parse5-parser-stream 7.1.2 → 8.1.0

package/README.md

@@ -16,7 +16,7 @@
 <br>
 
 <p align="center">
-  📖 <a href="https://parse5.js.org/modules/parse5_parser_stream.html"><b>Documentation</b></a> 📖
+  📖 <a href="https://parse5.js.org/modules/parse5-parser-stream.html"><b>Documentation</b></a> 📖
 </p>
 
 ---

package/dist/index.d.ts

@@ -1,4 +1,3 @@
-/// <reference types="node" resolution-mode="require"/>
 import { Writable } from 'node:stream';
 import { Parser, type ParserOptions, type TreeAdapterTypeMap, type DefaultTreeAdapterMap } from 'parse5';
 /**
@@ -34,6 +33,26 @@ export declare class ParserStream<T extends TreeAdapterTypeMap = DefaultTreeAdap
     /** The resulting document node. */
     get document(): T['document'];
     getFragment(): T['documentFragment'];
+    /**
+     * Synchronously delivers buffered character tokens to the tree builder without ending the input stream.
+     * Incomplete tags and character references are retained for subsequent writes. This does not resume a
+     * suspended script or process queued writes, and must not be called from a parser callback.
+     *
+     * Flushing splits character-token runs. Their source locations do not extend through subsequently
+     * ignored input, such as CDATA delimiters or the leading newline in a `pre` element, so text-node
+     * source ranges can differ from parsing without flushing.
+     *
+     * @example
+     *
+     * ```js
+     * const stream = new ParserStream();
+     * stream.write('<p>hello');
+     * stream.flushCharacters();
+     * // The paragraph's text is now available in `stream.document`.
+     * stream.end(' world</p>');
+     * ```
+     */
+    flushCharacters(): void;
     /**
      * @param options Parsing options.
      */
@@ -82,4 +101,3 @@ export interface ParserStream<T extends TreeAdapterTypeMap = DefaultTreeAdapterM
      */
     on(event: string, handler: (...args: any[]) => void): this;
 }
-//# sourceMappingURL=index.d.ts.map

package/dist/index.js

@@ -1,6 +1,5 @@
 import { Writable } from 'node:stream';
 import { Parser } from 'parse5';
-/* eslint-disable unicorn/consistent-function-scoping -- The rule seems to be broken here. */
 /**
  * Streaming HTML parser with scripting support.
  * A [writable stream](https://nodejs.org/api/stream.html#stream_class_stream_writable).
@@ -38,6 +37,28 @@ export class ParserStream extends Writable {
     getFragment() {
         return this.parser.getFragment();
     }
+    /**
+     * Synchronously delivers buffered character tokens to the tree builder without ending the input stream.
+     * Incomplete tags and character references are retained for subsequent writes. This does not resume a
+     * suspended script or process queued writes, and must not be called from a parser callback.
+     *
+     * Flushing splits character-token runs. Their source locations do not extend through subsequently
+     * ignored input, such as CDATA delimiters or the leading newline in a `pre` element, so text-node
+     * source ranges can differ from parsing without flushing.
+     *
+     * @example
+     *
+     * ```js
+     * const stream = new ParserStream();
+     * stream.write('<p>hello');
+     * stream.flushCharacters();
+     * // The paragraph's text is now available in `stream.document`.
+     * stream.end(' world</p>');
+     * ```
+     */
+    flushCharacters() {
+        this.parser.tokenizer.flushCharacters();
+    }
     /**
      * @param options Parsing options.
      */
@@ -83,4 +104,3 @@ export class ParserStream extends Writable {
         super.end(chunk || '', encoding, callback);
     }
 }
-//# sourceMappingURL=index.js.map

package/package.json

@@ -2,10 +2,15 @@
     "name": "parse5-parser-stream",
     "type": "module",
     "description": "Streaming HTML parser with scripting support.",
-    "version": "7.1.2",
+    "version": "8.1.0",
     "author": "Ivan Nikulin <ifaaan@gmail.com> (https://github.com/inikulin)",
-    "contributors": "https://github.com/inikulin/parse5/graphs/contributors",
-    "homepage": "https://github.com/inikulin/parse5",
+    "contributors": [
+        "James Garbutt (https://github.com/43081j)",
+        "Felix Boehm (https://github.com/fb55)",
+        "Ivan Nikulin (https://github.com/inikulin)",
+        "Titus (https://github.com/wooorm)"
+    ],
+    "homepage": "https://parse5.js.org",
     "funding": "https://github.com/inikulin/parse5?sponsor=1",
     "keywords": [
         "parse5",
@@ -14,26 +19,27 @@
         "streaming"
     ],
     "license": "MIT",
-    "main": "dist/cjs/index.js",
+    "main": "dist/index.js",
     "module": "dist/index.js",
     "types": "dist/index.d.ts",
     "exports": {
-        "import": "./dist/index.js",
-        "require": "./dist/cjs/index.js"
+        ".": {
+            "default": "./dist/index.js"
+        }
     },
     "dependencies": {
-        "parse5": "^7.0.0"
-    },
-    "scripts": {
-        "build:cjs": "tsc --module CommonJS --target ES6 --outDir dist/cjs && echo '{\"type\":\"commonjs\"}' > dist/cjs/package.json"
+        "parse5": "^8.0.0"
     },
+    "scripts": {},
     "repository": {
         "type": "git",
         "url": "git://github.com/inikulin/parse5.git"
     },
     "files": [
-        "dist/cjs/package.json",
         "dist/**/*.js",
         "dist/**/*.d.ts"
-    ]
+    ],
+    "devDependencies": {
+        "parse5-test-utils": "1.0.0"
+    }
 }

package/dist/cjs/index.d.ts

@@ -1,85 +0,0 @@
-/// <reference types="node" resolution-mode="require"/>
-import { Writable } from 'node:stream';
-import { Parser, type ParserOptions, type TreeAdapterTypeMap, type DefaultTreeAdapterMap } from 'parse5';
-/**
- * Streaming HTML parser with scripting support.
- * A [writable stream](https://nodejs.org/api/stream.html#stream_class_stream_writable).
- *
- * @example
- *
- * ```js
- * const ParserStream = require('parse5-parser-stream');
- * const http = require('http');
- * const { finished } = require('node:stream');
- *
- * // Fetch the page content and obtain it's <head> node
- * http.get('http://inikulin.github.io/parse5/', res => {
- *     const parser = new ParserStream();
- *
- *     finished(parser, () => {
- *         console.log(parser.document.childNodes[1].childNodes[0].tagName); //> 'head'
- *     });
- *
- *     res.pipe(parser);
- * });
- * ```
- *
- */
-export declare class ParserStream<T extends TreeAdapterTypeMap = DefaultTreeAdapterMap> extends Writable {
-    parser: Parser<T>;
-    static getFragmentStream<T extends TreeAdapterTypeMap>(fragmentContext?: T['parentNode'] | null, options?: ParserOptions<T>): ParserStream<T>;
-    private lastChunkWritten;
-    private writeCallback;
-    private pendingHtmlInsertions;
-    /** The resulting document node. */
-    get document(): T['document'];
-    getFragment(): T['documentFragment'];
-    /**
-     * @param options Parsing options.
-     */
-    constructor(options?: ParserOptions<T>, parser?: Parser<T>);
-    _write(chunk: string, _encoding: string, callback: () => void): void;
-    end(chunk?: any, encoding?: any, callback?: any): any;
-}
-export interface ParserStream<T extends TreeAdapterTypeMap = DefaultTreeAdapterMap> {
-    /**
-     * Raised when parser encounters a `<script>` element. If this event has listeners, parsing will be suspended once
-     * it is emitted. So, if `<script>` has the `src` attribute, you can fetch it, execute and then resume parsing just
-     * like browsers do.
-     *
-     * @example
-     *
-     * ```js
-     * const ParserStream = require('parse5-parser-stream');
-     * const http = require('http');
-     *
-     * const parser = new ParserStream();
-     *
-     * parser.on('script', (scriptElement, documentWrite, resume) => {
-     *     const src = scriptElement.attrs.find(({ name }) => name === 'src').value;
-     *
-     *     http.get(src, res => {
-     *         // Fetch the script content, execute it with DOM built around `parser.document` and
-     *         // `document.write` implemented using `documentWrite`.
-     *         ...
-     *         // Then resume parsing.
-     *         resume();
-     *     });
-     * });
-     *
-     * parser.end('<script src="example.com/script.js"></script>');
-     * ```
-     *
-     * @param event Name of the event
-     * @param handler
-     */
-    on(event: 'script', handler: (scriptElement: T['element'], documentWrite: (html: string) => void, resume: () => void) => void): void;
-    /**
-     * Base event handler.
-     *
-     * @param event Name of the event
-     * @param handler Event handler
-     */
-    on(event: string, handler: (...args: any[]) => void): this;
-}
-//# sourceMappingURL=index.d.ts.map

package/dist/cjs/index.js

@@ -1,90 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.ParserStream = void 0;
-const node_stream_1 = require("node:stream");
-const parse5_1 = require("parse5");
-/* eslint-disable unicorn/consistent-function-scoping -- The rule seems to be broken here. */
-/**
- * Streaming HTML parser with scripting support.
- * A [writable stream](https://nodejs.org/api/stream.html#stream_class_stream_writable).
- *
- * @example
- *
- * ```js
- * const ParserStream = require('parse5-parser-stream');
- * const http = require('http');
- * const { finished } = require('node:stream');
- *
- * // Fetch the page content and obtain it's <head> node
- * http.get('http://inikulin.github.io/parse5/', res => {
- *     const parser = new ParserStream();
- *
- *     finished(parser, () => {
- *         console.log(parser.document.childNodes[1].childNodes[0].tagName); //> 'head'
- *     });
- *
- *     res.pipe(parser);
- * });
- * ```
- *
- */
-class ParserStream extends node_stream_1.Writable {
-    static getFragmentStream(fragmentContext, options) {
-        const parser = parse5_1.Parser.getFragmentParser(fragmentContext, options);
-        const stream = new ParserStream(options, parser);
-        return stream;
-    }
-    /** The resulting document node. */
-    get document() {
-        return this.parser.document;
-    }
-    getFragment() {
-        return this.parser.getFragment();
-    }
-    /**
-     * @param options Parsing options.
-     */
-    constructor(options, parser = new parse5_1.Parser(options)) {
-        super({ decodeStrings: false });
-        this.parser = parser;
-        this.lastChunkWritten = false;
-        this.writeCallback = undefined;
-        this.pendingHtmlInsertions = [];
-        const resume = () => {
-            for (let i = this.pendingHtmlInsertions.length - 1; i >= 0; i--) {
-                this.parser.tokenizer.insertHtmlAtCurrentPos(this.pendingHtmlInsertions[i]);
-            }
-            this.pendingHtmlInsertions.length = 0;
-            //NOTE: keep parsing if we don't wait for the next input chunk
-            this.parser.tokenizer.resume(this.writeCallback);
-        };
-        const documentWrite = (html) => {
-            if (!this.parser.stopped) {
-                this.pendingHtmlInsertions.push(html);
-            }
-        };
-        const scriptHandler = (scriptElement) => {
-            if (this.listenerCount('script') > 0) {
-                this.parser.tokenizer.pause();
-                this.emit('script', scriptElement, documentWrite, resume);
-            }
-        };
-        this.parser.scriptHandler = scriptHandler;
-    }
-    //WritableStream implementation
-    _write(chunk, _encoding, callback) {
-        if (typeof chunk !== 'string') {
-            throw new TypeError('Parser can work only with string streams.');
-        }
-        this.writeCallback = callback;
-        this.parser.tokenizer.write(chunk, this.lastChunkWritten, this.writeCallback);
-    }
-    // TODO [engine:node@>=16]: Due to issues with Node < 16, we are overriding `end` instead of `_final`.
-    // eslint-disable-next-line @typescript-eslint/no-explicit-any
-    end(chunk, encoding, callback) {
-        this.lastChunkWritten = true;
-        super.end(chunk || '', encoding, callback);
-    }
-}
-exports.ParserStream = ParserStream;
-//# sourceMappingURL=index.js.map

package/dist/cjs/package.json

@@ -1 +0,0 @@
-{"type":"commonjs"}