def _metadata_is_consolidated(m: fsspec.FSMap) -> bool:
try:
zarr.open_consolidated(m)
consolidated = True
except KeyError:
# group with un-consolidated metadata, or array
consolidated = False
return consolidated
def test_zarr_array_to_parquet_table(dataset):
"""
Test converting from a zarr array to a parquet table, specifying a list of
variables to store and setting 'snappy' compression.
"""
with tempfile.TemporaryDirectory() as tmpdir:
zarrstore: str = os.path.join(tmpdir, "temp.zarr")
dataset.to_zarr(store=zarrstore, consolidated=True)
zarrarray: zarr.hierarchy.Group = zarr.open_consolidated(store=zarrstore)
parquetpath: str = os.path.join(tmpdir, "temp.parquet")
ndarray_to_parquet(
ndarray=zarrarray,
parquetpath=parquetpath,
variables=["longitude", "latitude", "h_corr", "delta_time"],
compression="snappy",
)
df: dask.dataframe.core.DataFrame = dask.dataframe.read_parquet(
path=parquetpath
)
assert len(df) == 1404
assert list(df.columns) == [
"longitude",
"latitude",
"h_corr_1",
"h_corr_2",
"delta_time_1",
"delta_time_2",
]
assert all(np.issubdtype(dtype, np.float64) for dtype in df.dtypes)
def _open_snp_sites(self):
if self._cache_snp_sites is None:
path = f"{self.path}/v3/snp_genotypes/all/sites/"
store = SafeStore(self.fs.get_mapper(path))
root = zarr.open_consolidated(store=store)
self._cache_snp_sites = root
return self._cache_snp_sites
def get_tracking_ids(zstore):
"""given a GC zarr location, fetch the associated dataset/netCDF tracking IDs"""
from requests import session
from zarr import open_consolidated
from fsspec import get_mapper
# request params
client = session()
base_url = 'http://hdl.handle.net/api/handles/'
dset_id_query = '?type=IS_PART_OF'
version_query = '?type=VERSION_NUMBER'
# get primary tracking id
netcdf_tracking_ids = open_consolidated(
get_mapper(zstore)).attrs['tracking_id'].split('\n')
file_tracking_id = netcdf_tracking_ids[0]
version_ids = []
dataset_ids = []
# query for dataset_tracking_id
dset_id_url = base_url + file_tracking_id[4:] + dset_id_query
r = client.get(dset_id_url)
r.raise_for_status()
dataset_tracking_id = r.json()['values'][0]['data']['value']
# query for version
if ';' in dataset_tracking_id:
# multiple dataset_ids erroneously reported
dataset_tracking_id = "ambiguous"
return dataset_tracking_id, netcdf_tracking_ids
def open_group(cls, store, mode='r', synchronizer=None, group=None,
consolidated=False, consolidate_on_close=False):
import zarr
min_zarr = '2.2'
if LooseVersion(zarr.__version__) < min_zarr: # pragma: no cover
raise NotImplementedError("Zarr version %s or greater is "
"required by xarray. See zarr "
"installation "
"http://zarr.readthedocs.io/en/stable/"
"#installation" % min_zarr)
if consolidated or consolidate_on_close:
if LooseVersion(
zarr.__version__) <= '2.2.1.dev2': # pragma: no cover
raise NotImplementedError("Zarr version 2.2.1.dev2 or greater "
"is required by for consolidated "
"metadata.")
open_kwargs = dict(mode=mode, synchronizer=synchronizer, path=group)
if consolidated:
# TODO: an option to pass the metadata_key keyword
zarr_group = zarr.open_consolidated(store, **open_kwargs)
else:
zarr_group = zarr.open_group(store, **open_kwargs)
return cls(zarr_group, consolidate_on_close)
def open_group(
cls,
store,
mode="r",
synchronizer=None,
group=None,
consolidated=False,
consolidate_on_close=False,
chunk_store=None,
append_dim=None,
write_region=None,
):
import zarr
# zarr doesn't support pathlib.Path objects yet. zarr-python#601
if isinstance(store, pathlib.Path):
store = os.fspath(store)
open_kwargs = dict(mode=mode, synchronizer=synchronizer, path=group)
if chunk_store:
open_kwargs["chunk_store"] = chunk_store
if consolidated:
# TODO: an option to pass the metadata_key keyword
zarr_group = zarr.open_consolidated(store, **open_kwargs)
else:
zarr_group = zarr.open_group(store, **open_kwargs)
return cls(zarr_group, consolidate_on_close, append_dim, write_region)
def get_schema(url, coords):
mapper = fsspec.get_mapper(url)
group = zarr.open_consolidated(mapper)
schema = synth.read_schema_from_zarr(group, coords)
return schema
def _open_zarr(self):
fmap = fsspec.get_mapper(
f's3://{self.bucket_name}/{self.dataset_id}',
**self.storage_options,
)
self._zarr_group = zarr.open_consolidated(fmap)
self._total_size = np.sum(
[arr.nbytes for _, arr in self._zarr_group.items()])
self._total_size_repr = memory_repr(self._total_size)
def consolidate_metadata(self, metadata_key='.zmetadata'):
'''
Wrapper over zarr.consolidate_metadata to pass chunk store when opening the zarr store
'''
zarr.consolidate_metadata(self.store, metadata_key=metadata_key)
store_mode_cons = 'r' if self.store_mode == 'r' else 'r+'
self.zgroup = zarr.open_consolidated(self.store, metadata_key=metadata_key,
mode=store_mode_cons, chunk_store=self.zgroup.chunk_store,
path=self.store_path)
return self.zgroup
def _open_snp_genotypes(self, *, sample_set):
try:
return self._cache_snp_genotypes[sample_set]
except KeyError:
release = self._lookup_release(sample_set=sample_set)
path = f"{self.path}/{release}/snp_genotypes/all/{sample_set}/"
store = SafeStore(self.fs.get_mapper(path))
root = zarr.open_consolidated(store=store)
self._cache_snp_genotypes[sample_set] = root
return root
def _open_site_filters(self, *, mask, analysis):
key = mask, analysis
try:
return self._cache_site_filters[key]
except KeyError:
path = f"{self.path}/v3/site_filters/{analysis}/{mask}/"
store = SafeStore(self.fs.get_mapper(path))
root = zarr.open_consolidated(store=store)
self._cache_site_filters[key] = root
return root
def fetch_zarr(zarr_url, storage_options={'anon': True}):
zg = zarr.open_consolidated(fsspec.get_mapper(zarr_url, **storage_options),
mode='r')
dimensions = {}
variable_arrays = {}
for k, a in zg.arrays():
if k in a.attrs['_ARRAY_DIMENSIONS']:
dimensions[k] = a.attrs['_ARRAY_DIMENSIONS']
else:
variable_arrays[k] = a.attrs['_ARRAY_DIMENSIONS']
return zg, dimensions, variable_arrays
def _open_with_xarray_or_zarr(
m: fsspec.FSMap, consolidated: bool
) -> Tuple[Union[xr.Dataset, zarr.hierarchy.Group, zarr.core.Array], bool]:
try:
result = xr.open_zarr(m, consolidated=consolidated)
is_xarray_dataset = True
except KeyError:
# xarray requires _ARRAY_DIMENSIONS attribute, assuming missing if KeyError
result = zarr.open_consolidated(m) if consolidated else zarr.open(m)
is_xarray_dataset = False
return result, is_xarray_dataset
def open_mask_group(self):
"""Open the zarr group that contains the masks
Returns
-------
mask_group : zarr.Group
"""
mapper = self.mask_fs.get_mapper(self.mask_path)
zgroup = zarr.open_consolidated(mapper)
return zgroup
def _load(self):
import zarr
if self._grp is None:
# obtain the zarr root group
if isinstance(self._urlpath, zarr.hierarchy.Group):
# use already-opened group, allows support for nested groups
# as catalogs
root = self._urlpath
else:
# obtain store
if isinstance(self._urlpath, str):
# open store from url
from fsspec import get_mapper
store = get_mapper(self._urlpath, **self._storage_options)
else:
# assume store passed directly
store = self._urlpath
# open root group
if self._consolidated:
# use consolidated metadata
root = zarr.open_consolidated(store=store, mode='r')
else:
root = zarr.open_group(store=store, mode='r')
# deal with component path
if self._component is None:
self._grp = root
else:
self._grp = root[self._component]
# use zarr attributes as metadata
self.metadata.update(self._grp.attrs.asdict())
# build catalog entries
entries = {}
for k, v in self._grp.items():
if isinstance(v, zarr.core.Array):
entry = LocalCatalogEntry(name=k,
description='',
driver='ndzarr',
args=dict(urlpath=v),
catalog=self)
else:
entry = LocalCatalogEntry(name=k,
description='',
driver='zarr_cat',
args=dict(urlpath=v))
entries[k] = entry
self._entries = entries
def structure_mesh(allen_id):
if allen_id in _cache:
return _cache[allen_id]
fs = HTTPFileSystem()
# Todo: Use AWS store after Scott / Lydia upload
store = fs.get_mapper(
"https://thewtex.github.io/allen-ccf-itk-vtk-zarr/meshes/{0}.zarr".
format(allen_id))
root = zarr.open_consolidated(store)
mesh = zarr_to_vtkjs(root)
_cache[allen_id] = mesh
return mesh
def get_version(zstore, method='fsspec'):
client = requests.session()
baseurl = 'http://hdl.handle.net/api/handles/'
query1 = '?type=IS_PART_OF'
query2 = '?type=VERSION_NUMBER'
# get the `netcdf_tracking_ids` from the zstore metadata
if method == 'fsspec':
mapper = fsspec.get_mapper(zstore)
else:
mapper = zstore
group = zarr.open_consolidated(mapper)
tracking_ids = group.attrs['tracking_id']
# query the dataset handler to obtain `dataset_tracking_id` and `version`
versions = []
datasets = []
for file_tracking_id in tracking_ids.split('\n')[0:1]:
url = baseurl + file_tracking_id[4:] + query1
r = client.get(url)
r.raise_for_status()
dataset_tracking_id = r.json()['values'][0]['data']['value']
datasets += [dataset_tracking_id]
if ';' in dataset_tracking_id:
# multiple dataset_ids erroneously reported
dtracks = dataset_tracking_id.split(';')
vs = []
for dtrack in dtracks:
url2 = baseurl + dtrack[4:] + query2
r = client.get(url2)
r.raise_for_status()
r.json()['values'][0]['data']['value']
vs += [r.json()['values'][0]['data']['value']]
v = sorted(vs)[-1]
else:
url2 = baseurl + dataset_tracking_id[4:] + query2
r = client.get(url2)
r.raise_for_status()
v = r.json()['values'][0]['data']['value']
versions += [v]
version_id = list(set(versions))
dataset_id = list(set(datasets))
assert len(version_id) == 1
return dataset_id[0], version_id[0]
def open_group(
cls,
store,
mode="r",
synchronizer=None,
group=None,
consolidated=False,
consolidate_on_close=False,
):
import zarr
open_kwargs = dict(mode=mode, synchronizer=synchronizer, path=group)
if consolidated:
# TODO: an option to pass the metadata_key keyword
zarr_group = zarr.open_consolidated(store, **open_kwargs)
else:
zarr_group = zarr.open_group(store, **open_kwargs)
return cls(zarr_group, consolidate_on_close)
def open_group(
cls,
store,
mode="r",
synchronizer=None,
group=None,
consolidated=False,
consolidate_on_close=False,
chunk_store=None,
storage_options=None,
append_dim=None,
write_region=None,
safe_chunks=True,
):
# zarr doesn't support pathlib.Path objects yet. zarr-python#601
if isinstance(store, pathlib.Path):
store = os.fspath(store)
open_kwargs = dict(
mode=mode,
synchronizer=synchronizer,
path=group,
)
if LooseVersion(zarr.__version__) >= "2.5.0":
open_kwargs["storage_options"] = storage_options
elif storage_options:
raise ValueError(
"Storage options only compatible with zarr>=2.5.0")
if chunk_store:
open_kwargs["chunk_store"] = chunk_store
if consolidated:
# TODO: an option to pass the metadata_key keyword
zarr_group = zarr.open_consolidated(store, **open_kwargs)
else:
zarr_group = zarr.open_group(store, **open_kwargs)
return cls(zarr_group, consolidate_on_close, append_dim, write_region,
safe_chunks)
def append_zarr_along_time(source_path: str,
target_path: str,
fs: fsspec.AbstractFileSystem,
dim: str = "time"):
"""Append local zarr store at source_path to zarr store at target_path along time.
Args:
source_path: Local path to zarr store that represents an xarray dataset.
target_path: Local or remote url for zarr store to be appended to.
fs: Filesystem for target_path.
dim: (optional) name of time dimension. Defaults to "time".
Raises:
ValueError: If the chunk size in time does not evenly divide length of time
dimension for zarr stores at source_path.
Warning:
The zarr store as source_path will be modified in place.
"""
merged_time = _get_merged_time_coordinate(source_path, target_path, dim,
fs)
if fs.exists(target_path):
source_store = zarr.open(source_path, mode="r+")
target_store = zarr.open_consolidated(fsspec.get_mapper(target_path))
_assert_chunks_match(source_store, target_store, dim)
_set_time_units_like(source_store, target_store)
_shift_store(source_store, dim, _get_dim_size(target_store, dim))
elif fs.protocol == "file":
os.makedirs(target_path)
upload_dir(source_path, target_path)
_overwrite_time_array_with_single_chunk(target_path, merged_time, dim)
_, _, absolute_target_paths = fsspec.get_fs_token_paths(target_path)
consolidate_metadata(fs, absolute_target_paths[0])
def update_slice(store: Union[str, MutableMapping],
insert_index: int,
dataslice: xr.Dataset,
mode: str,
dimension: str = "time") -> None:
"""
Update existing Zarr dataset with new data slice.
:param store: A Zarr store.
:param insert_index: index at which to insert
:param dataslice: slice to insert
:param mode: Update mode, 'insert' or 'replace'
:param dimension: name of dimension perpendicular to slice
"""
if mode not in ('insert', 'replace'):
raise ValueError(f'illegal mode value: {mode!r}')
insert_mode = mode == 'insert'
append_dim_var_names = []
encoding = {}
# Neither Zarr nor xarray offer an explicit API function to check whether
# a Zarr is consolidated. Here we use the workaround of attempting to
# open as consolidated, and catching the resulting exception if this
# isn't possible. In the case of a consolidated Zarr, there is a slight
# inefficiency, since the consolidated metadata object is fetched twice
# (by Zarr and thereafter by xarray). See comments on PR #48 for
# discussion of possible optimizations.
consolidated = True
try:
_ = zarr.open_consolidated(store)
except KeyError:
consolidated = False
with xr.open_zarr(store, consolidated=consolidated) as ds:
for var_name in ds.variables:
var = ds[var_name]
if var.ndim >= 1 and dimension in var.dims:
if var.dims[0] != dimension:
# TODO: Remove this restriction -- it's not fundamentally
# necessary. Removal should be accompanied by appropriate
# unit tests and the addition of a warning to the user
# about potential slowness / inefficiency.
raise ValueError(f"dimension '{dimension}' of variable "
f"{var_name!r} must be first dimension")
append_dim_var_names.append(var_name)
enc = dict(ds[var_name].encoding)
# xarray 0.17+ supports engine preferred chunks if exposed by
# the backend zarr does that, but when we use the new
# 'preferred_chunks' when writing to zarr it raises and says,
# 'preferred_chunks' is an unsupported encoding
if 'preferred_chunks' in enc:
del enc['preferred_chunks']
encoding[var_name] = enc
temp_dir = tempfile.TemporaryDirectory(prefix='nc2zarr-slice-',
suffix='.zarr')
dataslice.to_zarr(temp_dir.name, encoding=encoding)
slice_root_group = zarr.open(temp_dir.name, mode='r')
slice_arrays = dict(slice_root_group.arrays())
root_group = zarr.open(store, mode='r+')
for var_name, var_array in root_group.arrays():
if var_name in append_dim_var_names:
slice_array = slice_arrays[var_name]
if insert_mode:
# Add one empty step
empty = zarr.creation.empty(slice_array.shape,
dtype=var_array.dtype)
var_array.append(empty, axis=0)
# Shift contents
var_array[insert_index + 1:, ...] = \
var_array[insert_index:-1, ...]
# Replace slice
var_array[insert_index, ...] = slice_array[0]
if consolidated:
zarr.consolidate_metadata(store)
def open_group(
cls,
store,
mode="r",
synchronizer=None,
group=None,
consolidated=False,
consolidate_on_close=False,
chunk_store=None,
storage_options=None,
append_dim=None,
write_region=None,
safe_chunks=True,
stacklevel=2,
):
# zarr doesn't support pathlib.Path objects yet. zarr-python#601
if isinstance(store, pathlib.Path):
store = os.fspath(store)
open_kwargs = dict(
mode=mode,
synchronizer=synchronizer,
path=group,
)
if LooseVersion(zarr.__version__) >= "2.5.0":
open_kwargs["storage_options"] = storage_options
elif storage_options:
raise ValueError("Storage options only compatible with zarr>=2.5.0")
if chunk_store:
open_kwargs["chunk_store"] = chunk_store
if consolidated is None:
consolidated = False
if consolidated is None:
try:
zarr_group = zarr.open_consolidated(store, **open_kwargs)
except KeyError:
warnings.warn(
"Failed to open Zarr store with consolidated metadata, "
"falling back to try reading non-consolidated metadata. "
"This is typically much slower for opening a dataset. "
"To silence this warning, consider:\n"
"1. Consolidating metadata in this existing store with "
"zarr.consolidate_metadata().\n"
"2. Explicitly setting consolidated=False, to avoid trying "
"to read consolidate metadata, or\n"
"3. Explicitly setting consolidated=True, to raise an "
"error in this case instead of falling back to try "
"reading non-consolidated metadata.",
RuntimeWarning,
stacklevel=stacklevel,
)
zarr_group = zarr.open_group(store, **open_kwargs)
elif consolidated:
# TODO: an option to pass the metadata_key keyword
zarr_group = zarr.open_consolidated(store, **open_kwargs)
else:
zarr_group = zarr.open_group(store, **open_kwargs)
return cls(
zarr_group,
mode,
consolidate_on_close,
append_dim,
write_region,
safe_chunks,
)
def variation_main_pass(self):
path = self.base_dir / 'variation/main/zarr/pass/ag1000g.phase2.ar1.pass'
return zarr.open_consolidated(str(path))
def haplotypes_main(self):
path = self.base_dir / 'haplotypes/main/zarr/ag1000g.phase2.ar1.haplotypes'
return zarr.open_consolidated(str(path))
def load_cmip(
activity_ids: str = "CMIP",
experiment_ids: str = "historical",
member_ids: str = "r1i1p1f1",
source_ids: str = "MIROC6",
table_ids: str = "day",
grid_labels: str = "gn",
variable_ids: List[str] = ["tasmax"],
return_type: str = 'zarr',
) -> xr.Dataset:
"""Loads CMIP6 GCM dataset based on input criteria.
Parameters
----------
activity_ids : list, optional
activity_ids in CMIP6 catalog, by default ["CMIP", "ScenarioMIP"],
experiment_ids : list, optional
experiment_ids in CMIP6 catalog, by default ["historical", "ssp370"], ex:# "ssp126", "ssp245", "ssp585"
member_ids : list, optional
member_ids in CMIP6 catalog, by default ["r1i1p1f1"]
source_ids : list, optional
source_ids in CMIP6 catalog, by default ["MIROC6"]
table_ids : list, optional
table_ids in CMIP6 catalog, by default ["day"]
grid_labels : list, optional
grid_labels in CMIP6 catalog, by default ["gn"]
variable_ids : list, optional
variable_ids in CMIP6 catalog, by default ['tasmax']
Returns
-------
ds : xr.Dataset or zarr group
Dataset or zarr group with CMIP data
"""
if isinstance(variable_ids, str):
variable_ids = [variable_ids]
col = cat.cmip6()
for i, var in enumerate(variable_ids):
stores = (
col.search(
activity_id=activity_ids,
experiment_id=experiment_ids,
member_id=member_ids,
source_id=source_ids,
table_id=table_ids,
grid_label=grid_labels,
variable_id=[var],
)
.df['zstore']
.to_list()
)
storage_options = config.get('data_catalog.era5.storage_options')
if len(stores) > 1:
raise ValueError('can only get 1 store at a time')
if return_type == 'zarr':
ds = zarr.open_consolidated(stores[0], mode='r', storage_options=storage_options)
elif return_type == 'xr':
ds = xr.open_zarr(stores[0], consolidated=True, storage_options=storage_options)
# flip the lats if necessary and drop the extra dims/vars like bnds
ds = gcm_munge(ds)
ds = lon_to_180(ds)
# convert to mm/day - helpful to prevent rounding errors from very tiny numbers
if var == 'pr':
ds['pr'] *= 86400
if i == 0:
ds_out = ds
else:
ds_out[var] = ds[var]
return ds_out
def test_end_to_end_file_conversion(self, _callback_post):
"""
Full end-to-end test of the adapter from call to `main` to Harmony callbacks, including
ensuring the contents of the file are correct. Mocks S3 interactions using @mock_s3.
"""
conn = boto3.resource('s3')
conn.create_bucket(Bucket='example-bucket',
CreateBucketConfiguration={
'LocationConstraint':
os.environ['AWS_DEFAULT_REGION']
})
netcdf_file = create_full_dataset()
netcdf_file2 = create_full_dataset()
try:
message = mock_message_for(netcdf_file, netcdf_file2)
main([
'harmony_netcdf_to_zarr', '--harmony-action', 'invoke',
'--harmony-input', message
],
config=self.config)
finally:
os.remove(netcdf_file)
os.remove(netcdf_file2)
callbacks = parse_callbacks(_callback_post)
# -- Progress and Callback Assertions --
# Assert that we got three callbacks, one for first file, one for the second, and the final message
self.assertEqual(len(callbacks), 3)
self.assertEqual(callbacks[0]['progress'], '50')
self.assertEqual(callbacks[0]['item[type]'], 'application/x-zarr')
self.assertEqual(callbacks[1]['progress'], '100')
self.assertEqual(callbacks[1]['item[type]'], 'application/x-zarr')
self.assertEqual(callbacks[2], {'status': 'successful'})
self.assertNotEqual(callbacks[0]['item[href]'],
callbacks[1]['item[href]'])
self.assertTrue(callbacks[0]['item[href]'].endswith('.zarr'))
self.assertTrue(callbacks[1]['item[href]'].endswith('.zarr'))
# Now calls back with spatial and temporal if present in the incoming message
self.assertEqual(callbacks[0]['item[temporal]'],
'2020-01-01T00:00:00.000Z,2020-01-02T00:00:00.000Z')
self.assertEqual(callbacks[0]['item[bbox]'], '-11.1,-22.2,33.3,44.4')
# Open the Zarr file that the adapter called back with
zarr_location = callbacks[0]['item[href]']
store = s3fs.S3FileSystem().get_mapper(root=zarr_location, check=False)
out = zarr.open_consolidated(store)
# -- Hierarchical Structure Assertions --
contents = textwrap.dedent("""
/
├── data
│ ├── horizontal
│ │ ├── east (1, 3, 3) int64
│ │ └── west (1, 3, 3) float64
│ └── vertical
│ ├── north (1, 3, 3) float64
│ └── south (1, 3, 3) float64
├── location
│ ├── lat (3, 3) float64
│ └── lon (3, 3) float64
└── time (1,) float64
""").strip()
self.assertEqual(str(out.tree()), contents)
# -- Metadata Assertions --
# Root level values
self.assertEqual(dict(out.attrs), ROOT_METADATA_VALUES)
# Group metadata
self.assertEqual(out['data'].attrs['description'],
'Group to hold the data')
# Variable metadata
var = out['data/vertical/north']
self.assertEqual(var.attrs['coordinates'], 'lon lat')
# -- Data Assertions --
# Nested Byte Arrays
self.assertEqual(out['data/vertical/north'][0, 0, 2], 16)
self.assertEqual(out['data/vertical/north'][0, 2, 0], 0)
self.assertEqual(out['data/vertical/south'][0, 2, 0], 16)
self.assertEqual(out['data/vertical/south'][0, 0, 2], 0)
self.assertEqual(out['data/horizontal/east'][0, 2, 2],
16) # scale_factor = 2
self.assertEqual(out['data/horizontal/east'][0, 0, 0], 0)
self.assertEqual(out['data/horizontal/west'][0, 0, 0], 16)
self.assertEqual(out['data/horizontal/west'][0, 2, 2], 0)
# 'east' attributes scale_factor removed
self.assertFalse(hasattr(out['data/horizontal/east'], 'scale_factor'))
# 'east' attributes present and scaled
self.assertEqual(out['data/horizontal/east'].attrs['valid_range'],
[0.0, 50.0])
self.assertEqual(out['data/horizontal/east'].attrs['valid_min'], 0.0)
self.assertEqual(out['data/horizontal/east'].attrs['valid_max'], 50.0)
self.assertEqual(out['data/horizontal/east'].attrs['_FillValue'],
254.0)
self.assertFalse(hasattr(out['data/horizontal/east'], 'missing_value'))
# 2D Nested Float Arrays
self.assertEqual(out['location/lat'][0, 1], 5.5)
self.assertEqual(out['location/lon'][0, 1], -5.5)
# 1D Root-Level Float Array sharing its name with a dimension
self.assertEqual(out['time'][0], 166536)
def variation_main_pass_biallelic(self):
path = self.base_dir / 'variation/main/zarr/biallelic/ag1000g.phase2.ar1.pass.biallelic'
return zarr.open_consolidated(str(path))
def vcfzarr_to_zarr(
input: PathType,
output: PathType,
*,
contigs: Optional[List[str]] = None,
grouped_by_contig: bool = False,
consolidated: bool = False,
tempdir: Optional[PathType] = None,
) -> None:
"""Convert VCF Zarr files created using scikit-allel to a single Zarr on-disk store in sgkit Xarray format.
Parameters
----------
input
Path to the input Zarr file.
output
Path to the ouput Zarr file.
contigs
The contigs to convert. By default all contigs are converted.
grouped_by_contig
Whether there is one group for each contig in the Zarr file, by default False.
consolidated
Whether the Zarr file has consolidated metadata, by default False.
tempdir
Temporary directory where intermediate files are stored. The default None means
use the system default temporary directory.
"""
if consolidated:
vcfzarr = zarr.open_consolidated(str(input), mode="r")
else:
vcfzarr = zarr.open_group(str(input), mode="r")
if not grouped_by_contig:
ds = _vcfzarr_to_dataset(vcfzarr)
ds.to_zarr(str(output))
else:
# read each contig separately, concatenate, rechunk, then save to zarr
contigs = contigs or list(vcfzarr.group_keys())
# Index the contig names
_, variant_contig_names = encode_array(contigs)
variant_contig_names = list(variant_contig_names)
vars_to_rechunk = []
vars_to_copy = []
with tempfile.TemporaryDirectory(prefix="vcfzarr_to_zarr_",
suffix=".zarr",
dir=tempdir) as tmpdir:
zarr_files = []
for i, contig in enumerate(contigs):
# convert contig group to zarr and save in tmpdir
ds = _vcfzarr_to_dataset(vcfzarr[contig], contig,
variant_contig_names)
if i == 0:
for (var, arr) in ds.data_vars.items():
if arr.dims[0] == "variants":
vars_to_rechunk.append(var)
else:
vars_to_copy.append(var)
contig_zarr_file = Path(tmpdir) / contig
ds.to_zarr(contig_zarr_file)
zarr_files.append(str(contig_zarr_file))
concat_zarrs_optimized(zarr_files,
output,
vars_to_rechunk,
vars_to_copy,
fix_strings=True)
def vcfzarr_to_zarr(
input: PathType,
output: PathType,
*,
contigs: Optional[List[str]] = None,
grouped_by_contig: bool = False,
consolidated: bool = False,
tempdir: Optional[PathType] = None,
concat_algorithm: Optional[Literal["xarray_internal"]] = None,
) -> None:
"""Convert VCF Zarr files created using scikit-allel to a single Zarr on-disk store in sgkit Xarray format.
Parameters
----------
input
Path to the input Zarr file.
output
Path to the ouput Zarr file.
contigs
The contigs to convert. By default all contigs are converted.
grouped_by_contig
Whether there is one group for each contig in the Zarr file, by default False.
consolidated
Whether the Zarr file has consolidated metadata, by default False.
tempdir
Temporary directory where intermediate files are stored. The default None means
use the system default temporary directory.
concat_algorithm
The algorithm to use to concatenate and rechunk Zarr files. The default None means
use the optimized version suitable for large files, whereas ``xarray_internal`` will
use built-in Xarray APIs, which can exhibit high memory usage, see https://github.com/dask/dask/issues/6745.
"""
if consolidated:
vcfzarr = zarr.open_consolidated(str(input), mode="r")
else:
vcfzarr = zarr.open_group(str(input), mode="r")
if not grouped_by_contig:
ds = _vcfzarr_to_dataset(vcfzarr)
ds.to_zarr(str(output))
else:
# read each contig separately, concatenate, rechunk, then save to zarr
contigs = contigs or list(vcfzarr.group_keys())
# Index the contig names
_, variant_contig_names = encode_array(contigs)
variant_contig_names = list(variant_contig_names)
vars_to_rechunk = []
vars_to_copy = []
with tempfile.TemporaryDirectory(prefix="vcfzarr_to_zarr_",
suffix=".zarr",
dir=tempdir) as tmpdir:
zarr_files = []
for i, contig in enumerate(contigs):
# convert contig group to zarr and save in tmpdir
ds = _vcfzarr_to_dataset(vcfzarr[contig], contig,
variant_contig_names)
if i == 0:
for (var, arr) in ds.data_vars.items():
if arr.dims[0] == "variants":
vars_to_rechunk.append(var)
else:
vars_to_copy.append(var)
contig_zarr_file = Path(tmpdir) / contig
ds.to_zarr(contig_zarr_file)
zarr_files.append(str(contig_zarr_file))
if concat_algorithm == "xarray_internal":
ds = zarrs_to_dataset(zarr_files)
ds.to_zarr(output, mode="w")
else:
# Use the optimized algorithm in `concatenate_and_rechunk`
_concat_zarrs_optimized(zarr_files, output, vars_to_rechunk,
vars_to_copy)
def xds_from_zarr(store, columns=None, chunks=None, **kwargs):
"""
Reads the zarr data store in `store` and returns list of
Dataset's containing the data.
Parameters
----------
store : str or Path
Path containing the data
columns : list of str or str or None
Columns to read. `None` or `"ALL"` stores all columns on each dataset.
Otherwise, a list of columns should be supplied.
chunks: dict or list of dicts
chunking schema for each dataset
**kwargs: optional
Returns
-------
writes : Dataset or list of Datasets
Dataset(s) representing write operations
"""
if isinstance(store, DaskMSStore):
pass
elif isinstance(store, (Path, str)):
store = DaskMSStore(f"{store}", **kwargs.pop("storage_options", {}))
else:
raise TypeError(f"store '{store}' must be "
f"Path, str or DaskMSStore")
# If any kwargs are added, they should be popped prior to this check.
if len(kwargs) > 0:
warnings.warn(
f"The following unsupported kwargs were ignored in "
f"xds_from_zarr: {kwargs}", UserWarning)
columns = promote_columns(columns)
if chunks is None:
pass
elif isinstance(chunks, (tuple, list)):
if not all(isinstance(v, dict) for v in chunks):
raise TypeError("chunks must be None, a dict or a list of dicts")
elif isinstance(chunks, dict):
chunks = [chunks]
else:
raise TypeError("chunks must be None, a dict or a list of dicts")
datasets = []
numpy_vars = []
# NOTE(JSKenyon): Iterating over all the zarr groups/arrays is VERY
# expensive if the metadata has not been consolidated.
zc.consolidate_metadata(store.map)
table_path = store.table if store.table else "MAIN"
table_group = zarr.open_consolidated(store.map)[table_path]
for g, (group_name,
group) in enumerate(sorted(table_group.groups(),
key=group_sortkey)):
group_attrs = decode_attr(dict(group.attrs))
dask_ms_attrs = group_attrs.pop(DASKMS_ATTR_KEY)
natural_chunks = dask_ms_attrs["chunks"]
group_chunks = {d: tuple(dc) for d, dc in natural_chunks.items()}
if chunks:
# Defer to user-supplied chunking strategy
try:
group_chunks.update(chunks[g])
except IndexError:
group_chunks.update(chunks[-1]) # Reuse last chunking.
pass
data_vars = {}
coords = {}
for name, zarray in column_iterator(group, columns):
attrs = decode_attr(dict(zarray.attrs[DASKMS_ATTR_KEY]))
dims = attrs["dims"]
coordinate = attrs.get("coordinate", False)
array_chunks = tuple(
group_chunks.get(d, s) for d, s in zip(dims, zarray.shape))
array_chunks = da.core.normalize_chunks(array_chunks, zarray.shape)
ext_args = extent_args(dims, array_chunks)
token_name = f"read~{name}-{tokenize(zarray, *ext_args)}"
read = da.blockwise(zarr_getter,
dims,
zarray,
None,
*ext_args,
concatenate=False,
name=token_name,
meta=np.empty((0, ) * zarray.ndim,
zarray.dtype))
read = inlined_array(read, ext_args[::2])
var = Variable(dims, read, attrs)
(coords if coordinate else data_vars)[name] = var
# Save numpy arrays for reification
typ = decode_type(attrs["array_type"])
if typ is np.ndarray:
numpy_vars.append(var)
elif typ is da.Array:
pass
else:
raise TypeError(f"Unknown array_type '{attrs['array_type']}'")
datasets.append(Dataset(data_vars, coords=coords, attrs=group_attrs))
# Reify any numpy arrays directly into their variables
for v, a in zip(numpy_vars, dask.compute(v.data for v in numpy_vars)[0]):
v.data = a
return datasets
