cli: Add more useful defaults

two paths + reimport tasks at the package level so they are automatically
picked up by luigi without passing all module names on the CLI

swh/graph/cli.py

@@ -381,11 +381,11 @@ def luigi(
 
     default_values = dict(
         local_export_path=dataset_path,
-        export_task_type="ExportGraph",
-        compression_task_type="CompressGraph",
         local_graph_path=dataset_path / "compressed",
+        derived_datasets_path=dataset_path,
         topological_order_path=dataset_path / "topology/topological_order_dfs.csv.zst",
         origin_contributors_path=dataset_path / "datasets/contribution_graph.csv.zst",
+        origin_urls_path=dataset_path / "datasets/origin_urls.csv.zst",
         export_id=f"{dataset_name}-{secrets.token_hex(10)}",
     )
 

swh/graph/luigi/__init__.py

@@ -51,7 +51,9 @@ from typing import List
 
 import luigi
 
-from . import compressed_graph
+from .compressed_graph import *  # noqa
+from .misc_datasets import *  # noqa
+from .origin_contributors import *  # noqa
 
 
 class RunExportCompressUpload(luigi.Task):
@@ -64,9 +66,11 @@ class RunExportCompressUpload(luigi.Task):
         """
         from swh.dataset.luigi import RunExportAll
 
+        from .compressed_graph import UploadGraphToS3
+
         return [
             RunExportAll(),
-            compressed_graph.UploadGraphToS3(),
+            UploadGraphToS3(),
         ]
 
     def complete(self) -> bool: