Merge branch 'main' into no-wasm-simd

Author: Uzlopak

.github/workflows/nodejs.yml

@@ -134,6 +134,13 @@ jobs:
           NODE_V8_COVERAGE: ${{ inputs.codecov == true && './coverage/tmp' || '' }} 
           UNDICI_NO_WASM_SIMD: ${{ inputs['no-wasm-simd'] }}
 
+      - name: Test infra
+        run: npm run test:infra
+        id: test-infra
+        env:
+          CI: true
+          NODE_V8_COVERAGE: ${{ inputs.codecov == true && './coverage/tmp' || '' }} 
+
       - name: Test subresource-integrity
         run: npm run test:subresource-integrity
         id: test-subresource-integrity
@@ -166,14 +173,6 @@ jobs:
           NODE_V8_COVERAGE: ${{ inputs.codecov == true && './coverage/tmp' || '' }} 
           UNDICI_NO_WASM_SIMD: ${{ inputs['no-wasm-simd'] }}
 
-      - name: Test h2
-        run: npm run test:h2
-        id: test-h2
-        env:
-          CI: true
-          NODE_V8_COVERAGE: ${{ inputs.codecov == true && './coverage/tmp' || '' }} 
-          UNDICI_NO_WASM_SIMD: ${{ inputs['no-wasm-simd'] }}
-      
       - name: Test jest
         run: npm run test:jest
         id: test-jest

lib/encoding/index.js

@@ -0,0 +1,33 @@
+'use strict'
+
+const textDecoder = new TextDecoder()
+
+/**
+ * @see https://encoding.spec.whatwg.org/#utf-8-decode
+ * @param {Uint8Array} buffer
+ */
+function utf8DecodeBytes (buffer) {
+  if (buffer.length === 0) {
+    return ''
+  }
+
+  // 1. Let buffer be the result of peeking three bytes from
+  //    ioQueue, converted to a byte sequence.
+
+  // 2. If buffer is 0xEF 0xBB 0xBF, then read three
+  //    bytes from ioQueue. (Do nothing with those bytes.)
+  if (buffer[0] === 0xEF && buffer[1] === 0xBB && buffer[2] === 0xBF) {
+    buffer = buffer.subarray(3)
+  }
+
+  // 3. Process a queue with an instance of UTF-8’s
+  //    decoder, ioQueue, output, and "replacement".
+  const output = textDecoder.decode(buffer)
+
+  // 4. Return output.
+  return output
+}
+
+module.exports = {
+  utf8DecodeBytes
+}

lib/web/cookies/parse.js

@@ -1,8 +1,8 @@
 'use strict'
 
+const { collectASequenceOfCodePointsFast } = require('../infra')
 const { maxNameValuePairSize, maxAttributeValueSize } = require('./constants')
 const { isCTLExcludingHtab } = require('./util')
-const { collectASequenceOfCodePointsFast } = require('../fetch/data-url')
 const assert = require('node:assert')
 const { unescape: qsUnescape } = require('node:querystring')
 

lib/web/fetch/body.js

@@ -5,8 +5,7 @@ const {
   ReadableStreamFrom,
   readableStreamClose,
   fullyReadBody,
-  extractMimeType,
-  utf8DecodeBytes
+  extractMimeType
 } = require('./util')
 const { FormData, setFormDataState } = require('./formdata')
 const { webidl } = require('../webidl')
@@ -16,6 +15,8 @@ const { isArrayBuffer } = require('node:util/types')
 const { serializeAMimeType } = require('./data-url')
 const { multipartFormDataParser } = require('./formdata-parser')
 const { createDeferredPromise } = require('../../util/promise')
+const { parseJSONFromBytes } = require('../infra')
+const { utf8DecodeBytes } = require('../../encoding')
 
 let random
 
@@ -496,14 +497,6 @@ function bodyUnusable (object) {
   return body != null && (body.stream.locked || util.isDisturbed(body.stream))
 }
 
-/**
- * @see https://infra.spec.whatwg.org/#parse-json-bytes-to-a-javascript-value
- * @param {Uint8Array} bytes
- */
-function parseJSONFromBytes (bytes) {
-  return JSON.parse(utf8DecodeBytes(bytes))
-}
-
 /**
  * @see https://fetch.spec.whatwg.org/#concept-body-mime-type
  * @param {any} requestOrResponse internal state

lib/web/fetch/data-url.js

@@ -1,6 +1,7 @@
 'use strict'
 
 const assert = require('node:assert')
+const { forgivingBase64, collectASequenceOfCodePoints, collectASequenceOfCodePointsFast, isomorphicDecode, removeASCIIWhitespace, removeChars } = require('../infra')
 
 const encoder = new TextEncoder()
 
@@ -9,7 +10,6 @@ const encoder = new TextEncoder()
  */
 const HTTP_TOKEN_CODEPOINTS = /^[!#$%&'*+\-.^_|~A-Za-z0-9]+$/
 const HTTP_WHITESPACE_REGEX = /[\u000A\u000D\u0009\u0020]/ // eslint-disable-line
-const ASCII_WHITESPACE_REPLACE_REGEX = /[\u0009\u000A\u000C\u000D\u0020]/g // eslint-disable-line
 /**
  * @see https://mimesniff.spec.whatwg.org/#http-quoted-string-token-code-point
  */
@@ -136,49 +136,6 @@ function URLSerializer (url, excludeFragment = false) {
   return serialized
 }
 
-// https://infra.spec.whatwg.org/#collect-a-sequence-of-code-points
-/**
- * @param {(char: string) => boolean} condition
- * @param {string} input
- * @param {{ position: number }} position
- */
-function collectASequenceOfCodePoints (condition, input, position) {
-  // 1. Let result be the empty string.
-  let result = ''
-
-  // 2. While position doesn’t point past the end of input and the
-  // code point at position within input meets the condition condition:
-  while (position.position < input.length && condition(input[position.position])) {
-    // 1. Append that code point to the end of result.
-    result += input[position.position]
-
-    // 2. Advance position by 1.
-    position.position++
-  }
-
-  // 3. Return result.
-  return result
-}
-
-/**
- * A faster collectASequenceOfCodePoints that only works when comparing a single character.
- * @param {string} char
- * @param {string} input
- * @param {{ position: number }} position
- */
-function collectASequenceOfCodePointsFast (char, input, position) {
-  const idx = input.indexOf(char, position.position)
-  const start = position.position
-
-  if (idx === -1) {
-    position.position = input.length
-    return input.slice(start)
-  }
-
-  position.position = idx
-  return input.slice(start, position.position)
-}
-
 // https://url.spec.whatwg.org/#string-percent-decode
 /** @param {string} input */
 function stringPercentDecode (input) {
@@ -427,45 +384,6 @@ function parseMIMEType (input) {
   return mimeType
 }
 
-// https://infra.spec.whatwg.org/#forgiving-base64-decode
-/** @param {string} data */
-function forgivingBase64 (data) {
-  // 1. Remove all ASCII whitespace from data.
-  data = data.replace(ASCII_WHITESPACE_REPLACE_REGEX, '')
-
-  let dataLength = data.length
-  // 2. If data’s code point length divides by 4 leaving
-  // no remainder, then:
-  if (dataLength % 4 === 0) {
-    // 1. If data ends with one or two U+003D (=) code points,
-    // then remove them from data.
-    if (data.charCodeAt(dataLength - 1) === 0x003D) {
-      --dataLength
-      if (data.charCodeAt(dataLength - 1) === 0x003D) {
-        --dataLength
-      }
-    }
-  }
-
-  // 3. If data’s code point length divides by 4 leaving
-  // a remainder of 1, then return failure.
-  if (dataLength % 4 === 1) {
-    return 'failure'
-  }
-
-  // 4. If data contains a code point that is not one of
-  //  U+002B (+)
-  //  U+002F (/)
-  //  ASCII alphanumeric
-  // then return failure.
-  if (/[^+/0-9A-Za-z]/.test(data.length === dataLength ? data : data.substring(0, dataLength))) {
-    return 'failure'
-  }
-
-  const buffer = Buffer.from(data, 'base64')
-  return new Uint8Array(buffer.buffer, buffer.byteOffset, buffer.byteLength)
-}
-
 // https://fetch.spec.whatwg.org/#collect-an-http-quoted-string
 // tests: https://fetch.spec.whatwg.org/#example-http-quoted-string
 /**
@@ -608,71 +526,6 @@ function removeHTTPWhitespace (str, leading = true, trailing = true) {
   return removeChars(str, leading, trailing, isHTTPWhiteSpace)
 }
 
-/**
- * @see https://infra.spec.whatwg.org/#ascii-whitespace
- * @param {number} char
- */
-function isASCIIWhitespace (char) {
-  // "\r\n\t\f "
-  return char === 0x00d || char === 0x00a || char === 0x009 || char === 0x00c || char === 0x020
-}
-
-/**
- * @see https://infra.spec.whatwg.org/#strip-leading-and-trailing-ascii-whitespace
- * @param {string} str
- * @param {boolean} [leading=true]
- * @param {boolean} [trailing=true]
- */
-function removeASCIIWhitespace (str, leading = true, trailing = true) {
-  return removeChars(str, leading, trailing, isASCIIWhitespace)
-}
-
-/**
- * @param {string} str
- * @param {boolean} leading
- * @param {boolean} trailing
- * @param {(charCode: number) => boolean} predicate
- * @returns
- */
-function removeChars (str, leading, trailing, predicate) {
-  let lead = 0
-  let trail = str.length - 1
-
-  if (leading) {
-    while (lead < str.length && predicate(str.charCodeAt(lead))) lead++
-  }
-
-  if (trailing) {
-    while (trail > 0 && predicate(str.charCodeAt(trail))) trail--
-  }
-
-  return lead === 0 && trail === str.length - 1 ? str : str.slice(lead, trail + 1)
-}
-
-/**
- * @see https://infra.spec.whatwg.org/#isomorphic-decode
- * @param {Uint8Array} input
- * @returns {string}
- */
-function isomorphicDecode (input) {
-  // 1. To isomorphic decode a byte sequence input, return a string whose code point
-  //    length is equal to input’s length and whose code points have the same values
-  //    as the values of input’s bytes, in the same order.
-  const length = input.length
-  if ((2 << 15) - 1 > length) {
-    return String.fromCharCode.apply(null, input)
-  }
-  let result = ''; let i = 0
-  let addition = (2 << 15) - 1
-  while (i < length) {
-    if (i + addition > length) {
-      addition = length - i
-    }
-    result += String.fromCharCode.apply(null, input.subarray(i, i += addition))
-  }
-  return result
-}
-
 /**
  * @see https://mimesniff.spec.whatwg.org/#minimize-a-supported-mime-type
  * @param {Exclude<ReturnType<typeof parseMIMEType>, 'failure'>} mimeType
@@ -730,15 +583,11 @@ function minimizeSupportedMimeType (mimeType) {
 module.exports = {
   dataURLProcessor,
   URLSerializer,
-  collectASequenceOfCodePoints,
-  collectASequenceOfCodePointsFast,
   stringPercentDecode,
   parseMIMEType,
   collectAnHTTPQuotedString,
   serializeAMimeType,
-  removeChars,
   removeHTTPWhitespace,
   minimizeSupportedMimeType,
-  HTTP_TOKEN_CODEPOINTS,
-  isomorphicDecode
+  HTTP_TOKEN_CODEPOINTS
 }

lib/web/fetch/formdata-parser.js

@@ -1,11 +1,12 @@
 'use strict'
 
 const { bufferToLowerCasedHeaderName } = require('../../core/util')
-const { utf8DecodeBytes } = require('./util')
-const { HTTP_TOKEN_CODEPOINTS, isomorphicDecode } = require('./data-url')
+const { HTTP_TOKEN_CODEPOINTS } = require('./data-url')
 const { makeEntry } = require('./formdata')
 const { webidl } = require('../webidl')
 const assert = require('node:assert')
+const { isomorphicDecode } = require('../infra')
+const { utf8DecodeBytes } = require('../../encoding')
 
 const formDataNameBuffer = Buffer.from('form-data; name="')
 const filenameBuffer = Buffer.from('filename')

lib/web/fetch/index.js

@@ -35,7 +35,6 @@ const {
   isErrorLike,
   fullyReadBody,
   readableStreamClose,
-  isomorphicEncode,
   urlIsLocal,
   urlIsHttpHttpsScheme,
   urlHasHttpsScheme,
@@ -63,6 +62,7 @@ const { webidl } = require('../webidl')
 const { STATUS_CODES } = require('node:http')
 const { bytesMatch } = require('../subresource-integrity/subresource-integrity')
 const { createDeferredPromise } = require('../../util/promise')
+const { isomorphicEncode } = require('../infra')
 
 const hasZstd = typeof zlib.createZstdDecompress === 'function'
 

lib/web/fetch/response.js

@@ -9,9 +9,7 @@ const {
   isValidReasonPhrase,
   isCancelled,
   isAborted,
-  serializeJavascriptValueToJSONString,
   isErrorLike,
-  isomorphicEncode,
   environmentSettingsObject: relevantRealm
 } = require('./util')
 const {
@@ -22,6 +20,7 @@ const { webidl } = require('../webidl')
 const { URLSerializer } = require('./data-url')
 const { kConstruct } = require('../../core/symbols')
 const assert = require('node:assert')
+const { isomorphicEncode, serializeJavascriptValueToJSONString } = require('../infra')
 
 const textEncoder = new TextEncoder('utf-8')