fix(arg-parser): a quoted value closes at a quote that ends a word, also after whitespace

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

lib/parsers/arg-parser.mjs

@@ -137,15 +137,16 @@ const isSpace = (c) => /\s/.test(c);
  * Split argument text into words. Words are separated by whitespace. A word
  * that starts with `'` or `"` opens a quote. Its closing quote is the next
  * same-kind quote that ends a word (followed by whitespace or the end of the
- * text) and does not start one; a same-kind quote inside a word is skipped.
- * The search stops at a same-kind quote that starts a word (one preceded by
- * whitespace): then there is no pair. A pair groups the words between the
- * quotes into one value and is removed. A quote with no partner is a literal
- * character (`user's` stays `user's`). Examples: `'Fix the user's login'` is
- * one value, `Fix the user's login`; `"my dir"/x --since "HEAD~1"` gives the
- * literal words `"my` and `dir"/x`, then `--since` and `HEAD~1`, so a quoted
- * value never swallows a later flag. Nothing is expanded: the function only
- * slices the string.
+ * text), whatever stands before it: a letter, whitespace or a line break. A
+ * same-kind quote that starts a word (whitespace before it, a non-whitespace
+ * character after it) stops the search: then there is no pair. Any other
+ * same-kind quote (inside a word) is skipped. A pair groups the words between
+ * the quotes into one value and is removed. A quote with no partner is a
+ * literal character (`user's` stays `user's`). Examples: `"my dir "` is one
+ * value, `my dir `; `'Fix the user's login'` is one value,
+ * `Fix the user's login`; `"my dir"/x --since "HEAD~1"` gives the literal
+ * words `"my` and `dir"/x`, then `--since` and `HEAD~1`. Nothing is expanded:
+ * the function only slices the string.
  *
  * @param {string} text
  * @returns {Array<{ value: string, start: number, end: number }>}
@@ -163,8 +164,11 @@ export function tokenize(text) {
     const q = text[i];
     if (q === '"' || q === "'") {
       let j = text.indexOf(q, i + 1);
-      while (j !== -1 && !isSpace(text[j - 1]) && j + 1 < n && !isSpace(text[j + 1])) j = text.indexOf(q, j + 1);
-      if (j !== -1 && !isSpace(text[j - 1])) {
+      while (j !== -1 && j + 1 < n && !isSpace(text[j + 1])) {
+        if (isSpace(text[j - 1])) { j = -1; break; }
+        j = text.indexOf(q, j + 1);
+      }
+      if (j !== -1) {
         out.push({ value: text.slice(i + 1, j), start, end: j + 1 });
         i = j + 1;
         continue;

tests/lib/arg-parser-closing-quote.test.mjs

@@ -0,0 +1,29 @@
+// [CQW] Fix round 4.5b, part 3 (decision 89).
+// Defect: `tokenize` refused a closing quote that had whitespace or a line
+// break before it, so `"multi line` + line break + `"` (the form the `--stdin`
+// heredoc produces), `--project "my dir "`, `"a "` and `" "` never paired and
+// their quotes stayed literal characters. A same-kind quote that ends a word
+// must close the value, whatever stands before it.
+// This file must load at ce98f62: it imports only `parseArgs` and `tokenize`,
+// which exist there, so its red is an assertion, not a crash.
+import { test } from 'node:test';
+import { strict as assert } from 'node:assert';
+import { parseArgs, tokenize } from '../../lib/parsers/arg-parser.mjs';
+
+const values = (text) => tokenize(text).map((t) => t.value);
+
+test('[CQW] a quoted value closes at a quote that ends a word, also after whitespace or a line break', () => {
+  // Red at the start point: the closing quote has whitespace before it.
+  assert.deepEqual(values('"multi line\n" --since HEAD'), ['multi line\n', '--since', 'HEAD']);
+  const r = parseArgs('--project "my dir "', 'trekreview');
+  assert.equal(r.flags['--project'], 'my dir ');
+  assert.deepEqual(r.positional, []);
+  assert.deepEqual(values('"a "'), ['a ']);
+  assert.deepEqual(values('" "'), [' ']);
+
+  // Control: a quote that starts a word stops the search; a quote inside a
+  // word is skipped; a quote that ends a word after a letter still closes.
+  assert.equal(parseArgs('--project "my dir"/x --since "HEAD~1"', 'trekreview').flags['--since'], 'HEAD~1');
+  assert.deepEqual(values("'Fix the user's login'"), ["Fix the user's login"]);
+  assert.deepEqual(values("'the '90s were great'"), ["'the", '90s were great']);
+});