Add detector for .tar.gz (gunzipped tarball file) (#763)

Author: Borewit

core.js

@@ -4,7 +4,7 @@ Primary entry point, Node.js specific entry point is index.js
 
 import * as Token from 'token-types';
 import * as strtok3 from 'strtok3/core';
-import {ZipHandler} from '@tokenizer/inflate';
+import {ZipHandler, GzipHandler} from '@tokenizer/inflate';
 import {getUintBE} from 'uint8array-extras';
 import {
 	stringToBytes,
@@ -238,7 +238,7 @@ export class FileTypeParser {
 	}
 
 	async fromStream(stream) {
-		const tokenizer = await strtok3.fromWebStream(stream, this.tokenizerOptions);
+		const tokenizer = strtok3.fromWebStream(stream, this.tokenizerOptions);
 		try {
 			return await this.fromTokenizer(tokenizer);
 		} finally {
@@ -408,6 +408,21 @@ export class FileTypeParser {
 		}
 
 		if (this.check([0x1F, 0x8B, 0x8])) {
+			const gzipHandler = new GzipHandler(tokenizer);
+
+			const stream = gzipHandler.inflate();
+			try {
+				const compressedFileType = await this.fromStream(stream);
+				if (compressedFileType && compressedFileType.ext === 'tar') {
+					return {
+						ext: 'tar.gz',
+						mime: 'application/gzip',
+					};
+				}
+			} finally {
+				await stream.cancel();
+			}
+
 			return {
 				ext: 'gz',
 				mime: 'application/gzip',

fixture/fixture.gz

@@ -243,11 +243,12 @@
 		"jar",
 		"rm",
 		"ppsm",
-		"ppsx"
+		"ppsx",
+		"tar.gz"
 	],
 	"dependencies": {
-		"@tokenizer/inflate": "^0.2.7",
-		"strtok3": "^10.2.2",
+		"@tokenizer/inflate": "^0.3.1",
+		"strtok3": "^10.3.1",
 		"token-types": "^6.0.0",
 		"uint8array-extras": "^1.4.0"
 	},

package.json

@@ -607,7 +607,8 @@ abortController.abort(); // Abort file-type reading from the Blob stream.
 - [`sqlite`](https://www.sqlite.org/fileformat2.html) - SQLite file
 - [`stl`](https://en.wikipedia.org/wiki/STL_(file_format)) - Standard Tesselated Geometry File Format (ASCII only)
 - [`swf`](https://en.wikipedia.org/wiki/SWF) - Adobe Flash Player file
-- [`tar`](https://en.wikipedia.org/wiki/Tar_(computing)#File_format) - Tarball archive file
+- [`tar`](https://en.wikipedia.org/wiki/Tar_(computing)#File_format) - Tape archive or tarball
+- [`tar.gz`](https://en.wikipedia.org/wiki/Gzip) - Gzipped tape archive (tarball)
 - [`tif`](https://en.wikipedia.org/wiki/Tagged_Image_File_Format) - Tagged Image file
 - [`ttc`](https://en.wikipedia.org/wiki/TrueType#TrueType_Collection) - TrueType Collection font
 - [`ttf`](https://en.wikipedia.org/wiki/TrueType) - TrueType font

readme.md

@@ -174,6 +174,7 @@ export const extensions = [
 	'rm',
 	'ppsm',
 	'ppsx',
+	'tar.gz',
 ];
 
 export const mimeTypes = [

supported.js

@@ -113,7 +113,7 @@ const names = {
 		'fixture-bali',
 	],
 	gz: [
-		'fixture.tar',
+		'fixture',
 	],
 	xz: [
 		'fixture.tar',
@@ -279,6 +279,9 @@ const names = {
 	ppsm: [
 		'fixture',
 	],
+	'tar.gz': [
+		'fixture',
+	],
 };
 
 // Define an entry here only if the file type has potential