def test_set_options():
_config.set({'database_directory': '/tmp/collections'})
s1 = _config.get('database_directory')
assert s1 == os.path.abspath(os.path.expanduser('/tmp/collections'))
_config.set({'database_directory': '/tests/test_collections'})
s2 = _config.get('database_directory')
assert not s1 == s2
def test_to_xarray():
with config.set({'database-directory': './tests/test_collections'}):
col = intake.open_esm_metadatastore(collection_name='gmet_test')
cat = col.search(direct_access=True)
ds = cat.to_xarray(chunks={'time': 1}, decode_times=True)
assert isinstance(ds, xr.Dataset)
assert 'member_id' in ds.coords
def test_build_collection_dict():
with config.set({'database-directory': './tests/test_collections'}):
collection_definition = {
'collection_type': 'cmip5',
'data_sources': {
'BNU-ESM': {
'extra_attributes': {
'activity': 'CMIP5',
'institute': 'BNU',
'product': 'output1',
},
'locations': [{
'direct_access':
True,
'exclude_dirs': ['*/files/*', 'latest'],
'file_extension':
'.nc',
'loc_type':
'posix',
'name':
'SAMPLE-DATA',
'urlpath':
'./tests/sample_data/cmip/cmip5/output1/BNU/BNU-ESM',
}],
}
},
'name': 'cmip5_test_dict_collection',
}
col = intake.open_esm_metadatastore(
collection_input_definition=collection_definition,
overwrite_existing=True)
assert isinstance(col.df, pd.DataFrame)
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(
collection_name='cmip5_test_collection')
cat = c.search(model=['CanESM2', 'CSIRO-Mk3-6-0'])
assert isinstance(cat.ds, xr.Dataset)
assert len(cat.ds.index) > 0
def test_build_collection():
with config.set({'database-directory': './tests/test_collections'}):
collection_input_definition = os.path.join(here, 'gmet-test.yml')
col = intake.open_esm_metadatastore(
collection_input_definition=collection_input_definition,
overwrite_existing=True)
assert isinstance(col.ds, xr.Dataset)
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(
collection_name='cmip5_test_collection')
cat = c.search(model=['CanESM2', 'CSIRO-Mk3-6-0'])
assert isinstance(cat.query_results, pd.DataFrame)
assert not cat.query_results.empty
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(
collection_name='cmip6_test_collection')
cat = c.search(source_id=['CNRM-ESM2-1', 'GISS-E2-1-G'])
assert isinstance(cat.query_results, pd.DataFrame)
assert not cat.query_results.empty
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(collection_name='mpige_test')
cat = c.search(component='mpiom', stream='monitoring_ym')
assert isinstance(cat.ds, xr.Dataset)
assert len(cat.ds.index) > 0
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(collection_name='mpige_test')
cat = c.search(component='mpiom', stream='monitoring_ym')
assert isinstance(cat.query_results, pd.DataFrame)
assert not cat.query_results.empty
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
col = intake.open_esm_metadatastore(collection_name='gmet_test')
cat = col.search(member_id=[1, 2])
assert isinstance(cat.ds, xr.Dataset)
assert len(cat.ds.index) > 0
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
col = intake.open_esm_metadatastore(
collection_name='AWS-CESM1-LE', storage_options=storage_options
)
cat = col.search(variable=['RAIN', 'FSNO'])
assert len(cat.ds.index) > 0
def test_to_xarray_restore_non_coords(query):
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(collection_name='cesm1-le')
cat = c.search(**query)
dset = cat.to_xarray(decode_times=False)
_, ds = dset.popitem()
assert 'TAREA' in ds.data_vars
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(
collection_name='cesm_dple_test_collection')
cat = c.search(variable='O2', direct_access=True)
assert isinstance(cat.query_results, pd.DataFrame)
assert not cat.query_results.empty
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
col = intake.open_esm_metadatastore(collection_name='era5_test')
cat = col.search(variable_short_name=['mn2t', 'mx2t'],
forecast_initial_date=['2002-02-01', '2002-02-16'])
assert isinstance(cat.ds, xr.Dataset)
assert len(cat.ds.index) > 0
def test_build_collection_cesm1_aws_le():
with config.set({'database-directory': './tests/test_collections'}):
col = intake.open_esm_metadatastore(
collection_input_definition=cdef,
overwrite_existing=True,
storage_options=storage_options,
)
assert isinstance(col.ds, xr.Dataset)
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
col = intake.open_esm_metadatastore(collection_name='era5_test')
cat = col.search(variable_short_name=['mn2t', 'mx2t'],
forecast_initial_date=['2002-02-01', '2002-02-16'])
assert isinstance(cat.query_results, pd.DataFrame)
assert not cat.query_results.empty
def test_to_xarray():
with config.set({'database-directory': './tests/test_collections'}):
col = intake.open_esm_metadatastore(
collection_name='AWS-CESM1-LE', storage_options=storage_options
)
cat = col.search(variable='FSNO', experiment='20C', component='lnd')
dsets = cat.to_xarray()
_, ds = dsets.popitem()
assert isinstance(ds, xr.Dataset)
def test_open_collection_def_locally(filepath):
"""Opens all files listed in file_alias_dict."""
data_cache_dir = f'{TMPDIR}/intake-esm-tests'
with config.set({
'database-directory': './tests/test_collections',
'data-cache-directory': data_cache_dir
}):
d = load_collection_input_file(filepath, cache=False)
assert isinstance(d, dict)
assert len(d) > 0
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
col = intake.open_esm_metadatastore(collection_name='gmet_test')
cat = col.search(
member_id=[1, 2],
time_range=['19800101-19801231', '19810101-19811231', '19820101-19821231'],
)
assert isinstance(cat.query_results, pd.DataFrame)
assert not cat.query_results.empty
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
col = intake.open_esm_metadatastore(collection_name='cafe_test')
cat = col.search(
frequency='day',
variable_short_name=['ucomp', 'wt'],
start_date=['1988-03-01', ]
)
assert isinstance(cat.query_results, pd.DataFrame)
assert not cat.query_results.empty
def test_search():
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(
collection_name='cmip6_test_collection')
cat = c.search(source_id=['BCC-ESM1', 'CNRM-CM6-1', 'CNRM-ESM2-1'])
assert isinstance(cat.ds, xr.Dataset)
assert len(cat.ds.index) > 0
assert isinstance(cat.df, pd.DataFrame)
assert isinstance(cat.nunique(), pd.Series)
assert isinstance(cat.unique(), dict)
def test_to_xarray_cesm(chunks, expected_chunks):
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(collection_name='cesm1-le')
cat = c.search(
variable=['STF_O2', 'SHF'],
ensemble=[1, 3, 9],
experiment=['20C', 'RCP85'],
direct_access=True,
)
dset = cat.to_xarray(chunks=chunks)
ds = dset['pop.h.ocn']
assert ds['SHF'].data.chunksize == expected_chunks
def test_to_xarray_cmip(chunks, expected_chunks):
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(
collection_name='cmip6_test_collection')
# Test for data from multiple models
cat = c.search(source_id=['CNRM-ESM2-1', 'CNRM-CM6-1', 'BCC-ESM1'],
variable_id=['tasmax'])
ds = cat.to_xarray(decode_times=False, chunks=chunks)
print(ds)
assert isinstance(ds, dict)
_, dset = ds.popitem()
assert dset['tasmax'].data.chunksize == expected_chunks
def test_to_xarray_cmip(chunks, expected_chunks):
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(
collection_name='cmip6_test_collection')
# Test for data from multiple institutions
cat = c.search(source_id=['CNRM-ESM2-1', 'GISS-E2-1-G'],
variable_id=['prra', 'tasmax'])
ds = cat.to_xarray(decode_times=False, chunks=chunks)
print(ds)
assert isinstance(ds, dict)
nasa_dset = ds['NASA-GISS.GISS-E2-1-G.amip.Omon.gn']
assert nasa_dset['prra'].data.chunksize == expected_chunks
def test_to_xarray_cesm(chunks, expected_chunks):
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(collection_name='cesm1-le')
query = {
'variable': ['STF_O2', 'SHF'],
'member_id': [1, 3, 9],
'experiment': ['20C', 'RCP85'],
'direct_access': True,
}
cat = c.search(**query)
dset = cat.to_xarray(chunks=chunks)
_, ds = dset.popitem()
assert ds['SHF'].data.chunksize == expected_chunks
def test_to_xarray_cmip_empty():
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(collection_name='cmip5_test_collection')
cat = c.search(
model='CanESM2',
experiment='rcp85',
frequency='mon',
modeling_realm='atmos',
ensemble_member='r2i1p1',
)
with pytest.raises(ValueError):
cat.to_xarray()
def test_to_xarray():
with config.set({'database-directory': './tests/test_collections'}):
col = intake.open_esm_metadatastore(collection_name='mpige_test')
cat = col.search(component='mpiom', stream='monitoring_ym')
with pytest.warns(UserWarning):
ds = cat.to_xarray()
assert isinstance(ds, dict)
cat = col.search(
component=['mpiom', 'hamocc'],
stream='monitoring_ym',
experiment=['hist', 'rcp85'],
ensemble=[2, 3],
)
ds = cat.to_xarray(merge_exp=False)
assert 'experiment_id' in ds.coords
def test_to_xarray_cesm_hsi():
data_cache_dir = f'{TMPDIR}/intake-esm-tests/transferred-data'
with config.set({
'database-directory': './tests/test_collections',
'data-cache-directory': data_cache_dir
}):
collection_input_definition = os.path.join(
here, '../ensure-file-hsi-transfer-collection-input.yml')
col = intake.open_esm_metadatastore(
collection_input_definition=collection_input_definition,
overwrite_existing=False)
cat = col.search(variable=['SST'])
dset = cat.to_xarray(chunks={'time': 365})
_, ds = dset.popitem()
assert isinstance(ds['SST'], xr.DataArray)
def test_to_xarray_cmip(chunks, expected_chunks):
with config.set({'database-directory': './tests/test_collections'}):
c = intake.open_esm_metadatastore(collection_name='cmip5_test_collection')
cat = c.search(
variable=['hfls'], frequency='mon', modeling_realm='atmos', model=['CNRM-CM5']
)
dset = cat.to_xarray(decode_times=True, chunks=chunks)
ds = dset['CNRM-CERFACS.CNRM-CM5.historical.mon.atmos']
assert ds['hfls'].data.chunksize == expected_chunks
# Test for data from multiple institutions
cat = c.search(variable=['hfls'], frequency='mon', modeling_realm='atmos')
ds = cat.to_xarray(decode_times=False, chunks=chunks)
assert isinstance(ds, dict)
assert 'CCCma.CanCM4.historical.mon.atmos' in ds.keys()
def test_file_transfer_symlink():
data_cache_dir = f'{TMPDIR}/intake-esm-tests/transferred-data'
with config.set(
{'database-directory': './tests/test_collections', 'data-cache-directory': data_cache_dir}
):
collection_input_definition = os.path.join(here, 'copy-to-cache-collection-input.yml')
col = intake.open_esm_metadatastore(
collection_input_definition=collection_input_definition, overwrite_existing=True
)
cat = col.search(variable=['STF_O2', 'SHF'])
query_results = _ensure_file_access(cat.ds)
local_urlpaths = query_results['file_fullpath'].tolist()
assert isinstance(local_urlpaths, list)
assert len(local_urlpaths) > 0
shutil.rmtree(data_cache_dir)