Update to @cto.af/ucd generator. Upgraded to Unicode 16.0.0. BREAKING: now requires node 20+.

Author: hildjj

.github/workflows/node.js.yml

@@ -11,7 +11,7 @@ jobs:
 
     strategy:
       matrix:
-        node-version: [18, 20, 22, 23]
+        node-version: [20, 22, 24]
         os: [ubuntu-latest]
     runs-on: ${{ matrix.os }}
     steps:

.gitignore

@@ -3,5 +3,4 @@ coverage/
 docs/
 lib/widths.js
 node_modules/
-tools/*.txt
 types/

package.json

@@ -27,30 +27,24 @@
   ],
   "author": "Joe Hildebrand <[email protected]>",
   "license": "MIT",
-  "devDependencies": {
-    "@cto.af/eslint-config": "5.1.12",
-    "@types/node": "22.13.5",
-    "c8": "10.1.3",
-    "eslint": "9.21.0",
-    "eslint-plugin-mocha": "10.5.0",
-    "mocha": "11.1.0",
-    "typedoc": "0.27.9",
-    "typescript": "5.7.3"
-  },
   "dependencies": {
-    "@cto.af/unicode-trie": "2.0.1",
+    "@cto.af/unicode-trie-runtime": "3.2.2",
     "ansi-regex": "6.1.0",
     "emoji-regex": "10.4.0"
   },
-  "pnpm": {
-    "overrides": {
-      "chokidar": "4.0.3",
-      "fast-glob": "3.3.3",
-      "foreground-child": "3.3.1"
-    }
+  "devDependencies": {
+    "@cto.af/eslint-config": "6.0.4",
+    "@cto.af/unicode-trie": "3.2.2",
+    "@types/node": "22.15.24",
+    "c8": "10.1.3",
+    "eslint": "9.27.0",
+    "eslint-plugin-mocha": "11.1.0",
+    "mocha": "11.5.0",
+    "typedoc": "0.28.5",
+    "typescript": "5.8.3"
   },
-  "packageManager": "pnpm@10.5.2",
+  "packageManager": "pnpm@10.11.0",
   "engines": {
-    "node": ">= 18"
+    "node": ">= 20"
   }
 }

pnpm-lock.yaml

@@ -1,87 +1,64 @@
 import {AMBIGUOUS, POTENTIAL_EMOJI} from '../lib/constants.js';
-import {UnicodeTrieBuilder} from '@cto.af/unicode-trie/builder.js';
 import {fileURLToPath} from 'node:url';
-import fs from 'node:fs/promises';
 import path from 'node:path';
+import {writeFile} from '@cto.af/unicode-trie/file';
 
 const __filename = fileURLToPath(import.meta.url);
 const __dirname = path.dirname(__filename);
 
-async function processFile(name, trie, transform) {
-  const INPUT = path.join(__dirname, `${name.replaceAll('/', '_')}.txt`);
-
-  // Cache data in local file.  Requires Node v18+.
-  let txt = null;
-  try {
-    txt = await fs.readFile(INPUT, 'utf8');
-  } catch (_ignored) {
-    const url = `https://www.unicode.org/Public/UCD/latest/ucd/${name}.txt`;
-    const res = await fetch(url);
-    txt = await res.text();
-    await fs.writeFile(INPUT, txt, 'utf8');
-  }
-  // # LineBreak-15.0.0.txt
-  // # Date: 2022-07-28, 09:20:42 GMT [KW, LI]
-
-  const version = txt.match(/^#\s*\S+-(?<version>\d+\.\d+\.\d+).txt/)?.groups?.version;
-  const date = txt.match(/^#\s*Date: (?<date>[\d,: GMT-]+)/m)?.groups?.date;
-
-  const matches = txt.matchAll(
-    /^(?<start>\p{Hex}{4,6})(?:\.\.(?<end>\p{Hex}{4,6}))?\s*;\s*(?<val>\S+)/gmu
-  );
-  for (const match of matches) {
-    const val = transform(match.groups.val);
-    if (val == null) {
-      continue;
-    }
-    const start = parseInt(match.groups.start, 16);
-    if (match.groups.end) {
-      const end = parseInt(match.groups.end, 16);
-      trie.setRange(start, end, val);
-    } else {
-      trie.set(start, val);
-    }
-  }
-  return {version, date};
-}
-
-const trie = new UnicodeTrieBuilder(1, NaN);
-
-const {version, date} = await processFile('DerivedCoreProperties', trie, x => {
-  if (x === 'Default_Ignorable_Code_Point') {
-    return 0;
-  }
-  return null;
-});
-
-await processFile('Scripts', trie, x => {
-  if (x === 'Hangul') {
-    return 2;
-  }
-  return null;
-});
-
-await processFile('EastAsianWidth', trie, x => {
-  switch (x) {
-    case 'F':
-      return 2;
-    case 'W':
-      return 2;
-    case 'A':
-      return AMBIGUOUS;
-  }
-  return null;
+await writeFile([
+  {
+    name: 'DerivedCoreProperties.txt',
+    transform(x) {
+      if (x === 'Default_Ignorable_Code_Point') {
+        return 0;
+      }
+      return null;
+    },
+  },
+  {
+    name: 'Scripts.txt',
+    transform(x) {
+      return (x === 'Hangul') ? 2 : null;
+    },
+  },
+  {
+    name: 'EastAsianWidth.txt',
+    transform(x) {
+      switch (x) {
+        case 'F':
+          return 2;
+        case 'W':
+          return 2;
+        case 'A':
+          return AMBIGUOUS;
+      }
+      return null;
+    },
+  },
+  {
+    name: 'emoji/emoji-data.txt',
+    transform(x) {
+      return (x === 'Emoji') ? POTENTIAL_EMOJI : null;
+    },
+  },
+  {
+    transform(trie) {
+      trie.setRange(0, 20, 0); // C0 Controls
+      trie.set(0x8, -1); // Backspace
+      trie.set(0x7f, 0); // ESC
+      trie.set(0x2E3A, 3); // TWO-EM DASH, in some contexts and fonts
+      trie.set(0x2E3B, 4); // THREE-EM DASH, in some contexts and fonts
+    },
+  },
+], {
+  cacheDir: __dirname,
+  initialValue: 1,
+  errorValue: NaN,
+  className: 'Width',
+  out: path.resolve(__dirname, '..', 'lib', 'widths.js'),
+  quot: "'",
+  semi: ';',
+  verbose: true,
+  frequency: 0,
 });
-
-await processFile('emoji/emoji-data', trie, x => ((x === 'Emoji') ? POTENTIAL_EMOJI : null));
-
-trie.setRange(0, 20, 0); // C0 Controls
-trie.set(0x8, -1); // Backspace
-trie.set(0x7f, 0); // ESC
-trie.set(0x2E3A, 3); // TWO-EM DASH, in some contexts and fonts
-trie.set(0x2E3B, 4); // THREE-EM DASH, in some contexts and fonts
-
-const OUTPUT = path.resolve(__dirname, '..', 'lib', 'widths.js');
-await fs.writeFile(OUTPUT, trie.toModule({
-  version, date, name: 'Width', quot: "'", semi: ';',
-}));