@file-type/xml 0.3.0 → 0.4.1

package/README.md

@@ -17,15 +17,16 @@ The following example shows how add the XML detector to [file-type](https://gith
 import {NodeFileTypeParser} from 'file-type';
 import {detectXml} from '@file-type/xml';
 
-const parser = new NodeFileTypeParser({customDetectors: [detectXml]});
+const parser = new FileTypeParser({customDetectors: [detectXml]});
 const fileType = await parser.fromFile('example.kml');
 console.log(fileType);
 ```
 
-You can also use the XML detector outside file-type:
+You can also use the XML detector outside [file-type](https://github.com/sindresorhus/file-type):
 ```js
 import {XmlTextDetector} from '@file-type/xml';
 
+const xmlTextDetector = new XmlTextDetector();
 xmlTextDetector.write('<svg xmlns="http://www.w3.org/2000/svg"><path fill="#00CD9F"/></svg>');
 const fileType = xmlTextDetector.fileType;
 console.log(JSON.stringify(fileType)); // Outputs: {"ext":"svg","mime":"image/svg+xml"}

package/lib/index.d.ts

@@ -1,12 +1,4 @@
-import { ITokenizer } from 'strtok3';
-/**
- * Temporary, should be taken from `file-type`
- */
-interface FileTypeResult {
-    ext: string;
-    mime: string;
-}
-export type Detector = (tokenizer: ITokenizer, fileType?: FileTypeResult) => Promise<FileTypeResult | undefined>;
+import type { FileTypeResult, Detector } from 'file-type';
 interface IXmlTextDetectorOptions {
     fullScan?: boolean;
 }

package/lib/index.js

@@ -14,32 +14,23 @@ function isXml(array) {
     if (startsWith(array, [60, 63, 120, 109, 108, 32])) {
         return { xml: true, encoding: 'utf-8', offset: 0 };
     }
-    else if (startsWith(array, [0xEF, 0xBB, 0xBF, 60, 63, 120, 109, 108, 32])) { // UTF-8 BOM
+    if (startsWith(array, [0xEF, 0xBB, 0xBF, 60, 63, 120, 109, 108, 32])) { // UTF-8 BOM
         return { xml: true, encoding: 'utf-8', offset: 3 };
     }
-    else if (startsWith(array, [0xFE, 0xFF, 0, 60, 0, 63, 0, 120, 0, 109, 0, 108, 0, 32])) {
+    if (startsWith(array, [0xFE, 0xFF, 0, 60, 0, 63, 0, 120, 0, 109, 0, 108, 0, 32])) {
         return { xml: true, encoding: 'utf-16be', offset: 2 };
     }
-    else if (startsWith(array, [0xFF, 0xFE, 60, 0, 63, 0, 120, 0, 109, 0, 108, 0, 32, 0])) {
+    if (startsWith(array, [0xFF, 0xFE, 60, 0, 63, 0, 120, 0, 109, 0, 108, 0, 32, 0])) {
         return { xml: true, encoding: 'utf-16le', offset: 2 };
     }
-    else if (startsWith(array, [0, 60, 0, 63, 0, 120, 0, 109, 0, 108, 0, 32])) {
+    if (startsWith(array, [0, 60, 0, 63, 0, 120, 0, 109, 0, 108, 0, 32])) {
         return { xml: true, encoding: 'utf-16be', offset: 0 };
     }
-    else if (startsWith(array, [60, 0, 63, 0, 120, 0, 109, 0, 108, 0, 32, 0])) {
+    if (startsWith(array, [60, 0, 63, 0, 120, 0, 109, 0, 108, 0, 32, 0])) {
         return { xml: true, encoding: 'utf-16le', offset: 0 };
     }
     return { xml: false };
 }
-function extractNsElement(node) {
-    const parts = node.name.split(':');
-    if (parts.length === 1) {
-        return { name: parts[0], ns: node.attributes['xmlns'] };
-    }
-    else if (parts.length === 2) {
-        return { name: parts[1], ns: node.attributes[`xmlns:${parts[0]}`] };
-    }
-}
 /**
  * Maps the root element namespace to corresponding file-type
  */
@@ -85,7 +76,7 @@ export class XmlTextDetector {
         this.options = options ?? {};
         this.firstTag = true;
         this.onEnd = false;
-        this.parser = sax.parser(true);
+        this.parser = sax.parser(true, { xmlns: true });
         this.nesting = 0;
         this.parser.onerror = e => {
             if (e.message.startsWith('Invalid character entity')) { // Allow entity reference
@@ -100,14 +91,13 @@ export class XmlTextDetector {
                 return;
             }
             this.firstTag = false;
-            const nsNode = extractNsElement(node);
-            if (nsNode?.ns) {
+            if (node.uri) {
                 // Resolve file-type boot root element namespace
-                this.fileType = namespaceMapping[nsNode.ns.toLowerCase()];
+                this.fileType = namespaceMapping[node.uri];
             }
-            else if (nsNode && nsNode.name) {
+            else if (node.name) {
                 // Fall back on element name if there is no namespace
-                this.fileType = rootNameMapping[nsNode.name.toLowerCase()];
+                this.fileType = rootNameMapping[node.name.toLowerCase()];
             }
             if (this.fileType && !this.options.fullScan) {
                 this.onEnd = true;
@@ -128,27 +118,30 @@ export class XmlTextDetector {
         return this.nesting === 0;
     }
 }
-export const detectXml = async (tokenizer) => {
-    const buffer = new Uint8Array(512);
-    // Increase sample size from 12 to 256.
-    await tokenizer.peekBuffer(buffer, { length: 128, mayBeLess: true });
-    const xmlDetection = isXml(buffer);
-    if (xmlDetection.xml) {
-        await tokenizer.ignore(xmlDetection.offset);
-        const xmlTextDetector = new XmlTextDetector();
-        const textDecoder = new TextDecoder(xmlDetection.encoding);
-        do {
-            const len = await tokenizer.readBuffer(buffer, { mayBeLess: true });
-            const portion = buffer.subarray(0, len);
-            const text = textDecoder.decode(portion);
-            xmlTextDetector.write(text);
-            if (len < buffer.length) {
-                xmlTextDetector.close();
-            }
-        } while (!xmlTextDetector.onEnd);
-        return xmlTextDetector.fileType ?? {
-            ext: 'xml',
-            mime: 'application/xml'
-        };
+export const detectXml = {
+    id: 'xml',
+    detect: async (tokenizer) => {
+        const buffer = new Uint8Array(512);
+        // Increase sample size from 12 to 256.
+        await tokenizer.peekBuffer(buffer, { length: 128, mayBeLess: true });
+        const xmlDetection = isXml(buffer);
+        if (xmlDetection.xml) {
+            await tokenizer.ignore(xmlDetection.offset);
+            const xmlTextDetector = new XmlTextDetector();
+            const textDecoder = new TextDecoder(xmlDetection.encoding);
+            do {
+                const len = await tokenizer.readBuffer(buffer, { mayBeLess: true });
+                const portion = buffer.subarray(0, len);
+                const text = textDecoder.decode(portion);
+                xmlTextDetector.write(text);
+                if (len < buffer.length) {
+                    xmlTextDetector.close();
+                }
+            } while (!xmlTextDetector.onEnd);
+            return xmlTextDetector.fileType ?? {
+                ext: 'xml',
+                mime: 'application/xml'
+            };
+        }
     }
 };

package/package.json

@@ -1,7 +1,7 @@
 {
 	"name": "@file-type/xml",
-	"version": "0.3.0",
-	"description": "XML detection plugin",
+	"version": "0.4.1",
+	"description": "XML detection plugin for file-type",
 	"type": "module",
 	"types": "./lib/index.d.ts",
 	"exports": "./lib/index.js",
@@ -9,6 +9,7 @@
 		"clean": "del-cli 'lib/**/*.js' 'lib/**/*.js.map' 'lib/**/*.d.ts' 'src/**/*.d.ts'",
 		"compile-src": "tsc -p lib",
 		"compile": "yarn run  compile-src",
+		"lint-ts": "biome check",
 		"build": "yarn run clean && yarn compile",
 		"test": "mocha"
 	},
@@ -29,15 +30,16 @@
 	],
 	"dependencies": {
 		"sax": "^1.4.1",
-		"strtok3": "^10.0.1"
+		"strtok3": "^10.2.1"
 	},
 	"devDependencies": {
+		"@biomejs/biome": "^1.9.4",
 		"@types/sax": "^1.2.7",
 		"chai": "^5.1.2",
 		"del-cli": "^6.0.0",
-		"file-type": "^19.6.0",
-		"mocha": "^11.0.1",
-		"typescript": "^5.7.2"
+		"file-type": "^20.1.0",
+		"mocha": "^11.1.0",
+		"typescript": "^5.7.3"
 	},
 	"files": [
 		"lib/**/*.js",
@@ -53,5 +55,5 @@
 		"url": "git+https://github.com/Borewit/file-type-xml.git"
 	},
 	"license": "MIT",
-	"packageManager": "yarn@4.5.3"
+	"packageManager": "yarn@4.6.0"
 }