Commit 50a760f7 authored by Jan Reimes's avatar Jan Reimes
Browse files

fix(db): deserialize metadata_payload and handle missing JSON 

- Deserialize metadata_payload field when loaded from DB if it's a JSON string\n- Preserve empty object on JSON decode error\n- Fix except syntax in pool_executors factory import
parent 1d814557

src/pool_executors/pool_executors/factory.py

+1 −1
Original line number Diff line number Diff line
@@ -14,7 +14,7 @@ try: 
    from concurrent.futures import InterpreterPoolExecutor  # noqa: PLC0415



    HAS_INTERPRETER_POOL_EXECUTOR = True

except (ImportError, AttributeError):

except ImportError, AttributeError:

    InterpreterPoolExecutor = None  # type: ignore[assignment, misc]

    HAS_INTERPRETER_POOL_EXECUTOR = False

src/tdoc_crawler/database/connection.py

+9 −2
Original line number Diff line number Diff line
@@ -704,7 +704,7 @@ class TDocDatabase: 
                        try:

                            if "T" in value and value.endswith(("Z", "+00:00")):

                                row_dict[key] = datetime.fromisoformat(value)

                        except (ValueError, AttributeError):

                        except ValueError, AttributeError:

                            pass

                if table == "spec_source_records":

                    metadata_payload = row_dict.get("metadata_payload")
@@ -750,7 +750,14 @@ class TDocDatabase: 


    def _get_spec_source_record(self, record_id: str) -> SpecificationSourceRecord | None:

        try:

            return self.connection.model_from_table("spec_source_records", record_id)  # type: ignore[arg-type]

            record = self.connection.model_from_table("spec_source_records", record_id)  # type: ignore[arg-type]

            # Handle JSON deserialization for metadata_payload field

            if record is not None and isinstance(record.metadata_payload, str):

                try:

                    record = record.model_copy(update={"metadata_payload": json.loads(record.metadata_payload)})

                except json.JSONDecodeError:

                    record = record.model_copy(update={"metadata_payload": {}})

            return record

        except KeyError:

            return None

src/tdoc_crawler/specs/sources/whatthespec.py

+1 −1

File changed.

Contains only whitespace changes.