net/http: add more audio/video mime sniffing

Following the spec at
  https://mimesniff.spec.whatwg.org/#matching-an-audio-or-video-type-pattern

Adds signatures for:
+ audio/aiff
+ audio/basic
+ audio/midi
+ audio/mpeg
+ video/avi

Updates the signature for:
+ application/ogg

Also updates the pattern matching algorithm in
  https://mimesniff.spec.whatwg.org/#matching-a-mime-type-pattern
by implementing clause 4 that dictates that the number of bytes in
the pattern must match the number of bytes in the mask.

Fixes #13383

Change-Id: Ie321f392e6570299c17176adf1c75f62f357e1e8
Reviewed-on: https://go-review.googlesource.com/17132
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

2 files changed, 47 insertions, 1 deletions

diff --git a/src/net/http/sniff.go b/src/net/http/sniff.go
index 54986b9956..0d21b44a56 100644
--- a/src/net/http/sniff.go
+++ b/src/net/http/sniff.go
@@ -91,12 +91,41 @@ var sniffSignatures = []sniffSig{
 		ct:   "image/webp",
 	},
 	&exactSig{[]byte("\x00\x00\x01\x00"), "image/vnd.microsoft.icon"},
-	&exactSig{[]byte("\x4F\x67\x67\x53\x00"), "application/ogg"},
 	&maskedSig{
 		mask: []byte("\xFF\xFF\xFF\xFF\x00\x00\x00\x00\xFF\xFF\xFF\xFF"),
 		pat:  []byte("RIFF\x00\x00\x00\x00WAVE"),
 		ct:   "audio/wave",
 	},
+	&maskedSig{
+		mask: []byte("\xFF\xFF\xFF\xFF\x00\x00\x00\x00\xFF\xFF\xFF\xFF"),
+		pat:  []byte("FORM\x00\x00\x00\x00AIFF"),
+		ct:   "audio/aiff",
+	},
+	&maskedSig{
+		mask: []byte("\xFF\xFF\xFF\xFF"),
+		pat:  []byte(".snd"),
+		ct:   "audio/basic",
+	},
+	&maskedSig{
+		mask: []byte("OggS\x00"),
+		pat:  []byte("\x4F\x67\x67\x53\x00"),
+		ct:   "application/ogg",
+	},
+	&maskedSig{
+		mask: []byte("\xFF\xFF\xFF\xFF\xFF\xFF\xFF\xFF"),
+		pat:  []byte("MThd\x00\x00\x00\x06"),
+		ct:   "audio/midi",
+	},
+	&maskedSig{
+		mask: []byte("\xFF\xFF\xFF"),
+		pat:  []byte("ID3"),
+		ct:   "audio/mpeg",
+	},
+	&maskedSig{
+		mask: []byte("\xFF\xFF\xFF\xFF\x00\x00\x00\x00\xFF\xFF\xFF\xFF"),
+		pat:  []byte("RIFF\x00\x00\x00\x00AVI "),
+		ct:   "video/avi",
+	},
 	&exactSig{[]byte("\x1A\x45\xDF\xA3"), "video/webm"},
 	&exactSig{[]byte("\x52\x61\x72\x20\x1A\x07\x00"), "application/x-rar-compressed"},
 	&exactSig{[]byte("\x50\x4B\x03\x04"), "application/zip"},
@@ -126,9 +155,15 @@ type maskedSig struct {
 }
 
 func (m *maskedSig) match(data []byte, firstNonWS int) string {
+	// pattern matching algorithm section 6
+	// https://mimesniff.spec.whatwg.org/#pattern-matching-algorithm
+
 	if m.skipWS {
 		data = data[firstNonWS:]
 	}
+	if len(m.pat) != len(m.mask) {
+		return ""
+	}
 	if len(data) < len(m.mask) {
 		return ""
 	}
diff --git a/src/net/http/sniff_test.go b/src/net/http/sniff_test.go
index e0085516da..ac404bfa72 100644
--- a/src/net/http/sniff_test.go
+++ b/src/net/http/sniff_test.go
@@ -39,7 +39,18 @@ var sniffTests = []struct {
 	{"GIF 87a", []byte(`GIF87a`), "image/gif"},
 	{"GIF 89a", []byte(`GIF89a...`), "image/gif"},
 
+	// Audio types.
+	{"MIDI audio", []byte("MThd\x00\x00\x00\x06\x00\x01"), "audio/midi"},
+	{"MP3 audio/MPEG audio", []byte("ID3\x03\x00\x00\x00\x00\x0f"), "audio/mpeg"},
+	{"WAV audio #1", []byte("RIFFb\xb8\x00\x00WAVEfmt \x12\x00\x00\x00\x06"), "audio/wave"},
+	{"WAV audio #2", []byte("RIFF,\x00\x00\x00WAVEfmt \x12\x00\x00\x00\x06"), "audio/wave"},
+	{"AIFF audio #1", []byte("FORM\x00\x00\x00\x00AIFFCOMM\x00\x00\x00\x12\x00\x01\x00\x00\x57\x55\x00\x10\x40\x0d\xf3\x34"), "audio/aiff"},
+	{"OGG audio", []byte("OggS\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x7e\x46\x00\x00\x00\x00\x00\x00\x1f\xf6\xb4\xfc\x01\x1e\x01\x76\x6f\x72"), "application/ogg"},
+
+	// Video types.
 	{"MP4 video", []byte("\x00\x00\x00\x18ftypmp42\x00\x00\x00\x00mp42isom<\x06t\xbfmdat"), "video/mp4"},
+	{"AVI video #1", []byte("RIFF,O\n\x00AVI LISTÀ"), "video/avi"},
+	{"AVI video #2", []byte("RIFF,\n\x00\x00AVI LISTÀ"), "video/avi"},
 }
 
 func TestDetectContentType(t *testing.T) {