JS implementation of tgz->zip

packages/yarnpkg-core/sources/Cache.ts

@@ -381,7 +381,7 @@ export class Cache {
       ? () => makeMockPackage()
       : () => new ZipFS(cachePath, {baseFs, readOnly: true});
 
-    const lazyFs = new LazyFS<PortablePath>(() => miscUtils.prettifySyncErrors(() => {
+    const lazyFs = new LazyFS(() => miscUtils.prettifySyncErrors(() => {
       return zipFs = zipFsBuilder();
     }, message => {
       return `Failed to open the cache entry for ${structUtils.prettyLocator(this.configuration, locator)}: ${message}`;

packages/yarnpkg-core/sources/FastZip.ts

@@ -0,0 +1,108 @@
+import {PortablePath} from '@yarnpkg/fslib';
+import fs             from 'fs';
+
+import * as tgzUtils  from './tgzUtils';
+
+export const TarFileTypeMap = {
+  0: `File`,
+  5: `Directory`,
+} as const;
+
+export type TarFileType = (typeof TarFileTypeMap)[keyof typeof TarFileTypeMap];
+
+export function parseTarEntries(source: Buffer) {
+  const entries: Array<{
+    type: TarFileType;
+    path: PortablePath;
+    offset: number;
+    size: number;
+    mode: number;
+  }> = [];
+
+  let offset = 0;
+
+  while (offset < source.byteLength) {
+    const path = source.toString(`utf8`, offset, offset + 100).replace(/\0.*$/, ``);
+    if (!path) {
+      offset += 512;
+      continue;
+    }
+
+    const mode = parseInt(source.toString(`utf8`, offset + 100, offset + 108), 8);
+    const size = parseInt(source.toString(`utf8`, offset + 124, offset + 136), 8);
+    const type = (source[offset + 156] || `0`);
+    const prefix = source.toString(`utf8`, offset + 345, offset + 500).toString().replace(/\0.*$/, ``);
+
+    entries.push({
+      path: `${prefix}${path}` as PortablePath,
+      type: (TarFileTypeMap as any)[type] ?? `Unknown`,
+      offset: offset + 512,
+      size,
+      mode,
+    });
+
+    offset += 512 + Math.ceil(size / 512) * 512;
+  }
+
+  return entries;
+}
+
+async function testConvertToZip3rdParty(tgz: Buffer) {
+  const start2 = Date.now();
+
+  await tgzUtils.convertToZip3rdParty(tgz, {
+    compressionLevel: 0,
+  });
+
+  console.log(`convertToZip3rdParty: ${Date.now() - start2}ms`);
+}
+
+async function testConvertToZipCustomJs(tgz: Buffer) {
+  const start2 = Date.now();
+
+  await tgzUtils.convertToZipCustomJs(tgz, {
+    compressionLevel: 0,
+  });
+
+  console.log(`testConvertToZipCustomJs: ${Date.now() - start2}ms`);
+}
+
+function testConvertToZipWasm(tgz: Buffer) {
+  const start2 = Date.now();
+
+  tgzUtils.convertToZipWasm(tgz, {
+    compressionLevel: 0,
+  });
+
+  console.log(`testConvertToZipWasm: ${Date.now() - start2}ms`);
+}
+
+async function test() {
+  console.log(`Let's test the performance of the various tgz -> zip converters!`);
+
+  const files = fs.readdirSync(`.`).filter(name => name.endsWith(`.tgz`));
+  for (const f of files) {
+    const tgz = fs.readFileSync(f);
+
+    console.log();
+    console.log(`=== ${f} ${tgz.byteLength / 1024} KiB ===`);
+    console.log();
+
+    for (let t = 0; t < 10; ++t)
+      await testConvertToZip3rdParty(tgz);
+
+    console.log();
+
+    for (let t = 0; t < 10; ++t)
+      await testConvertToZipCustomJs(tgz);
+
+    console.log();
+
+    for (let t = 0; t < 10; ++t) {
+      testConvertToZipWasm(tgz);
+    }
+  }
+}
+
+//if (require.main === module)
+//  test();

packages/yarnpkg-core/sources/crc32.ts

@@ -0,0 +1,63 @@
+var CRC_TABLE = new Int32Array([
+  0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
+  0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
+  0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
+  0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
+  0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
+  0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
+  0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
+  0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
+  0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
+  0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
+  0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
+  0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
+  0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
+  0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
+  0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
+  0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
+  0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
+  0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
+  0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
+  0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
+  0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
+  0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
+  0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
+  0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
+  0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
+  0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
+  0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
+  0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
+  0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
+  0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
+  0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
+  0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
+  0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
+  0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
+  0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
+  0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
+  0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
+  0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
+  0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
+  0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
+  0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
+  0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
+  0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
+  0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
+  0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
+  0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
+  0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
+  0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
+  0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
+  0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
+  0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
+  0x2d02ef8d,
+]);
+
+export function crc32(buf: Buffer, offset: number, length: number) {
+  let crc = ~0;
+
+  for (let n = offset, N = offset + length; n < N; ++n)
+    crc = CRC_TABLE[(crc ^ buf[n]) & 0xff] ^ (crc >>> 8);
+
+  return (crc ^ -1) >>> 0;
+}

packages/yarnpkg-core/sources/tgzUtils.ts

@@ -1,11 +1,16 @@
-import {FakeFS, PortablePath, NodeFS, ppath, xfs, npath, constants} from '@yarnpkg/fslib';
-import {ZipCompression, ZipFS}                                      from '@yarnpkg/libzip';
-import {PassThrough, Readable}                                      from 'stream';
-import tar                                                          from 'tar';
+import {FakeFS, PortablePath, NodeFS, ppath, xfs, npath, constants, statUtils} from '@yarnpkg/fslib';
+import {ZipCompression, ZipFS}                                                 from '@yarnpkg/libzip';
+import {PassThrough, Readable}                                                 from 'stream';
+import tar                                                                     from 'tar';
+import {promisify}                                                             from 'util';
+import {gunzip, gunzipSync}                                                    from 'zlib';
 
-import {WorkerPool}                                                 from './WorkerPool';
-import * as miscUtils                                               from './miscUtils';
-import {getContent as getZipWorkerSource, ConvertToZipPayload}      from './worker-zip';
+import {parseTarEntries}                                                       from './FastZip';
+import {WorkerPool}                                                            from './WorkerPool';
+import * as miscUtils                                                          from './miscUtils';
+import {getContent as getZipWorkerSource, ConvertToZipPayload}                 from './worker-zip';
+
+const gunzipP = promisify(gunzip);
 
 interface MakeArchiveFromDirectoryOptions {
   baseFs?: FakeFS<PortablePath>;
@@ -32,22 +37,92 @@ export async function makeArchiveFromDirectory(source: PortablePath, {baseFs = n
 }
 
 export interface ExtractBufferOptions {
+  algorithm?: `3rdParty` | `CustomJs` | `Wasm`;
   compressionLevel?: ZipCompression;
+  destination?: PortablePath;
+  enableWorkerPool?: boolean;
   prefixPath?: PortablePath;
   stripComponents?: number;
 }
 
 let workerPool: WorkerPool<ConvertToZipPayload, PortablePath> | null;
 
+const defaultEnableWorkers = process.env.CONVERT_TO_ZIP_WORKERS === `1`;
+const defaultConvertAlgorithm = process.env.CONVERT_TO_ZIP_ALGORITHM ?? `3rdParty`;
+
+const convertAlgorithms: Record<string, (tgz: Buffer, opts: ExtractBufferOptions) => Promise<ZipFS>> = {
+  [`3rdParty`]: convertToZip3rdParty,
+  [`CustomJs`]: convertToZipCustomJs,
+  [`Wasm`]: convertToZipWasm,
+};
+
 export async function convertToZip(tgz: Buffer, opts: ExtractBufferOptions) {
-  const tmpFolder = await xfs.mktempPromise();
-  const tmpFile = ppath.join(tmpFolder, `archive.zip`);
+  if (opts.enableWorkerPool ?? defaultEnableWorkers) {
+    return convertToZipWorker(tgz, opts);
+  } else {
+    return convertToZipNoWorker(tgz, opts);
+  }
+}
+
+export async function convertToZipWorker(tgz: Buffer, opts: ExtractBufferOptions) {
+  const destination = opts.destination ?? ppath.join(await xfs.mktempPromise(), `archive.zip`);
 
   workerPool ||= new WorkerPool(getZipWorkerSource());
 
-  await workerPool.run({tmpFile, tgz, opts});
+  await workerPool.run({tgz, opts: {...opts, destination}});
+
+  return new ZipFS(destination, {level: opts.compressionLevel});
+}
+
+export async function convertToZipNoWorker(tgz: Buffer, opts: ExtractBufferOptions) {
+  const algorithm = opts.algorithm ?? defaultConvertAlgorithm;
+
+  return await convertAlgorithms[algorithm](tgz, opts);
+}
+
+export async function convertToZip3rdParty(tgz: Buffer, opts: ExtractBufferOptions) {
+  const destination = opts.destination ?? ppath.join(await xfs.mktempPromise(), `archive.zip`);
+
+  const zipFs = new ZipFS(destination, {create: true, level: opts.compressionLevel, stats: statUtils.makeDefaultStats()});
 
-  return new ZipFS(tmpFile, {level: opts.compressionLevel});
+  // Buffers sent through Node are turned into regular Uint8Arrays
+  const tgzBuffer = Buffer.from(tgz.buffer, tgz.byteOffset, tgz.byteLength);
+  await extractArchiveTo(0, tgzBuffer, zipFs, opts);
+
+  zipFs.saveAndClose();
+
+  return new ZipFS(destination, {level: opts.compressionLevel});
+}
+
+export async function convertToZipCustomJs(tgz: Buffer, opts: ExtractBufferOptions) {
+  const destination = opts.destination ?? ppath.join(await xfs.mktempPromise(), `archive.zip`);
+
+  const zipFs = new ZipFS(destination, {create: true, level: opts.compressionLevel, stats: statUtils.makeDefaultStats()});
+
+  // Buffers sent through Node are turned into regular Uint8Arrays
+  const tgzBuffer = Buffer.from(tgz.buffer, tgz.byteOffset, tgz.byteLength);
+  await extractArchiveTo(1, tgzBuffer, zipFs, opts);
+
+  zipFs.saveAndClose();
+
+  return new ZipFS(destination, {level: opts.compressionLevel});
+}
+
+export async function convertToZipWasm(tgz: Buffer, opts: ExtractBufferOptions) {
+  const zip = new ZipFS({
+    type: `tar`,
+    buffer: await gunzipP(tgz),
+    skipComponents: opts.stripComponents,
+    prefixPath: opts.prefixPath,
+  }, {
+    level: opts.compressionLevel,
+    stats: statUtils.makeDefaultStats(),
+  });
+
+  const destination = opts.destination ?? ppath.join(await xfs.mktempPromise(), `archive.zip`);
+  await xfs.writeFilePromise(destination, zip.getBufferAndClose());
+
+  return new ZipFS(destination, {level: opts.compressionLevel});
 }
 
 async function * parseTar(tgz: Buffer) {
@@ -79,8 +154,8 @@ async function * parseTar(tgz: Buffer) {
   }
 }
 
-export async function extractArchiveTo<T extends FakeFS<PortablePath>>(tgz: Buffer, targetFs: T, {stripComponents = 0, prefixPath = PortablePath.dot}: ExtractBufferOptions = {}): Promise<T> {
-  function ignore(entry: tar.ReadEntry) {
+export async function extractArchiveTo<T extends FakeFS<PortablePath>>(v: number, tgz: Buffer, targetFs: T, {stripComponents = 0, prefixPath = PortablePath.dot}: ExtractBufferOptions = {}): Promise<T> {
+  function ignore(entry: {path: string}) {
     // Disallow absolute paths; might be malicious (ex: /etc/passwd)
     if (entry.path[0] === `/`)
       return true;
@@ -97,7 +172,11 @@ export async function extractArchiveTo<T extends FakeFS<PortablePath>>(tgz: Buff
     return false;
   }
 
-  for await (const entry of parseTar(tgz)) {
+  const entries = v === 0
+    ? parseTar(tgz)
+    : parseTarEntries(gunzipSync(tgz));
+
+  for await (const entry of entries) {
     if (ignore(entry))
       continue;
 

packages/yarnpkg-core/sources/worker-zip/Worker.ts

@@ -1,25 +1,20 @@
-import {PortablePath, statUtils}                from '@yarnpkg/fslib';
-import {ZipFS}                                  from '@yarnpkg/libzip';
-import {parentPort}                             from 'worker_threads';
+import {parentPort}  from 'worker_threads';
 
-import {extractArchiveTo, ExtractBufferOptions} from '../tgzUtils';
+import * as tgzUtils from '../tgzUtils';
 
-export type ConvertToZipPayload = {tmpFile: PortablePath, tgz: Buffer | Uint8Array, opts: ExtractBufferOptions};
+export type ConvertToZipPayload = {
+  tgz: Buffer | Uint8Array;
+  opts: tgzUtils.ExtractBufferOptions;
+};
 
 if (!parentPort)
   throw new Error(`Assertion failed: Expected parentPort to be set`);
 
 parentPort.on(`message`, async (data: ConvertToZipPayload) => {
-  const {opts, tgz, tmpFile} = data;
-  const {compressionLevel, ...bufferOpts} = opts;
+  const {opts, tgz} = data;
 
-  const zipFs = new ZipFS(tmpFile, {create: true, level: compressionLevel, stats: statUtils.makeDefaultStats()});
-
-  // Buffers sent through Node are turned into regular Uint8Arrays
   const tgzBuffer = Buffer.from(tgz.buffer, tgz.byteOffset, tgz.byteLength);
-  await extractArchiveTo(tgzBuffer, zipFs, bufferOpts);
-
-  zipFs.saveAndClose();
+  await tgzUtils.convertToZipNoWorker(tgzBuffer, opts);
 
-  parentPort!.postMessage(data.tmpFile);
+  parentPort!.postMessage(true);
 });

packages/yarnpkg-libzip/artifacts/exported.json

@@ -50,5 +50,6 @@
     "_zipstruct_stat_comp_size",
     "_zipstruct_stat_comp_method",
 
-    "_zip_ext_count_symlinks"
+    "_zip_ext_count_symlinks",
+    "_zip_ext_import_tar"
 ]