest"
 )

 // Shared test fixture strings.
@@ -12,9 +13,6 @@ const (
 	testMIMEJPEG       = "image/jpeg"
 	testMIMEJPEGParams = "image/jpeg; charset=utf-8"
 	testMIMEPNG        = "image/png"
-	testMIMEWebP       = "image/webp"
-	testMIMEGIF        = "image/gif"
-	testMIMEAVIF       = "image/avif"
 )

 // pad appends zero bytes so data is comfortably above MinMagicBytes.
@@ -36,37 +34,22 @@ func TestDetectFormat(t *testing.T) {
 	avis := pad(0x00, 0x00, 0x00, 0x1C, 0x66, 0x74, 0x79, 0x70, 0x61, 0x76, 0x69, 0x73)

 	tests := []struct {
-		name     string
-		data     []byte
-		wantMIME MIMEType
-		wantErr  error
+		name       string
+		data       []byte
+		wantFormat imagerequest.Format
+		wantErr    error
 	}{
-		{name: "JPEG", data: jpeg, wantMIME: MIMETypeJPEG},
-		{name: "PNG", data: png, wantMIME: MIMETypePNG},
-		{name: "GIF87a", data: gif87a, wantMIME: MIMETypeGIF},
-		{name: "GIF89a", data: gif89a, wantMIME: MIMETypeGIF},
-		{name: "WebP", data: webp, wantMIME: MIMETypeWebP},
-		{name: "AVIF", data: avif, wantMIME: MIMETypeAVIF},
-		{name: "AVIF sequence", data: avis, wantMIME: MIMETypeAVIF},
+		{name: "JPEG", data: jpeg, wantFormat: imagerequest.FormatJPEG},
+		{name: "PNG", data: png, wantFormat: imagerequest.FormatPNG},
+		{name: "GIF87a", data: gif87a, wantFormat: imagerequest.FormatGIF},
+		{name: "GIF89a", data: gif89a, wantFormat: imagerequest.FormatGIF},
+		{name: "WebP", data: webp, wantFormat: imagerequest.FormatWebP},
+		{name: "AVIF", data: avif, wantFormat: imagerequest.FormatAVIF},
+		{name: "AVIF sequence", data: avis, wantFormat: imagerequest.FormatAVIF},
 		{
-			name:     "SVG with XML declaration",
-			data:     []byte(`<?xml version="1.0"?><svg></svg>`),
-			wantMIME: MIMETypeSVG,
-		},
-		{
-			name:     "SVG without declaration",
-			data:     []byte(`<svg xmlns="http://www.w3.org/2000/svg"></svg>`),
-			wantMIME: MIMETypeSVG,
-		},
-		{
-			name:     "SVG with whitespace",
-			data:     []byte(`   <?xml version="1.0"?><svg></svg>`),
-			wantMIME: MIMETypeSVG,
-		},
-		{
-			name:     "SVG with BOM",
-			data:     append([]byte{0xEF, 0xBB, 0xBF}, []byte(`<svg></svg>`)...),
-			wantMIME: MIMETypeSVG,
+			name:    "SVG",
+			data:    []byte(`<svg xmlns="http://www.w3.org/2000/svg"></svg>`),
+			wantErr: ErrUnknownFormat,
 		},
 		{
 			name:    "unknown format",
@@ -88,8 +71,8 @@ func TestDetectFormat(t *testing.T) {
 				return
 			}

-			if got != tt.wantMIME {
-				t.Errorf("DetectFormat() = %v, want %v", got, tt.wantMIME)
+			if got != tt.wantFormat {
+				t.Errorf("DetectFormat() = %v, want %v", got, tt.wantFormat)
 			}
 		})
 	}
@@ -105,24 +88,28 @@ func TestValidateMagicBytes(t *testing.T) {
 		name         string
 		data         []byte
 		declaredType string
+		wantFormat   imagerequest.Format
 		wantErr      error
 	}{
 		{
 			name:         "matching JPEG",
 			data:         jpegData,
 			declaredType: testMIMEJPEG,
+			wantFormat:   imagerequest.FormatJPEG,
 			wantErr:      nil,
 		},
 		{
 			name:         "matching JPEG with params",
 			data:         jpegData,
 			declaredType: testMIMEJPEGParams,
+			wantFormat:   imagerequest.FormatJPEG,
 			wantErr:      nil,
 		},
 		{
 			name:         "matching PNG",
 			data:         pngData,
 			declaredType: testMIMEPNG,
+			wantFormat:   imagerequest.FormatPNG,
 			wantErr:      nil,
 		},
 		{
@@ -143,45 +130,14 @@ func TestValidateMagicBytes(t *testing.T) {
 		t.Run(tt.name, func(t *testing.T) {
 			t.Parallel()

-			err := ValidateMagicBytes(tt.data, tt.declaredType)
+			got, err := ValidateMagicBytes(tt.data, tt.declaredType)

 			if !errors.Is(err, tt.wantErr) {
 				t.Errorf("ValidateMagicBytes() error = %v, wantErr %v", err, tt.wantErr)
 			}
-		})
-	}
-}
-
-func TestMIMEToImageFormat(t *testing.T) {
-	t.Parallel()
-
-	tests := []struct {
-		mimeType   string
-		wantFormat ImageFormat
-		wantOk     bool
-	}{
-		{testMIMEJPEG, FormatJPEG, true},
-		{testMIMEPNG, FormatPNG, true},
-		{testMIMEWebP, FormatWebP, true},
-		{testMIMEGIF, FormatGIF, true},
-		{testMIMEAVIF, FormatAVIF, true},
-		{"image/svg+xml", "", false}, // SVG doesn't convert to ImageFormat
-		{"image/tiff", "", false},
-		{"text/plain", "", false},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.mimeType, func(t *testing.T) {
-			t.Parallel()
-
-			got, ok := MIMEToImageFormat(tt.mimeType)
-
-			if ok != tt.wantOk {
-				t.Errorf("MIMEToImageFormat(%q) ok = %v, want %v", tt.mimeType, ok, tt.wantOk)
-			}

 			if got != tt.wantFormat {
-				t.Errorf("MIMEToImageFormat(%q) = %v, want %v", tt.mimeType, got, tt.wantFormat)
+				t.Errorf("ValidateMagicBytes() = %q, want %q", got, tt.wantFormat)
 			}
 		})
 	}
@@ -214,109 +170,6 @@ func TestNormalizeMIMEType(t *testing.T) {
 	}
 }

-func TestDetectSVG(t *testing.T) {
-	t.Parallel()
-
-	tests := []struct {
-		name string
-		data string
-		want bool
-	}{
-		{"xml declaration", `<?xml version="1.0"?><svg></svg>`, true},
-		{"svg element", `<svg xmlns="http://www.w3.org/2000/svg"></svg>`, true},
-		{
-			"doctype",
-			`<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" ` +
-				`"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">`,
-			true,
-		},
-		{"with whitespace", `
-  <?xml version="1.0"?><svg></svg>`, true},
-		{"uppercase", `<SVG></SVG>`, true},
-		{"not svg", `<html></html>`, false},
-		{"random text", `hello world`, false},
-		{testNameEmpty, ``, false},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			t.Parallel()
-
-			got := detectSVG([]byte(tt.data))
-
-			if got != tt.want {
-				t.Errorf("detectSVG(%q) = %v, want %v", tt.data, got, tt.want)
-			}
-		})
-	}
-}
-
-func TestSkipBOM(t *testing.T) {
-	t.Parallel()
-
-	tests := []struct {
-		name string
-		data []byte
-		want []byte
-	}{
-		{"with BOM", []byte{0xEF, 0xBB, 0xBF, 'h', 'e', 'l', 'l', 'o'}, []byte("hello")},
-		{"without BOM", []byte("hello"), []byte("hello")},
-		{testNameEmpty, []byte{}, []byte{}},
-		{"only BOM", []byte{0xEF, 0xBB, 0xBF}, []byte{}},
-		{"partial BOM", []byte{0xEF, 0xBB}, []byte{0xEF, 0xBB}},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			t.Parallel()
-
-			got := skipBOM(tt.data)
-
-			if !bytes.Equal(got, tt.want) {
-				t.Errorf("skipBOM() = %v, want %v", got, tt.want)
-			}
-		})
-	}
-}
-
-func TestRealWorldSVGPatterns(t *testing.T) {
-	t.Parallel()
-
-	// Test various real-world SVG patterns
-	svgPatterns := []string{
-		`<?xml version="1.0" encoding="UTF-8"?>
-<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 100 100">
-  <circle cx="50" cy="50" r="40"/>
-</svg>`,
-		`<svg viewBox="0 0 24 24" fill="none" xmlns="http://www.w3.org/2000/svg">
-  <path d="M12 2L2 7l10 5 10-5-10-5z"/>
-</svg>`,
-		`<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" ` +
-			`"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">` + `
-<svg xmlns="http://www.w3.org/2000/svg">
-</svg>`,
-	}
-
-	for i, pattern := range svgPatterns {
-		data := []byte(pattern)
-		if len(data) < MinMagicBytes {
-			// Pad short SVGs for detection
-			data = append(data, make([]byte, MinMagicBytes-len(data))...)
-		}
-
-		got, err := DetectFormat(data)
-		if err != nil {
-			t.Errorf("Pattern %d: DetectFormat() error = %v", i, err)
-
-			continue
-		}
-
-		if got != MIMETypeSVG {
-			t.Errorf("Pattern %d: DetectFormat() = %v, want %v", i, got, MIMETypeSVG)
-		}
-	}
-}
-
 func TestDetectFormatRIFFNotWebP(t *testing.T) {
 	t.Parallel()