file-type 7.7.0 → 9.0.0

package/index.js

@@ -1,11 +1,11 @@
 'use strict';
-const toBytes = s => Array.from(s).map(c => c.charCodeAt(0));
+const toBytes = s => [...s].map(c => c.charCodeAt(0));
 const xpiZipFilename = toBytes('META-INF/mozilla.rsa');
 const oxmlContentTypes = toBytes('[Content_Types].xml');
 const oxmlRels = toBytes('_rels/.rels');
 
 module.exports = input => {
-	const buf = (input instanceof Uint8Array) ? input : new Uint8Array(input);
+	const buf = input instanceof Uint8Array ? input : new Uint8Array(input);
 
 	if (!(buf && buf.length > 1)) {
 		return null;
@@ -151,41 +151,54 @@ module.exports = input => {
 			};
 		}
 
-		// https://github.com/file/file/blob/master/magic/Magdir/msooxml
-		if (check(oxmlContentTypes, {offset: 30}) || check(oxmlRels, {offset: 30})) {
-			const sliced = buf.subarray(4, 4 + 2000);
-			const nextZipHeaderIndex = arr => arr.findIndex((el, i, arr) => arr[i] === 0x50 && arr[i + 1] === 0x4B && arr[i + 2] === 0x3 && arr[i + 3] === 0x4);
-			const header2Pos = nextZipHeaderIndex(sliced);
-
-			if (header2Pos !== -1) {
-				const slicedAgain = buf.subarray(header2Pos + 8, header2Pos + 8 + 1000);
-				const header3Pos = nextZipHeaderIndex(slicedAgain);
-
-				if (header3Pos !== -1) {
-					const offset = 8 + header2Pos + header3Pos + 30;
-
-					if (checkString('word/', {offset})) {
-						return {
-							ext: 'docx',
-							mime: 'application/vnd.openxmlformats-officedocument.wordprocessingml.document'
-						};
-					}
-
-					if (checkString('ppt/', {offset})) {
-						return {
-							ext: 'pptx',
-							mime: 'application/vnd.openxmlformats-officedocument.presentationml.presentation'
-						};
-					}
-
-					if (checkString('xl/', {offset})) {
-						return {
-							ext: 'xlsx',
-							mime: 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet'
-						};
-					}
+		// The docx, xlsx and pptx file types extend the Office Open XML file format:
+		// https://en.wikipedia.org/wiki/Office_Open_XML_file_formats
+		// We look for:
+		// - one entry named '[Content_Types].xml' or '_rels/.rels',
+		// - one entry indicating specific type of file.
+		// MS Office, OpenOffice and LibreOffice may put the parts in different order, so the check should not rely on it.
+		const findNextZipHeaderIndex = (arr, startAt = 0) => arr.findIndex((el, i, arr) => i >= startAt && arr[i] === 0x50 && arr[i + 1] === 0x4B && arr[i + 2] === 0x3 && arr[i + 3] === 0x4);
+
+		let zipHeaderIndex = 0; // The first zip header was already found at index 0
+		let oxmlFound = false;
+		let type = null;
+
+		do {
+			const offset = zipHeaderIndex + 30;
+
+			if (!oxmlFound) {
+				oxmlFound = (check(oxmlContentTypes, {offset}) || check(oxmlRels, {offset}));
+			}
+
+			if (!type) {
+				if (checkString('word/', {offset})) {
+					type = {
+						ext: 'docx',
+						mime: 'application/vnd.openxmlformats-officedocument.wordprocessingml.document'
+					};
+				} else if (checkString('ppt/', {offset})) {
+					type = {
+						ext: 'pptx',
+						mime: 'application/vnd.openxmlformats-officedocument.presentationml.presentation'
+					};
+				} else if (checkString('xl/', {offset})) {
+					type = {
+						ext: 'xlsx',
+						mime: 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet'
+					};
 				}
 			}
+
+			if (oxmlFound && type) {
+				return type;
+			}
+
+			zipHeaderIndex = findNextZipHeaderIndex(buf, offset);
+		} while (zipHeaderIndex >= 0);
+
+		// No more zip parts available in the buffer, but maybe we are almost certain about the type?
+		if (type) {
+			return type;
 		}
 	}
 
@@ -278,7 +291,7 @@ module.exports = input => {
 
 		if (idPos !== -1) {
 			const docTypePos = idPos + 3;
-			const findDocType = type => Array.from(type).every((c, i) => sliced[docTypePos + i] === c.charCodeAt(0));
+			const findDocType = type => [...type].every((c, i) => sliced[docTypePos + i] === c.charCodeAt(0));
 
 			if (findDocType('matroska')) {
 				return {
@@ -307,14 +320,27 @@ module.exports = input => {
 		};
 	}
 
-	if (
-		check([0x52, 0x49, 0x46, 0x46]) &&
-		check([0x41, 0x56, 0x49], {offset: 8})
-	) {
-		return {
-			ext: 'avi',
-			mime: 'video/x-msvideo'
-		};
+	// RIFF file format which might be AVI, WAV, QCP, etc
+	if (check([0x52, 0x49, 0x46, 0x46])) {
+		if (check([0x41, 0x56, 0x49], {offset: 8})) {
+			return {
+				ext: 'avi',
+				mime: 'video/vnd.avi'
+			};
+		}
+		if (check([0x57, 0x41, 0x56, 0x45], {offset: 8})) {
+			return {
+				ext: 'wav',
+				mime: 'audio/vnd.wave'
+			};
+		}
+		// QLCM, QCP file
+		if (check([0x51, 0x4C, 0x43, 0x4D], {offset: 8})) {
+			return {
+				ext: 'qcp',
+				mime: 'audio/qcelp'
+			};
+		}
 	}
 
 	if (check([0x30, 0x26, 0xB2, 0x75, 0x8E, 0x66, 0xCF, 0x11, 0xA6, 0xD9])) {
@@ -361,15 +387,33 @@ module.exports = input => {
 				mime: 'audio/mpeg'
 			};
 		}
+
+		if (
+			check([0xFF, 0xF8], {offset: start, mask: [0xFF, 0xFC]}) // MPEG 2 layer 0 using ADTS
+		) {
+			return {
+				ext: 'mp2',
+				mime: 'audio/mpeg'
+			};
+		}
+
+		if (
+			check([0xFF, 0xF0], {offset: start, mask: [0xFF, 0xFC]}) // MPEG 4 layer 0 using ADTS
+		) {
+			return {
+				ext: 'mp4',
+				mime: 'audio/mpeg'
+			};
+		}
 	}
 
 	if (
 		check([0x66, 0x74, 0x79, 0x70, 0x4D, 0x34, 0x41], {offset: 4}) ||
 		check([0x4D, 0x34, 0x41, 0x20])
 	) {
-		return {
+		return { // MPEG-4 layer 3 (audio)
 			ext: 'm4a',
-			mime: 'audio/m4a'
+			mime: 'audio/mp4' // RFC 4337
 		};
 	}
 
@@ -437,13 +481,17 @@ module.exports = input => {
 		};
 	}
 
-	if (
-		check([0x52, 0x49, 0x46, 0x46]) &&
-		check([0x57, 0x41, 0x56, 0x45], {offset: 8})
-	) {
+	if (check([0x4D, 0x41, 0x43, 0x20])) { // 'MAC '
+		return {
+			ext: 'ape',
+			mime: 'audio/ape'
+		};
+	}
+
+	if (check([0x77, 0x76, 0x70, 0x6B])) { // 'wvpk'
 		return {
-			ext: 'wav',
-			mime: 'audio/x-wav'
+			ext: 'wv',
+			mime: 'audio/wavpack'
 		};
 	}
 
@@ -528,7 +576,7 @@ module.exports = input => {
 	) {
 		return {
 			ext: 'eot',
-			mime: 'application/octet-stream'
+			mime: 'application/vnd.ms-fontobject'
 		};
 	}
 
@@ -556,7 +604,7 @@ module.exports = input => {
 	if (check([0x00, 0x00, 0x02, 0x00])) {
 		return {
 			ext: 'cur',
-			mime: 'image/x-win-bitmap'
+			mime: 'image/x-icon'
 		};
 	}
 
@@ -770,5 +818,12 @@ module.exports = input => {
 		}
 	}
 
+	if (check([0xAB, 0x4B, 0x54, 0x58, 0x20, 0x31, 0x31, 0xBB, 0x0D, 0x0A, 0x1A, 0x0A])) {
+		return {
+			ext: 'ktx',
+			mime: 'image/ktx'
+		};
+	}
+
 	return null;
 };

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "file-type",
-	"version": "7.7.0",
+	"version": "9.0.0",
 	"description": "Detect the file type of a Buffer/Uint8Array",
 	"license": "MIT",
 	"repository": "sindresorhus/file-type",
@@ -10,7 +10,7 @@
 		"url": "sindresorhus.com"
 	},
 	"engines": {
-		"node": ">=4"
+		"node": ">=6"
 	},
 	"scripts": {
 		"test": "xo && ava"

package/readme.md

@@ -76,7 +76,7 @@ Or `null` when no match.
 
 Type: `Buffer` `Uint8Array`
 
-It only needs the first 4100 bytes.
+It only needs the first 4100 bytes. The exception is detection of `docx`, `pptx`, and `xlsx` which potentially requires reading the whole file.
 
 
 ## Supported file types
@@ -114,6 +114,7 @@ It only needs the first 4100 bytes.
 - [`opus`](https://en.wikipedia.org/wiki/Opus_(audio_format))
 - [`flac`](https://en.wikipedia.org/wiki/FLAC)
 - [`wav`](https://en.wikipedia.org/wiki/WAV)
+- [`qcp`](https://en.wikipedia.org/wiki/QCP)
 - [`amr`](https://en.wikipedia.org/wiki/Adaptive_Multi-Rate_audio_codec)
 - [`pdf`](https://en.wikipedia.org/wiki/Portable_Document_Format)
 - [`epub`](https://en.wikipedia.org/wiki/EPUB)
@@ -161,6 +162,9 @@ It only needs the first 4100 bytes.
 - [`xml`](https://en.wikipedia.org/wiki/XML)
 - [`heic`](http://nokiatech.github.io/heif/technical.html)
 - [`cur`](https://en.wikipedia.org/wiki/ICO_(file_format))
+- [`ktx`](https://www.khronos.org/opengles/sdk/tools/KTX/file_format_spec/)
+- [`ape`](https://en.wikipedia.org/wiki/Monkey%27s_Audio) - Monkey's Audio
+- [`wv`](https://en.wikipedia.org/wiki/WavPack) - WavPack
 
 *SVG isn't included as it requires the whole file to be read, but you can get it [here](https://github.com/sindresorhus/is-svg).*