fix(pipeline): respect quotes when parsing template expressions

src/pipeline/template.test.ts

@@ -100,6 +100,29 @@ describe('evalExpr', () => {
   it('applies join filter', () => {
     expect(evalExpr('item.tags | join(,)', { item: { tags: ['a', 'b', 'c'] } })).toBe('a,b,c');
   });
+  it('keeps a | inside a quoted filter argument', () => {
+    expect(evalExpr("item.tags | join(' | ')", { item: { tags: ['a', 'b', 'c'] } })).toBe('a | b | c');
+    expect(evalExpr("item.name | default('a|b') | upper", { item: {} })).toBe('A|B');
+  });
+  it('keeps a | inside a quoted string literal', () => {
+    expect(evalExpr("'a|b'", {})).toBe('a|b');
+    expect(evalExpr("item.name || 'a|b'", { item: {} })).toBe('a|b');
+  });
+  it('still splits filters after an unterminated quote', () => {
+    expect(evalExpr("item.name | default(it's) | upper", { item: {} })).toBe("IT'S");
+  });
+  it('returns a plain quoted string literal', () => {
+    expect(evalExpr("'plain'", {})).toBe('plain');
+    expect(evalExpr('"it\'s"', {})).toBe("it's");
+  });
+  it('evaluates concatenation that starts and ends with a string literal', () => {
+    expect(evalExpr("'https://x.test/' + item.id + '/'", { item: { id: 7 } })).toBe('https://x.test/7/');
+    expect(evalExpr('"a" + "b"', {})).toBe('ab');
+    expect(evalExpr("'x' === 'x' ? 'yes' : 'no'", {})).toBe('yes');
+  });
+  it('evaluates escapes in a string literal', () => {
+    expect(evalExpr("'it\\'s'", {})).toBe("it's");
+  });
   it('applies upper filter', () => {
     expect(evalExpr('item.name | upper', { item: { name: 'hello' } })).toBe('HELLO');
   });
@@ -147,6 +170,9 @@ describe('render', () => {
   it('renders inline helper expressions', () => {
     expect(render('https://example.com/search?q=${{ encodeURIComponent(args.keyword) }}', { args: { keyword: 'hello world' } })).toBe('https://example.com/search?q=hello%20world');
   });
+  it('renders a concatenation wrapped in string literals', () => {
+    expect(render("${{ 'https://x.test/' + item.id + '/' }}", { item: { id: 7 } })).toBe('https://x.test/7/');
+  });
   it('renders full multiline expressions', () => {
     expect(render("${{\n  args.topic ? `https://medium.com/tag/${args.topic}` : 'https://medium.com/tag/technology'\n}}", { args: { topic: 'ai' } })).toBe('https://medium.com/tag/ai');
   });

src/pipeline/template.ts

@@ -41,7 +41,7 @@ export function evalExpr(expr: string, ctx: RenderContext): unknown {
 
   // ── Pipe filters: expr | filter1(arg) | filter2 ──
   // Split on single | (not ||) so "item.a || item.b | upper" works correctly.
-  const pipeSegments = expr.split(/(?<!\|)\|(?!\|)/).map(s => s.trim());
+  const pipeSegments = splitPipes(expr);
   if (pipeSegments.length > 1) {
     let result = evalExpr(pipeSegments[0], ctx);
     for (let i = 1; i < pipeSegments.length; i++) {
@@ -50,8 +50,11 @@ export function evalExpr(expr: string, ctx: RenderContext): unknown {
     return result;
   }
 
-  // Fast path: quoted string literal — skip VM overhead
-  const strLit = expr.match(/^(['"])(.*)\1$/);
+  // Fast path: quoted string literal — skip VM overhead.
+  // The body must not contain the delimiter or a backslash: "'a' + item.id + 'b'"
+  // also starts and ends with a quote but is a concatenation, and escapes need
+  // real JS parsing. Both fall through to the VM below.
+  const strLit = expr.match(/^(['"])((?:(?!\1)[^\\])*)\1$/);
   if (strLit) return strLit[2];
 
   // Fast path: numeric literal
@@ -66,6 +69,35 @@ export function evalExpr(expr: string, ctx: RenderContext): unknown {
   return evalJsExpr(expr, { args, item, data, root, index });
 }
 
+const PIPE_SEPARATOR = /(?<!\|)\|(?!\|)/;
+
+/**
+ * Split an expression into its filter segments on single `|` (not `||`),
+ * skipping any `|` inside a quoted string so `join(' | ')` and `'a|b'` stay
+ * in one segment. An unterminated quote (e.g. `default(it's)`) means the
+ * quotes are not string delimiters, so fall back to the plain split.
+ */
+function splitPipes(expr: string): string[] {
+  const segments: string[] = [];
+  let quote = '';
+  let start = 0;
+  for (let i = 0; i < expr.length; i++) {
+    const ch = expr[i];
+    if (quote) {
+      if (ch === '\\') i++;
+      else if (ch === quote) quote = '';
+    } else if (ch === "'" || ch === '"' || ch === '`') {
+      quote = ch;
+    } else if (ch === '|' && expr[i - 1] !== '|' && expr[i + 1] !== '|') {
+      segments.push(expr.slice(start, i));
+      start = i + 1;
+    }
+  }
+  if (quote) return expr.split(PIPE_SEPARATOR).map(s => s.trim());
+  segments.push(expr.slice(start));
+  return segments.map(s => s.trim());
+}
+
 /**
  * Apply a named filter to a value.
  * Supported filters: