fix: match literal MIME types with regex metacharacters in DocumentTypeRouter (#11973)

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Co-authored-by: Stefano Fiorucci <stefanofiorucci@gmail.com>

Author: 3 people

haystack/components/routers/document_type_router.py

@@ -134,9 +134,12 @@ def run(self, documents: list[Document]) -> dict[str, list[Document]]:
 
             matched = False
             if mime_type:
-                for pattern in self._mime_type_patterns:
-                    if pattern.fullmatch(mime_type):
-                        mime_types[pattern.pattern].append(doc)
+                for bucket_key, pattern in zip(self.mime_types, self._mime_type_patterns, strict=True):
+                    # Match an exact MIME type first, so literal types containing regex
+                    # metacharacters (e.g. the '+' in 'image/svg+xml') are not misread as
+                    # a regex; fall back to regex matching for patterns like 'audio/.*'.
+                    if mime_type == bucket_key or pattern.fullmatch(mime_type):
+                        mime_types[bucket_key].append(doc)
                         matched = True
                         break
             if not matched:

releasenotes/notes/fix-document-type-router-literal-mime-141fa900eda54ec5.yaml

@@ -0,0 +1,9 @@
+---
+fixes:
+  - |
+    Fixed ``DocumentTypeRouter`` misrouting documents whose MIME type contains a
+    regex metacharacter. A declared type such as the standard IANA ``image/svg+xml``
+    was compiled as a regex, so the ``+`` was treated as a quantifier and the
+    document fell into ``unclassified`` instead of its own bucket. Declared MIME
+    types are now matched by exact equality first, falling back to regex matching
+    so patterns like ``audio/.*`` keep working.

test/components/routers/test_document_type_router.py

@@ -107,6 +107,24 @@ def test_run_with_mime_type_meta_field(self):
         assert result["audio/x-wav"][0].content == "Audio content"
         assert result["unclassified"][0].content == "Unknown type"
 
+    def test_run_with_literal_mime_type_containing_regex_metacharacter(self):
+        # 'image/svg+xml' is a standard IANA type; the '+' must be treated as a
+        # literal, not a regex quantifier. A regex pattern like 'audio/.*' must
+        # still match by regex.
+        docs = [
+            Document(content="An SVG", meta={"mime_type": "image/svg+xml"}),
+            Document(content="Some audio", meta={"mime_type": "audio/mpeg"}),
+        ]
+
+        router = DocumentTypeRouter(mime_type_meta_field="mime_type", mime_types=["image/svg+xml", "audio/.*"])
+        result = router.run(documents=docs)
+
+        assert "unclassified" not in result
+        assert len(result["image/svg+xml"]) == 1
+        assert result["image/svg+xml"][0].content == "An SVG"
+        assert len(result["audio/.*"]) == 1
+        assert result["audio/.*"][0].content == "Some audio"
+
     def test_run_with_file_path_meta_field(self):
         docs = [
             Document(content="Example text", meta={"file_path": "example.txt"}),