Hiroyuki Yamanaka

Go net/http DetectContentType

https://golang.org/pkg/net/http/#DetectContentType

DetectContentType implements the algorithm described at https://mimesniff.spec.whatwg.org/ to determine the Content-Type of the given data. It considers at most the first 512 bytes of data. DetectContentType always returns a valid MIME type: if it cannot determine a more specific one, it returns “application/octet-stream”.

https://mimesniff.spec.whatwg.org/
Implement the algorithms described here to determine Content-Type
Judgment is made at the first 512 bytes
If it can not be determined “application/octet-stream” is returned

DetectContentType

func DetectContentType(data []byte) string {
	if len(data) > sniffLen {
		data = data[:sniffLen]
	}

	// Index of the first non-whitespace byte in data.
	firstNonWS := 0
	for ; firstNonWS < len(data) && isWS(data[firstNonWS]); firstNonWS++ {
	}

	for _, sig := range sniffSignatures {
		if ct := sig.match(data, firstNonWS); ct != "" {
			return ct
		}
	}

	return "application/octet-stream" // fallback
}

sniffLen

const sniffLen = 512

isWS

Determine whitespace

func isWS(b byte) bool {
	switch b {
	case '\t', '\n', '\x0c', '\r', ' ':
		return true
	}
	return false
}

sniffSignatures

Contains concrete judgment data

var sniffSignatures = []sniffSig{
	htmlSig("<!DOCTYPE HTML"),
	htmlSig("<HTML"),
	htmlSig("<HEAD"),
	htmlSig("<SCRIPT"),
	htmlSig("<IFRAME"),
	htmlSig("<H1"),
	htmlSig("<DIV"),
	htmlSig("<FONT"),
	htmlSig("<TABLE"),
	htmlSig("<A"),
	htmlSig("<STYLE"),
	htmlSig("<TITLE"),
	htmlSig("<B"),
	htmlSig("<BODY"),
	htmlSig("<BR"),
	htmlSig("<P"),
	htmlSig("<!--"),

	&maskedSig{mask: []byte("\xFF\xFF\xFF\xFF\xFF"), pat: []byte("<?xml"), skipWS: true, ct: "text/xml; charset=utf-8"},

	&exactSig{[]byte("%PDF-"), "application/pdf"},
	&exactSig{[]byte("%!PS-Adobe-"), "application/postscript"},

	// UTF BOMs.
	&maskedSig{mask: []byte("\xFF\xFF\x00\x00"), pat: []byte("\xFE\xFF\x00\x00"), ct: "text/plain; charset=utf-16be"},
	&maskedSig{mask: []byte("\xFF\xFF\x00\x00"), pat: []byte("\xFF\xFE\x00\x00"), ct: "text/plain; charset=utf-16le"},
	&maskedSig{mask: []byte("\xFF\xFF\xFF\x00"), pat: []byte("\xEF\xBB\xBF\x00"), ct: "text/plain; charset=utf-8"},

	&exactSig{[]byte("GIF87a"), "image/gif"},
	&exactSig{[]byte("GIF89a"), "image/gif"},
	&exactSig{[]byte("\x89\x50\x4E\x47\x0D\x0A\x1A\x0A"), "image/png"},
	&exactSig{[]byte("\xFF\xD8\xFF"), "image/jpeg"},
	&exactSig{[]byte("BM"), "image/bmp"},
	&maskedSig{
		mask: []byte("\xFF\xFF\xFF\xFF\x00\x00\x00\x00\xFF\xFF\xFF\xFF\xFF\xFF"),
		pat:  []byte("RIFF\x00\x00\x00\x00WEBPVP"),
		ct:   "image/webp",
	},
	&exactSig{[]byte("\x00\x00\x01\x00"), "image/vnd.microsoft.icon"},

	&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("\xFF\xFF\xFF\xFF\xFF"),
		pat:  []byte("OggS\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",
	},

	// Fonts
	&maskedSig{
		// 34 NULL bytes followed by the string "LP"
		pat: []byte("\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x4C\x50"),
		// 34 NULL bytes followed by \xF\xF
		mask: []byte("\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xFF\xFF"),
		ct:   "application/vnd.ms-fontobject",
	},
	&exactSig{[]byte("\x00\x01\x00\x00"), "font/ttf"},
	&exactSig{[]byte("OTTO"), "font/otf"},
	&exactSig{[]byte("ttcf"), "font/collection"},
	&exactSig{[]byte("wOFF"), "font/woff"},
	&exactSig{[]byte("wOF2"), "font/woff2"},

	&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"},
	&exactSig{[]byte("\x1F\x8B\x08"), "application/x-gzip"},

	&exactSig{[]byte("\x00\x61\x73\x6D"), "application/wasm"},

	mp4Sig{},

	textSig{}, // should be last
}