chore: align `Headers` to spec (#10199)

This commit aligns `Headers` to spec. It also removes the now unused
03_dom_iterable.js file. We now pass all relevant `Headers` WPT. We do
not implement any sort of header filtering, as we are a server side
runtime.

This is likely not the most efficient implementation of `Headers` yet.
It is however spec compliant. Once all the APIs in the `HTTP` hot loop
are correct we can start optimizing them. It is likely that this commit
reduces bench throughput temporarily.

1 files changed, 104 insertions, 0 deletions

diff --git a/op_crates/web/00_infra.js b/op_crates/web/00_infra.js
index 0590ffd03..ff9cb7cd4 100644
--- a/op_crates/web/00_infra.js
+++ b/op_crates/web/00_infra.js
@@ -8,6 +8,74 @@
 "use strict";
 
 ((window) => {
+  const ASCII_DIGIT = ["\u0030-\u0039"];
+  const ASCII_UPPER_ALPHA = ["\u0041-\u005A"];
+  const ASCII_LOWER_ALPHA = ["\u0061-\u007A"];
+  const ASCII_ALPHA = [...ASCII_UPPER_ALPHA, ...ASCII_LOWER_ALPHA];
+  const ASCII_ALPHANUMERIC = [...ASCII_DIGIT, ...ASCII_ALPHA];
+
+  const HTTP_TAB_OR_SPACE = ["\u0009", "\u0020"];
+  const HTTP_WHITESPACE = ["\u000A", "\u000D", ...HTTP_TAB_OR_SPACE];
+
+  const HTTP_TOKEN_CODE_POINT = [
+    "\u0021",
+    "\u0023",
+    "\u0024",
+    "\u0025",
+    "\u0026",
+    "\u0027",
+    "\u002A",
+    "\u002B",
+    "\u002D",
+    "\u002E",
+    "\u005E",
+    "\u005F",
+    "\u0060",
+    "\u007C",
+    "\u007E",
+    ...ASCII_ALPHANUMERIC,
+  ];
+  const HTTP_TOKEN_CODE_POINT_RE = new RegExp(
+    `^[${regexMatcher(HTTP_TOKEN_CODE_POINT)}]+$`,
+  );
+  const HTTP_QUOTED_STRING_TOKEN_POINT = [
+    "\u0009",
+    "\u0020-\u007E",
+    "\u0080-\u00FF",
+  ];
+  const HTTP_QUOTED_STRING_TOKEN_POINT_RE = new RegExp(
+    `^[${regexMatcher(HTTP_QUOTED_STRING_TOKEN_POINT)}]+$`,
+  );
+  const HTTP_WHITESPACE_MATCHER = regexMatcher(HTTP_WHITESPACE);
+  const HTTP_WHITESPACE_PREFIX_RE = new RegExp(
+    `^[${HTTP_WHITESPACE_MATCHER}]+`,
+    "g",
+  );
+  const HTTP_WHITESPACE_SUFFIX_RE = new RegExp(
+    `[${HTTP_WHITESPACE_MATCHER}]+$`,
+    "g",
+  );
+
+  /**
+   * Turn a string of chars into a regex safe matcher.
+   * @param {string[]} chars 
+   * @returns {string}
+   */
+  function regexMatcher(chars) {
+    const matchers = chars.map((char) => {
+      if (char.length === 1) {
+        return `\\u${char.charCodeAt(0).toString(16).padStart(4, "0")}`;
+      } else if (char.length === 3 && char[1] === "-") {
+        return `\\u${char.charCodeAt(0).toString(16).padStart(4, "0")}-\\u${
+          char.charCodeAt(2).toString(16).padStart(4, "0")
+        }`;
+      } else {
+        throw TypeError("unreachable");
+      }
+    });
+    return matchers.join("");
+  }
+
   /**
    * https://infra.spec.whatwg.org/#collect-a-sequence-of-code-points
    * @param {string} input
@@ -25,7 +93,43 @@
     return { result: input.slice(start, position), position };
   }
 
+  /**
+   * @param {string} s
+   * @returns {string}
+   */
+  function byteUpperCase(s) {
+    return String(s).replace(/[a-z]/g, function byteUpperCaseReplace(c) {
+      return c.toUpperCase();
+    });
+  }
+
+  /**
+   * @param {string} s
+   * @returns {string}
+   */
+  function byteLowerCase(s) {
+    return String(s).replace(/[A-Z]/g, function byteUpperCaseReplace(c) {
+      return c.toLowerCase();
+    });
+  }
+
   window.__bootstrap.infra = {
     collectSequenceOfCodepoints,
+    ASCII_DIGIT,
+    ASCII_UPPER_ALPHA,
+    ASCII_LOWER_ALPHA,
+    ASCII_ALPHA,
+    ASCII_ALPHANUMERIC,
+    HTTP_TAB_OR_SPACE,
+    HTTP_WHITESPACE,
+    HTTP_TOKEN_CODE_POINT,
+    HTTP_TOKEN_CODE_POINT_RE,
+    HTTP_QUOTED_STRING_TOKEN_POINT,
+    HTTP_QUOTED_STRING_TOKEN_POINT_RE,
+    HTTP_WHITESPACE_PREFIX_RE,
+    HTTP_WHITESPACE_SUFFIX_RE,
+    regexMatcher,
+    byteUpperCase,
+    byteLowerCase,
   };
 })(globalThis);