def dataset(self, req):
ds = Dataset('%s-%s-%s' % (
req.dataset.id, self.obj.__class__.__name__.lower(), self.obj.id))
cols = self.columns(req)
ds.fields = tuple(col['name'] if isinstance(col, dict) else col for col in cols)
ds.table.schema.aboutUrl = url_template(req, 'value', 'ID')
for col in cols:
if isinstance(col, dict):
name = col.pop('name')
for attr, value in col.items():
setattr(ds.table.schema.columns[name], attr, value)
ds.metadata['dc:bibliographicCitation '] = text_citation(req, self.obj)
ds.metadata['dc:publisher'] = '%s, %s' % (
req.dataset.publisher_name, req.dataset.publisher_place)
ds.metadata['dc:license'] = req.dataset.license
ds.metadata['dc:issued'] = req.dataset.published.isoformat()
ds.metadata['dc:title'] = self.obj.name
ds.metadata['dc:creator'] = self.obj.formatted_contributors()
ds.metadata['dc:identifier'] = req.resource_url(self.obj)
ds.metadata['dc:isPartOf'] = req.resource_url(req.dataset)
ds.metadata['dcat:accessURL'] = req.route_url('download')
for value in self.value_query():
refs, sources = self.refs_and_sources(req, value)
row = self.row(req, value, refs)
if row:
ds.sources.add(*sources)
ds.add_row(row)
return ds
def make_dataset(name='test', fields=None, rows=None):
ds = Dataset(name)
ds.fields = fields or tuple(f[0] for f in REQUIRED_FIELDS)
if rows:
for row in rows:
ds.add_row(row)
return ds
def test_write_read(self):
from pycldf.dataset import Dataset, REQUIRED_FIELDS
row = ['1', 'abcd1234', 'fid', 'yes']
ds = Dataset('name')
ds.fields = tuple(v[0] for v in REQUIRED_FIELDS)
ds.add_row(row)
ds.write(self.tmp_path())
self.assertTrue(self.tmp_path('name.csv').exists())
ds2 = Dataset.from_file(self.tmp_path('name.csv'))
self.assertEqual(list(ds2[0].values()), row)
self.assertEqual(list(ds2['1'].values()), row)
def add_row(self, row):
#
# add segments column, value cleaned from "<>=..."
#
row = CldfDatasetBase.add_row(self, row)
if row:
for col, validator in self.validators.items():
if not validator(row):
del self._rows[row['ID']]
return
return row
