feat(profiles): ✨ detect invalid JSON in file descriptor and abort validation

Make the "File Descriptor JSON format" check (RO-Crate 1.1 and 1.2) catch
JSONDecodeError explicitly: report a precise issue with the parse error
message and line/column, then call context.abort_validation() so no metadata
is read and downstream checks don't emit false positives.

Author: kikkomep

rocrate_validator/profiles/ro-crate/1.1/must/0_file_descriptor_format.py

@@ -14,6 +14,7 @@
 
 # pylint: disable=invalid-name  # profile filename uses digit prefix (load-order convention)
 
+import json
 import re
 from typing import Any
 from urllib.parse import urljoin
@@ -77,6 +78,17 @@ def check(self, context: ValidationContext) -> bool:
             logger.debug("Checking validity of JSON file at %s", context.ro_crate.metadata)
             context.ro_crate.metadata.as_dict()
             return True
+        except json.JSONDecodeError as e:
+            context.result.add_issue(
+                f'RO-Crate file descriptor "{context.rel_fd_path}" is not valid JSON: '
+                f"{e.msg} at line {e.lineno}, column {e.colno} (char {e.pos})",
+                self,
+            )
+            if logger.isEnabledFor(logging.DEBUG):
+                logger.debug("RO-Crate file descriptor is not valid JSON", exc_info=True)
+            # The metadata cannot be parsed: abort to avoid false positives downstream.
+            context.abort_validation(f"file descriptor is not valid JSON: {e}")
+            return False
         except Exception:
             context.result.add_issue(
                 f'RO-Crate file descriptor "{context.rel_fd_path}" is not in the correct format', self

rocrate_validator/profiles/ro-crate/1.2/must/0_file_descriptor_format.py

@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import json
 import re
 from http import HTTPStatus
 from pathlib import Path
@@ -128,6 +129,17 @@ def check(self, context: ValidationContext) -> bool:
             logger.debug("Checking validity of JSON file at %s", context.ro_crate.metadata)
             context.ro_crate.metadata.as_dict()
             return True
+        except json.JSONDecodeError as e:
+            context.result.add_issue(
+                f'RO-Crate file descriptor "{context.rel_fd_path}" is not valid JSON: '
+                f"{e.msg} at line {e.lineno}, column {e.colno} (char {e.pos})",
+                self,
+            )
+            if logger.isEnabledFor(logging.DEBUG):
+                logger.debug("RO-Crate file descriptor is not valid JSON", exc_info=True)
+            # The metadata cannot be parsed: abort to avoid false positives downstream.
+            context.abort_validation(f"file descriptor is not valid JSON: {e}")
+            return False
         except Exception:
             context.result.add_issue(
                 f'RO-Crate file descriptor "{context.rel_fd_path}" is not in the correct format', self