feat(datasets): edit dataset metadata (#549)

Author: jsam

renku/api/datasets.py

@@ -87,6 +87,7 @@ def dataset_path(self, name):
             path = LinkReference(
                 client=self, name='datasets/' + name
             ).reference
+
         return path
 
     def load_dataset(self, name=None):

renku/cli/_format/datasets.py

@@ -31,7 +31,7 @@ def tabular(client, datasets):
         tabulate(
             datasets,
             headers=OrderedDict((
-                ('short_id', 'id'),
+                ('uid', 'id'),
                 ('display_name', None),
                 ('version', None),
                 ('created', None),

renku/cli/dataset.py

@@ -159,12 +159,15 @@
 from urllib.parse import ParseResult
 
 import click
+import editor
 import requests
+import yaml
 from click import BadParameter
 from tqdm import tqdm
 
 from renku.cli._providers import ProviderFactory
 from renku.models._tabulate import tabulate
+from renku.models.datasets import Dataset
 
 from .._compat import Path
 from ._client import pass_local_client
@@ -217,6 +220,27 @@ def create(client, name):
         click.secho('OK', fg='green')
 
 
+@dataset.command()
+@click.argument('id')
+@pass_local_client(clean=True, commit=True)
+def edit(client, id):
+    """Edit dataset metadata."""
+    dataset_ = client.load_dataset(id)
+
+    if dataset_:
+        metadata_edited = editor.edit(
+            contents=bytes(
+                yaml.safe_dump(dataset_.editable), encoding='utf-8'
+            )
+        )
+
+        edited = yaml.safe_load(metadata_edited)
+        updated_ = Dataset(**edited)
+
+        dataset_.update_metadata(updated_)
+        dataset_.to_yaml()
+
+
 @dataset.command()
 @click.argument('name')
 @click.argument('urls', nargs=-1)

renku/models/datasets.py

@@ -268,6 +268,11 @@ class Dataset(CreatorsMixin):
 
     SUPPORTED_SCHEMES = ('', 'file', 'http', 'https', 'git+https', 'git+ssh')
 
+    EDITABLE_FIELDS = [
+        'creator', 'date_published', 'description', 'in_language', 'keywords',
+        'license', 'name', 'url', 'version', 'created', 'files'
+    ]
+
     _id = jsonld.ib(default=None, context='@id')
 
     creator = jsonld.container.list(
@@ -334,7 +339,7 @@ def to_unix(el):
     @property
     def uid(self):
         """UUID part of identifier."""
-        return self._id.split('/')[-1]
+        return self.identifier.split('/')[-1]
 
     @created.default
     def _now(self):
@@ -353,6 +358,13 @@ def creators_csv(self):
         """Comma-separated list of creators associated with dataset."""
         return ','.join(creator.short_name for creator in self.creator)
 
+    @property
+    def editable(self):
+        """Subset of attributes which user can edit."""
+        obj = self.asjsonld()
+        data = {field_: obj.pop(field_) for field_ in self.EDITABLE_FIELDS}
+        return data
+
     def find_file(self, file_path, return_index=False):
         """Find a file in files container."""
         for index, file_ in enumerate(self.files):
@@ -370,13 +382,7 @@ def update_metadata(self, other_dataset):
         if is_doi(other_dataset.identifier):
             self._id = other_dataset.identifier
 
-        update_fields = [
-            'creator', 'date_published', 'description', 'identifier',
-            'in_language', 'keywords', 'license', 'name', 'url', 'version',
-            'created', 'files'
-        ]
-
-        for field_ in update_fields:
+        for field_ in self.EDITABLE_FIELDS:
             val = getattr(other_dataset, field_)
             if val:
                 setattr(self, field_, val)

setup.py

@@ -80,6 +80,7 @@
     'PyYAML>=3.12',
     'pyld>=1.0.3',
     'python-dateutil>=2.6.1',
+    'python-editor>=1.0.4',
     'rdflib-jsonld>=0.4.0',
     'requests>=2.21.0',
     'tabulate>=0.7.7',

tests/cli/test_datasets.py

@@ -673,6 +673,20 @@ def test_dataset_overwrite_confirm(runner, project):
     assert 'OK' in result.output
 
 
+def test_dataset_edit(runner, client, project):
+    """Check dataset metadata editing."""
+    result = runner.invoke(cli.cli, ['dataset', 'create', 'dataset'])
+    assert 0 == result.exit_code
+    assert 'OK' in result.output
+
+    dataset = client.load_dataset(name='dataset')
+
+    result = runner.invoke(
+        cli.cli, ['dataset', 'edit', dataset.identifier], input='wq'
+    )
+    assert 0 == result.exit_code
+
+
 @pytest.mark.integration
 def test_dataset_import_real_doi(runner, project):
     """Test dataset import for existing DOI."""