def test_write_read_archive(self):
from pycldf.dataset import Dataset
from pycldf.util import Archive
ds = Dataset.from_file(FIXTURES.joinpath('ds1.csv'))
out = self.tmp_path()
with self.assertRaises(ValueError):
ds.write(out.joinpath('non-existing'), '.tsv', archive=True)
with Archive(self.tmp_path('archive.zip').as_posix(), 'w') as archive:
ds.write('.', archive=archive)
ds2 = Dataset.from_file(FIXTURES.joinpath('ds1.csv'))
ds2.name = 'new_name'
ds2.write('.', archive=archive)
ds_out = Dataset.from_zip(self.tmp_path('archive.zip'), name='ds1')
self.assertEqual(ds.rows, ds_out.rows)
self.assertEqual(ds.metadata, ds_out.metadata)
with Archive(self.tmp_path('archive.zip')) as archive:
ds_out = Dataset.from_metadata('ds1.csv-metadata.json', container=archive)
self.assertEqual(ds.rows, ds_out.rows)
self.assertEqual(ds.metadata, ds_out.metadata)
ds.write(out, '.tsv', archive=True)
ds_out = Dataset.from_zip(out.joinpath('ds1.zip'))
self.assertEqual(ds.rows, ds_out.rows)
self.assertEqual(ds.metadata, ds_out.metadata)
def test_dataset_from_file(self):
from pycldf.dataset import Dataset
ds = Dataset.from_file(FIXTURES.joinpath('ds1.csv'))
self.assertIn('ds1', repr(ds))
self.assertEqual(len(ds), 2)
self.assertEqual(ds.table.url, 'ds1.csv')
self.assertEqual(ds.metadata['dc:creator'], 'The Author')
row = ['3', 'abcd1234', 'fid2', 'maybe', '', 'new[4]']
with self.assertRaises(ValueError):
ds.add_row(row)
ds.sources.add('@book{new,\nauthor={new author}}')
res = ds.add_row(row)
self.assertEqual(res.url, 'http://example.org/valuesets/3')
self.assertEqual(len(res.refs), 1)
self.assertEqual(
res.valueUrl('Language_ID'),
'http://glottolog.org/resource/languoid/id/abcd1234')
res = ds.add_row(['4', None, None, None, None, None])
self.assertEqual(res.valueUrl('Language_ID'), None)
out = self.tmp_path()
ds.write(out, '.tsv')
self.assertTrue(out.joinpath('ds1.bib').exists())
md = load(out.joinpath('ds1.tsv-metadata.json'))
self.assertEqual('ds1.tsv', md['tables'][0]['url'])
Dataset.from_file(out.joinpath('ds1.tsv'))
def test_invalid_dataset_from_file(self):
from pycldf.dataset import Dataset
log = Mock(warn=Mock())
with patch('pycldf.dataset.log', log):
Dataset.from_file(FIXTURES.joinpath('invalid.csv'), skip_on_error=True)
self.assertEqual(log.warn.call_count, 2)
with self.assertRaises(ValueError):
Dataset.from_file(FIXTURES.joinpath('invalid.csv'))
def stats(args):
"""
cldf stats <DATASET>
Print basic stats for CLDF dataset <DATASET>, where <DATASET> may be the path to
- a CLDF metadata file
- a CLDF core data file
- a CLDF zip archive
"""
if len(args.args) < 1:
raise ParserError('not enough arguments')
fname = Path(args.args[0])
if not fname.exists() or not fname.is_file():
raise ParserError('%s is not an existing directory' % fname)
if fname.suffix == '.zip':
ds = Dataset.from_zip(fname)
elif fname.name.endswith(MD_SUFFIX):
ds = Dataset.from_metadata(fname)
else:
ds = Dataset.from_file(fname)
print(fname)
stats_ = ds.stats
print("""
Name: %s
Different languages: %s
Different parameters: %s
Rows: %s
""" % (
ds.name,
len(stats_['languages']),
len(stats_['parameters']),
stats_['rowcount']
))
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)