Merge pull request #200 from dbt-labs/fix/validate-identifier-special-chars

Author: b-per

src/dbt_jobs_as_code/loader/load.py

@@ -8,6 +8,7 @@
 from ruamel.yaml import YAML
 
 from dbt_jobs_as_code.schemas.config import Config
+from dbt_jobs_as_code.schemas.job import VALID_IDENTIFIER_RE
 
 
 class LoadingJobsYAMLError(Exception):
@@ -49,22 +50,26 @@ def load_job_configuration(config_files: List[str], vars_file: Optional[List[str
 
 
 def _validate_job_identifiers(jobs: dict) -> None:
-    """Validate that job identifiers don't contain spaces.
+    """Validate that job identifiers only contain allowed characters.
+
+    Identifiers are embedded as [[identifier]] in job names and extracted via regex,
+    so they must only contain letters, digits, underscores, and hyphens.
 
     Args:
         jobs: Dictionary of jobs with identifiers as keys
 
     Raises:
-        LoadingJobsYAMLError: If any job identifier contains spaces
+        LoadingJobsYAMLError: If any job identifier contains invalid characters
     """
     invalid_identifiers = []
     for identifier in jobs.keys():
-        if " " in identifier:
+        if not VALID_IDENTIFIER_RE.match(identifier):
             invalid_identifiers.append(identifier)
 
     if invalid_identifiers:
         raise LoadingJobsYAMLError(
-            f"Job identifiers cannot contain spaces. Invalid identifiers: {', '.join(invalid_identifiers)}"
+            f"Job identifiers can only contain letters, digits, underscores, and hyphens. "
+            f"Invalid identifiers: {', '.join(invalid_identifiers)}"
         )
 
 

src/dbt_jobs_as_code/schemas/job.py

@@ -22,6 +22,10 @@
 
 JOB_TYPES_WITHOUT_SCHEDULE = ["ci", "merge"]
 
+# Characters allowed in a YAML identifier key (embedded as [[identifier]] in job names).
+# Must match the character class used when extracting identifiers from job names.
+VALID_IDENTIFIER_RE = re.compile(r"^[a-zA-Z0-9_-]+$")
+
 
 @dataclass
 class IdentifierInfo:

tests/loader/test_loader.py

@@ -73,7 +73,9 @@ def test_load_job_configuration_identifier_with_spaces_error(
         with pytest.raises(LoadingJobsYAMLError) as exc_info:
             load_job_configuration([str(config_file)], None)
 
-        assert "Job identifiers cannot contain spaces" in str(exc_info.value)
+        assert "Job identifiers can only contain letters, digits, underscores, and hyphens" in str(
+            exc_info.value
+        )
         assert "job with spaces" in str(exc_info.value)
 
     def test_load_job_configuration_multiple_identifiers_with_spaces_error(
@@ -86,7 +88,9 @@ def test_load_job_configuration_multiple_identifiers_with_spaces_error(
         with pytest.raises(LoadingJobsYAMLError) as exc_info:
             load_job_configuration([str(config_file)], None)
 
-        assert "Job identifiers cannot contain spaces" in str(exc_info.value)
+        assert "Job identifiers can only contain letters, digits, underscores, and hyphens" in str(
+            exc_info.value
+        )
         assert "job with spaces" in str(exc_info.value)
         assert "another invalid job" in str(exc_info.value)
 
@@ -124,7 +128,9 @@ def test_validate_job_identifiers_with_spaces(self):
         with pytest.raises(LoadingJobsYAMLError) as exc_info:
             _validate_job_identifiers(jobs)
 
-        assert "Job identifiers cannot contain spaces" in str(exc_info.value)
+        assert "Job identifiers can only contain letters, digits, underscores, and hyphens" in str(
+            exc_info.value
+        )
         assert "job with spaces" in str(exc_info.value)
         assert "another job" in str(exc_info.value)
 
@@ -139,9 +145,26 @@ def test_validate_job_identifiers_mixed_valid_invalid(self):
         with pytest.raises(LoadingJobsYAMLError) as exc_info:
             _validate_job_identifiers(jobs)
 
-        assert "Job identifiers cannot contain spaces" in str(exc_info.value)
+        assert "Job identifiers can only contain letters, digits, underscores, and hyphens" in str(
+            exc_info.value
+        )
         assert "job with spaces" in str(exc_info.value)
 
+    def test_validate_job_identifiers_with_special_chars(self):
+        """Test that identifiers with special characters (e.g. &, @, .) are rejected."""
+        jobs = {
+            "job&name": {},
+            "job@name": {},
+        }
+
+        with pytest.raises(LoadingJobsYAMLError) as exc_info:
+            _validate_job_identifiers(jobs)
+
+        assert "Job identifiers can only contain letters, digits, underscores, and hyphens" in str(
+            exc_info.value
+        )
+        assert "job&name" in str(exc_info.value)
+
     def test_validate_job_identifiers_empty_dict(self):
         """Test that validation passes for empty jobs dictionary."""
         jobs = {}