Sanitize job-derived operation identifiers in polling labels

Co-authored-by: Shri Sukhani <shrisukhani@users.noreply.github.com>

Author: cursoragent

hyperbrowser/client/polling.py

@@ -68,7 +68,15 @@ def _validate_operation_name(operation_name: str) -> None:
 
 
 def build_operation_name(prefix: str, identifier: str) -> str:
-    operation_name = f"{prefix}{identifier}"
+    normalized_identifier = identifier.strip()
+    if not normalized_identifier:
+        normalized_identifier = "unknown"
+    normalized_identifier = "".join(
+        "?" if ord(character) < 32 or ord(character) == 127 else character
+        for character in normalized_identifier
+    )
+
+    operation_name = f"{prefix}{normalized_identifier}"
     if len(operation_name) <= _MAX_OPERATION_NAME_LENGTH:
         return operation_name
     available_identifier_length = (

tests/test_polling.py

@@ -71,6 +71,18 @@ def test_build_operation_name_truncates_overlong_prefixes():
     assert operation_name == long_prefix[:200]
 
 
+def test_build_operation_name_sanitizes_identifier_whitespace_and_control_chars():
+    operation_name = build_operation_name("crawl job ", " \nabc\tdef\x7f ")
+
+    assert operation_name == "crawl job abc?def?"
+
+
+def test_build_operation_name_uses_unknown_for_blank_identifier():
+    operation_name = build_operation_name("crawl job ", " \n\t ")
+
+    assert operation_name == "crawl job unknown"
+
+
 def test_poll_until_terminal_status_allows_immediate_terminal_on_zero_max_wait():
     status = poll_until_terminal_status(
         operation_name="sync immediate zero wait",