fix(project): keep non-ASCII characters in slugified identifiers

packages/project/src/util/slugify.ts

@@ -1,5 +1,21 @@
+/**
+ * Convert a human-readable name into a slug used as an identifier.
+ *
+ * Any run of characters which is neither a letter, a digit nor an underscore
+ * becomes a separator. Letters and digits are matched with the Unicode
+ * properties `\p{L}` and `\p{N}` rather than `\w`, which is ASCII-only and
+ * would drop every non-Latin character: `café` slugified to `caf`, and a name
+ * written entirely in a non-Latin script collapsed to an empty string.
+ *
+ * For ASCII-only input this produces exactly the same slug as before, so
+ * identifiers in existing projects are unaffected.
+ */
 export default function slugify(text: string) {
   return (
-    text?.replace(/\W/g, ' ').trim().replace(/\s+/g, '-').toLowerCase() ?? ''
+    text
+      ?.replace(/[^\p{L}\p{N}_]+/gu, ' ')
+      .trim()
+      .replace(/\s+/g, '-')
+      .toLowerCase() ?? ''
   );
 }

packages/project/test/util/slugify.test.ts

@@ -0,0 +1,42 @@
+import test from 'ava';
+import slugify from '../../src/util/slugify';
+
+test('lowercases and hyphenates a plain name', (t) => {
+  t.is(slugify('My Workflow'), 'my-workflow');
+});
+
+test('collapses runs of separators', (t) => {
+  t.is(slugify('My   Workflow -- Two'), 'my-workflow-two');
+});
+
+test('keeps underscores and digits', (t) => {
+  t.is(slugify('step_1 of 2'), 'step_1-of-2');
+});
+
+test('trims leading and trailing separators', (t) => {
+  t.is(slugify('  ...My Workflow!  '), 'my-workflow');
+});
+
+test('keeps accented latin characters', (t) => {
+  t.is(slugify('café'), 'café');
+  t.is(slugify('résumé'), 'résumé');
+});
+
+test('keeps a whole accented phrase readable', (t) => {
+  t.is(slugify("Vérifier l'état du patient"), 'vérifier-l-état-du-patient');
+});
+
+test('keeps non-latin scripts instead of collapsing to an empty string', (t) => {
+  t.is(slugify('患者確認'), '患者確認');
+  t.is(slugify('проверка пациента'), 'проверка-пациента');
+  t.is(slugify('التحقق من المريض'), 'التحقق-من-المريض');
+});
+
+test('still drops characters which are neither letters nor digits', (t) => {
+  t.is(slugify('deploy 🚀 now'), 'deploy-now');
+});
+
+test('handles empty and nullish input', (t) => {
+  t.is(slugify(''), '');
+  t.is(slugify(undefined as unknown as string), '');
+});

packages/project/tsconfig.json

@@ -3,6 +3,9 @@
   "include": ["src/**/*.ts", "test/**/*test.ts"],
   "compilerOptions": {
     "module": "ESNext",
+    // Unicode property escapes in slugify need ES2018 or later. Build output is
+    // controlled by tsup; this only affects type validation.
+    "target": "ES2020",
     "sourceMap": true
   }
 }