default fill_value 0; refactor errors; resolves #49

Author: alimanfoo

docs/tutorial.rst

@@ -230,7 +230,7 @@ the delta filter::
     ...                chunks=(1000, 1000), compressor=compressor)
     >>> z
     Array((10000, 10000), int32, chunks=(1000, 1000), order=C)
-      nbytes: 381.5M; nbytes_stored: 248.9K; ratio: 1569.6; initialized: 100/100
+      nbytes: 381.5M; nbytes_stored: 248.9K; ratio: 1569.7; initialized: 100/100
       compressor: LZMA(format=1, check=-1, preset=None, filters=[{'dist': 4, 'id': 3}, {'preset': 1, 'id': 33}])
       store: dict
 
@@ -327,7 +327,7 @@ provided that all processes have access to a shared file system. E.g.::
     ...                     synchronizer=synchronizer)
     >>> z
     Array((10000, 10000), int32, chunks=(1000, 1000), order=C)
-      nbytes: 381.5M; nbytes_stored: 326; ratio: 1226993.9; initialized: 0/100
+      nbytes: 381.5M; nbytes_stored: 323; ratio: 1238390.1; initialized: 0/100
       compressor: Blosc(cname='lz4', clevel=5, shuffle=1)
       store: DirectoryStore; synchronizer: ProcessSynchronizer
 

zarr/core.py

@@ -13,7 +13,7 @@
 from zarr.storage import array_meta_key, attrs_key, listdir, getsize
 from zarr.meta import decode_array_metadata, encode_array_metadata
 from zarr.attrs import Attributes
-from zarr.errors import PermissionError
+from zarr.errors import PermissionError, err_read_only, err_array_not_found
 from zarr.compat import reduce
 from zarr.codecs import get_codec
 
@@ -108,7 +108,7 @@ def _load_metadata(self):
             mkey = self._key_prefix + array_meta_key
             meta_bytes = self._store[mkey]
         except KeyError:
-            raise ValueError('store has no metadata')
+            err_array_not_found(self._path)
         else:
 
             # decode and store metadata
@@ -135,7 +135,7 @@ def _load_metadata(self):
 
     def _flush_metadata(self):
         if self._is_view:
-            raise PermissionError('operation not permitted for views')
+            raise PermissionError('not permitted for views')
 
         if self._compressor:
             compressor_config = self._compressor.get_config()
@@ -496,7 +496,7 @@ def __setitem__(self, key, value):
 
         # guard conditions
         if self._read_only:
-            raise PermissionError('array is read-only')
+            err_read_only()
 
         # refresh metadata
         if not self._cache_metadata:
@@ -790,7 +790,7 @@ def _write_op(self, f, *args, **kwargs):
 
         # guard condition
         if self._read_only:
-            raise PermissionError('array is read-only')
+            err_read_only()
 
         # synchronization
         if self._synchronizer is None:
@@ -1057,7 +1057,7 @@ def view(self, shape=None, chunks=None, dtype=None,
             ...     v.resize(20000)
             ... except PermissionError as e:
             ...     print(e)
-            operation not permitted for views
+            not permitted for views
 
         """  # flake8: noqa
 

zarr/creation.py

@@ -8,12 +8,14 @@
 
 from zarr.core import Array
 from zarr.storage import DirectoryStore, init_array, contains_array, \
-    contains_group, default_compressor, _require_parent_group
+    contains_group, default_compressor, normalize_storage_path
 from zarr.codecs import codec_registry
+from zarr.errors import err_contains_array, err_contains_group, \
+    err_array_not_found
 
 
 def create(shape, chunks=None, dtype=None, compressor='default',
-           fill_value=None, order='C', store=None, synchronizer=None,
+           fill_value=0, order='C', store=None, synchronizer=None,
            overwrite=False, path=None, chunk_store=None, filters=None,
            cache_metadata=True, **kwargs):
     """Create an array.
@@ -65,7 +67,7 @@ def create(shape, chunks=None, dtype=None, compressor='default',
         >>> z = zarr.create((10000, 10000), chunks=(1000, 1000))
         >>> z
         Array((10000, 10000), float64, chunks=(1000, 1000), order=C)
-          nbytes: 762.9M; nbytes_stored: 326; ratio: 2453987.7; initialized: 0/100
+          nbytes: 762.9M; nbytes_stored: 323; ratio: 2476780.2; initialized: 0/100
           compressor: Blosc(cname='lz4', clevel=5, shuffle=1)
           store: dict
 
@@ -295,7 +297,7 @@ def array(data, **kwargs):
 
 
 def open_array(store=None, mode='a', shape=None, chunks=None, dtype=None,
-               compressor='default', fill_value=None, order='C',
+               compressor='default', fill_value=0, order='C',
                synchronizer=None, filters=None, cache_metadata=True,
                path=None, **kwargs):
     """Open array using mode-like semantics.
@@ -374,6 +376,7 @@ def open_array(store=None, mode='a', shape=None, chunks=None, dtype=None,
 
     # handle polymorphic store arg
     store = _handle_store_arg(store)
+    path = normalize_storage_path(path)
 
     # compatibility
     compressor, fill_value = _handle_kwargs(compressor, fill_value, kwargs)
@@ -382,9 +385,9 @@ def open_array(store=None, mode='a', shape=None, chunks=None, dtype=None,
 
     if mode in ['r', 'r+']:
         if contains_group(store, path=path):
-            raise ValueError('store contains group')
+            err_contains_group(path)
         elif not contains_array(store, path=path):
-            raise ValueError('array does not exist')
+            err_array_not_found(path)
 
     elif mode == 'w':
         init_array(store, shape=shape, chunks=chunks, dtype=dtype,
@@ -393,17 +396,17 @@ def open_array(store=None, mode='a', shape=None, chunks=None, dtype=None,
 
     elif mode == 'a':
         if contains_group(store, path=path):
-            raise ValueError('store contains group')
+            err_contains_group(path)
         elif not contains_array(store, path=path):
             init_array(store, shape=shape, chunks=chunks, dtype=dtype,
                        compressor=compressor, fill_value=fill_value,
                        order=order, filters=filters, path=path)
 
     elif mode in ['w-', 'x']:
         if contains_group(store, path=path):
-            raise ValueError('store contains group')
+            err_contains_group(path)
         elif contains_array(store, path=path):
-            raise ValueError('store contains array')
+            err_contains_array(path)
         else:
             init_array(store, shape=shape, chunks=chunks, dtype=dtype,
                        compressor=compressor, fill_value=fill_value,

zarr/errors.py

@@ -17,3 +17,36 @@ class PermissionError(Exception):
 
 class MetadataError(Exception):
     pass
+
+
+def err_contains_group(path):
+    raise KeyError('path %r contains a group' % path)
+
+
+def err_contains_array(path):
+    raise KeyError('path %r contains an array' % path)
+
+
+def err_array_not_found(path):
+    raise KeyError('array not found at path %r' % path)
+
+
+def err_group_not_found(path):
+    raise KeyError('group not found at path %r' % path)
+
+
+def err_path_not_found(path):
+    raise KeyError('path %r not found' % path)
+
+
+def err_bad_compressor(compressor):
+    raise ValueError('bad compressor; expected Codec object, found %r' %
+                     compressor)
+
+
+def err_fspath_exists_notdir(fspath):
+    raise ValueError('path exists but is not a directory: %r' % fspath)
+
+
+def err_read_only():
+    raise PermissionError('object is read-only')

zarr/hierarchy.py

@@ -9,11 +9,12 @@
 from zarr.attrs import Attributes
 from zarr.core import Array
 from zarr.storage import contains_array, contains_group, init_group, \
-    DictStore, DirectoryStore, group_meta_key, attrs_key, listdir, rmdir
+    DictStore, DirectoryStore, group_meta_key, attrs_key, listdir
 from zarr.creation import array, create, empty, zeros, ones, full, \
     empty_like, zeros_like, ones_like, full_like
 from zarr.util import normalize_storage_path, normalize_shape
-from zarr.errors import PermissionError
+from zarr.errors import PermissionError, err_contains_array, \
+    err_contains_group, err_group_not_found, err_read_only
 from zarr.meta import decode_group_metadata
 
 
@@ -91,14 +92,14 @@ def __init__(self, store, path=None, read_only=False, chunk_store=None,
 
         # guard conditions
         if contains_array(store, path=self._path):
-            raise ValueError('store contains an array')
+            err_contains_array(path)
 
         # initialize metadata
         try:
             mkey = self._key_prefix + group_meta_key
             meta_bytes = store[mkey]
         except KeyError:
-            raise ValueError('store has no metadata')
+            err_group_not_found(path)
         else:
             meta = decode_group_metadata(meta_bytes)
             self._meta = meta
@@ -285,7 +286,7 @@ def __getitem__(self, item):
           store: DictStore
         >>> g1['foo/bar/baz']
         Array(/foo/bar/baz, (100,), float64, chunks=(10,), order=C)
-          nbytes: 800; nbytes_stored: 293; ratio: 2.7; initialized: 0/10
+          nbytes: 800; nbytes_stored: 290; ratio: 2.8; initialized: 0/10
           compressor: Blosc(cname='lz4', clevel=5, shuffle=1)
           store: DictStore
 
@@ -396,7 +397,7 @@ def _write_op(self, f, *args, **kwargs):
 
         # guard condition
         if self._read_only:
-            raise PermissionError('group is read-only')
+            err_read_only()
 
         # synchronization
         if self._synchronizer is None:
@@ -494,7 +495,7 @@ def require_groups(self, *names):
         return tuple(self.require_group(name) for name in names)
 
     def create_dataset(self, name, data=None, shape=None, chunks=None,
-                       dtype=None, compressor='default', fill_value=None,
+                       dtype=None, compressor='default', fill_value=0,
                        order='C', synchronizer=None, filters=None,
                        overwrite=False, cache_metadata=True, **kwargs):
         """Create an array.
@@ -543,7 +544,7 @@ def create_dataset(self, name, data=None, shape=None, chunks=None,
         ...                        chunks=(1000, 1000))
         >>> d1
         Array(/foo, (10000, 10000), float64, chunks=(1000, 1000), order=C)
-          nbytes: 762.9M; nbytes_stored: 326; ratio: 2453987.7; initialized: 0/100
+          nbytes: 762.9M; nbytes_stored: 323; ratio: 2476780.2; initialized: 0/100
           compressor: Blosc(cname='lz4', clevel=5, shuffle=1)
           store: DictStore
 
@@ -558,7 +559,7 @@ def create_dataset(self, name, data=None, shape=None, chunks=None,
 
     def _create_dataset_nosync(self, name, data=None, shape=None, chunks=None,
                                dtype=None, compressor='default',
-                               fill_value=None, order='C', synchronizer=None,
+                               fill_value=0, order='C', synchronizer=None,
                                filters=None, overwrite=False,
                                cache_metadata=True, **kwargs):
 
@@ -804,6 +805,7 @@ def group(store=None, overwrite=False, chunk_store=None, synchronizer=None,
 
     # handle polymorphic store arg
     store = _handle_store_arg(store)
+    path = normalize_storage_path(path)
 
     # require group
     if overwrite or not contains_group(store):
@@ -857,29 +859,30 @@ def open_group(store=None, mode='a', synchronizer=None, path=None):
 
     # handle polymorphic store arg
     store = _handle_store_arg(store)
+    path = normalize_storage_path(path)
 
     # ensure store is initialized
 
     if mode in ['r', 'r+']:
         if contains_array(store, path=path):
-            raise ValueError('store contains array')
+            err_contains_array(path)
         elif not contains_group(store, path=path):
-            raise ValueError('group does not exist')
+            err_group_not_found(path)
 
     elif mode == 'w':
         init_group(store, overwrite=True, path=path)
 
     elif mode == 'a':
         if contains_array(store, path=path):
-            raise ValueError('store contains array')
+            err_contains_array(path)
         if not contains_group(store, path=path):
             init_group(store, path=path)
 
     elif mode in ['w-', 'x']:
         if contains_array(store, path=path):
-            raise ValueError('store contains array')
+            err_contains_array(path)
         elif contains_group(store, path=path):
-            raise ValueError('store contains group')
+            err_contains_group(path)
         else:
             init_group(store, path=path)
 

zarr/storage.py

@@ -16,7 +16,9 @@
 from zarr.meta import encode_array_metadata, encode_group_metadata
 from zarr.compat import PY2, binary_type
 from zarr.codecs import codec_registry
-from zarr.errors import PermissionError
+from zarr.errors import PermissionError, err_contains_group, \
+    err_contains_array, err_path_not_found, err_bad_compressor, \
+    err_fspath_exists_notdir, err_read_only
 
 
 array_meta_key = '.zarray'
@@ -267,9 +269,9 @@ def _init_array_metadata(store, shape, chunks, dtype=None,
         if chunk_store is not None and chunk_store != store:
             rmdir(chunk_store, path)
     elif contains_array(store, path):
-        raise KeyError('path %r contains an array' % path)
+        err_contains_array(path)
     elif contains_group(store, path):
-        raise KeyError('path %r contains a group' % path)
+        err_contains_group(path)
 
     # normalize metadata
     shape = normalize_shape(shape)
@@ -287,8 +289,7 @@ def _init_array_metadata(store, shape, chunks, dtype=None,
         try:
             compressor_config = compressor.get_config()
         except AttributeError:
-            raise ValueError('bad compressor argument; expected Codec object, '
-                             'found %r' % compressor)
+            err_bad_compressor(compressor)
     else:
         compressor_config = None
 
@@ -352,9 +353,9 @@ def _init_group_metadata(store, overwrite=False, path=None, chunk_store=None):
         if chunk_store is not None and chunk_store != store:
             rmdir(chunk_store, path)
     elif contains_array(store, path):
-        raise KeyError('path %r contains an array' % path)
+        err_contains_array(path)
     elif contains_group(store, path):
-        raise KeyError('path %r contains a group' % path)
+        err_contains_group(path)
 
     # initialize metadata
     # N.B., currently no metadata properties are needed, however there may
@@ -532,7 +533,7 @@ def getsize(self, path=None):
                 parent, key = self._get_parent(path)
                 value = parent[key]
             except KeyError:
-                raise KeyError('path %r not found' % path)
+                err_path_not_found(path)
         else:
             value = self.root
 
@@ -597,7 +598,7 @@ def __init__(self, path):
         # guard conditions
         path = os.path.abspath(path)
         if os.path.exists(path) and not os.path.isdir(path):
-            raise ValueError('path exists but is not a directory')
+            err_fspath_exists_notdir(path)
 
         self.path = path
 
@@ -713,7 +714,7 @@ def getsize(self, path=None):
                     size += os.path.getsize(child_fs_path)
             return size
         else:
-            raise KeyError('path %r not found' % path)
+            err_path_not_found(path)
 
 
 # noinspection PyPep8Naming
@@ -776,7 +777,7 @@ def __getitem__(self, key):
 
     def __setitem__(self, key, value):
         if self.mode == 'r':
-            raise PermissionError('mapping is read-only')
+            err_read_only()
         value = ensure_bytes(value)
         with zipfile.ZipFile(self.path, mode='a',
                              compression=self.compression,
@@ -846,7 +847,7 @@ def getsize(self, path=None):
                     info = zf.getinfo(path)
                     return info.compress_size
                 except KeyError:
-                    raise KeyError('path %r not found' % path)
+                    err_path_not_found(path)
             else:
                 return 0