webpack/lib/optimize/MergeDuplicateChunksPlugin.js

Packs CommonJs/AMD modules for the browser. Allows to split your codebase into multiple bundles, which can be loaded on demand. Support loaders to preprocess files, i.e. json, jsx, es7, css, less, ... and your custom stuff.

/*
	MIT License http://www.opensource.org/licenses/mit-license.php
	Author Tobias Koppers @sokra
*/

"use strict";

const { STAGE_BASIC } = require("../OptimizationStages");
const { runtimeEqual } = require("../util/runtime");

/** @typedef {import("../Compiler")} Compiler */

class MergeDuplicateChunksPlugin {
	/**
	 * @param {Compiler} compiler the compiler
	 * @returns {void}
	 */
	apply(compiler) {
		compiler.hooks.compilation.tap(
			"MergeDuplicateChunksPlugin",
			compilation => {
				compilation.hooks.optimizeChunks.tap(
					{
						name: "MergeDuplicateChunksPlugin",
						stage: STAGE_BASIC
					},
					chunks => {
						const { chunkGraph, moduleGraph } = compilation;

						// remember already tested chunks for performance
						const notDuplicates = new Set();

						// for each chunk
						for (const chunk of chunks) {
							// track a Set of all chunk that could be duplicates
							let possibleDuplicates;
							for (const module of chunkGraph.getChunkModulesIterable(chunk)) {
								if (possibleDuplicates === undefined) {
									// when possibleDuplicates is not yet set,
									// create a new Set from chunks of the current module
									// including only chunks with the same number of modules
									for (const dup of chunkGraph.getModuleChunksIterable(
										module
									)) {
										if (
											dup !== chunk &&
											chunkGraph.getNumberOfChunkModules(chunk) ===
												chunkGraph.getNumberOfChunkModules(dup) &&
											!notDuplicates.has(dup)
										) {
											// delay allocating the new Set until here, reduce memory pressure
											if (possibleDuplicates === undefined) {
												possibleDuplicates = new Set();
											}
											possibleDuplicates.add(dup);
										}
									}
									// when no chunk is possible we can break here
									if (possibleDuplicates === undefined) break;
								} else {
									// validate existing possible duplicates
									for (const dup of possibleDuplicates) {
										// remove possible duplicate when module is not contained
										if (!chunkGraph.isModuleInChunk(module, dup)) {
											possibleDuplicates.delete(dup);
										}
									}
									// when all chunks has been removed we can break here
									if (possibleDuplicates.size === 0) break;
								}
							}

							// when we found duplicates
							if (
								possibleDuplicates !== undefined &&
								possibleDuplicates.size > 0
							) {
								outer: for (const otherChunk of possibleDuplicates) {
									if (otherChunk.hasRuntime() !== chunk.hasRuntime()) continue;
									if (chunkGraph.getNumberOfEntryModules(chunk) > 0) continue;
									if (chunkGraph.getNumberOfEntryModules(otherChunk) > 0)
										continue;
									if (!runtimeEqual(chunk.runtime, otherChunk.runtime)) {
										for (const module of chunkGraph.getChunkModulesIterable(
											chunk
										)) {
											const exportsInfo = moduleGraph.getExportsInfo(module);
											if (
												!exportsInfo.isEquallyUsed(
													chunk.runtime,
													otherChunk.runtime
												)
											) {
												continue outer;
											}
										}
									}
									// merge them
									if (chunkGraph.canChunksBeIntegrated(chunk, otherChunk)) {
										chunkGraph.integrateChunks(chunk, otherChunk);
										compilation.chunks.delete(otherChunk);
									}
								}
							}

							// don't check already processed chunks twice
							notDuplicates.add(chunk);
						}
					}
				);
			}
		);
	}
}
module.exports = MergeDuplicateChunksPlugin;