Create tasks and review parsing reports

Screenshot note: Screenshots are illustrative. Controls or labels may remain untranslated in the demo release, and pages without a faithful English capture retain the source illustration. Control names and layouts can vary by release.

Lineage tasks organize source files, preprocessing, engines, and object matching. Prepare metadata and namespaces, then validate a small file sample.

Create a task

1. Open System Management > Lineage Management > Lineage Space > Lineage Management.
2. Select Add Task in the target folder.
3. Choose Script or ETL. Custom Excel lineage uses ETL. Container of User-Defined-Object contains saved manual lineage and cannot be parsed again.
4. Configure task, engine, and namespace; import the required format.
5. Check preprocessing variables and map every connection to its namespace.

Use actual task types and objects rather than screenshot values.

The current English environment did not provide a stable task-form capture. Confirm the exact field labels and available engines in your deployed release.

Import modes

Mode	Behavior
Full	Imports the directory's scripts, overwrites duplicate files, and resets them to unparsed; supports nested folders and matching names in different folders
Incremental	Compares content, size, path, and name; skips identical files and updates changed files

Check counts, paths, and state after import rather than judging duplicates by name alone.

Metadata constraints

Forced metadata binding is off by default. When enabled, an unmatched table or field can fail the task. Verify harvesting, schemas, names, and mappings. Any connection lacking a namespace can cause parsing failure; check source and target connections individually for cross-system ETL.

Execute and inspect

Each parsing run records start and end times, state, and counts for success, partial success, syntax errors, missing metadata, and other errors.

1. Review totals and file errors.
2. Open highlighted failed fragments for partial files.
3. Compare original and preprocessed text.
4. Correct engine, connections, metadata, or preprocessing and parse again.
5. Sample actual graph nodes, directions, and field relationships.

Parsing success and metadata coverage measure different aspects. Verify object identity even for successfully parsed files.

Related: namespaces and preprocessing, lineage views.