encode infinity fill value; spec tweaks

Author: alimanfoo

docs/spec/v2.rst

@@ -103,14 +103,23 @@ library::
 Data type encoding
 ~~~~~~~~~~~~~~~~~~
 
-Simple data types are encoded within the array metadata resource as a string, 
+Simple data types are encoded within the array metadata as a string,
 following the `NumPy array protocol type string (typestr) format 
 <http://docs.scipy.org/doc/numpy/reference/arrays.interface.html>`_. The format 
-consists of 3 parts: a character describing the byteorder of the data (``<``: 
-little-endian, ``>``: big-endian, ``|``: not-relevant), a character code giving 
-the basic type of the array, and an integer providing the number of bytes the 
-type uses. The byte order MUST be specified. E.g., ``"<f8"``, ``">i4"``, 
-``"|b1"`` and ``"|S12"`` are valid data types.
+consists of 3 parts:
+
+* One character describing the byteorder of the data (``"<"``: little-endian;
+  ``">"``: big-endian; ``"|"``: not-relevant)
+* One character code giving the basic type of the array (``"b"``: Boolean (integer
+  type where all values are only True or False); ``"i"``: integer; ``"u"``: unsigned
+  integer; ``"f"``: floating point; ``"c"``: complex floating point; ``"m"``: timedelta;
+  ``"M"``: datetime; ``"S"``: string (fixed-length sequence of char); ``"U"``: unicode
+  (fixed-length sequence of Py_UNICODE); ``"V"``: other (void * – each item is a
+  fixed-size chunk of memory))
+* An integer specifying the number of bytes the type uses.
+
+The byte order MUST be specified. E.g., ``"<f8"``, ``">i4"``, ``"|b1"`` and
+``"|S12"`` are valid data type encodings.
 
 Structured data types (i.e., with multiple named fields) are encoded as a list
 of two-element lists, following `NumPy array protocol type descriptions (descr) 
@@ -122,11 +131,17 @@ data type composed of three single-byte unsigned integers labelled "r", "g" and
 Fill value encoding
 ~~~~~~~~~~~~~~~~~~~
 
-Not a Number (NaN) MUST be encoded as the JSON string "NaN" if used as the
-value of the "fill_value" field.
+For simple floating point data types, the following table MUST be used to
+encode values of the "fill_value" field:
+
+=================  ===============
+Value              JSON encoding
+=================  ===============
+Not a Number       ``"NaN"``
+Positive Infinity  ``"Infinity"``
+Negative Infinity  ``"-Infinity"``
+=================  ===============
 
-When decoding the "fill_value" field, the JSON string "NaN" MUST be decoded
-as Not a Number (NaN) if the dtype basic type is floating point ("f").
 
 Chunks
 ~~~~~~

zarr/meta.py

@@ -102,20 +102,36 @@ def encode_group_metadata(meta=None):
     return b
 
 
+FLOAT_FILLS = {
+    'NaN': np.nan,
+    'Infinity': np.PINF,
+    '-Infinity': np.NINF
+}
+
+
 def decode_fill_value(v, dtype):
-    if v == 'NaN' and dtype.kind == 'f':
-        return np.nan
+    if dtype.kind == 'f':
+        if v == 'NaN':
+            return np.nan
+        elif v == 'Infinity':
+            return np.PINF
+        elif v == '-Infinity':
+            return np.NINF
+        else:
+            return v
     else:
         return v
 
 
 def encode_fill_value(v):
     try:
-        isnan = np.isnan(v)
-    except TypeError:
-        return v
-    else:
-        if isnan:
+        if np.isnan(v):
             return 'NaN'
+        elif np.isposinf(v):
+            return 'Infinity'
+        elif np.isneginf(v):
+            return '-Infinity'
         else:
             return v
+    except TypeError:
+        return v

zarr/tests/test_meta.py

@@ -106,37 +106,45 @@ def test_encode_decode_array_2():
     eq(meta['fill_value'], meta_dec['fill_value'])
 
 
-def test_encode_decode_array_nan_fill_value():
+def test_encode_decode_array_fill_values():
 
-    meta = dict(
-        shape=(100,),
-        chunks=(10,),
-        dtype=np.dtype('f8'),
-        compression='zlib',
-        compression_opts=1,
-        fill_value=np.nan,
-        order='C'
+    fills = (
+        (np.nan, "NaN", np.isnan),
+        (np.NINF, "-Infinity", np.isneginf),
+        (np.PINF, "Infinity", np.isposinf),
     )
 
-    meta_json = '''{
-        "chunks": [10],
-        "compression": "zlib",
-        "compression_opts": 1,
-        "dtype": "<f8",
-        "fill_value": "NaN",
-        "order": "C",
-        "shape": [100],
-        "zarr_format": %s
-    }''' % ZARR_FORMAT
-
-    # test encoding
-    meta_enc = encode_array_metadata(meta)
-    assert_json_eq(meta_json, meta_enc)
-
-    # test decoding
-    meta_dec = decode_array_metadata(meta_enc)
-    actual = meta_dec['fill_value']
-    assert np.isnan(actual)
+    for v, s, f in fills:
+
+        meta = dict(
+            shape=(100,),
+            chunks=(10,),
+            dtype=np.dtype('f8'),
+            compression='zlib',
+            compression_opts=1,
+            fill_value=v,
+            order='C'
+        )
+
+        meta_json = '''{
+            "chunks": [10],
+            "compression": "zlib",
+            "compression_opts": 1,
+            "dtype": "<f8",
+            "fill_value": "%s",
+            "order": "C",
+            "shape": [100],
+            "zarr_format": %s
+        }''' % (s, ZARR_FORMAT)
+
+        # test encoding
+        meta_enc = encode_array_metadata(meta)
+        assert_json_eq(meta_json, meta_enc)
+
+        # test decoding
+        meta_dec = decode_array_metadata(meta_enc)
+        actual = meta_dec['fill_value']
+        assert f(actual)
 
 
 def test_decode_array_unsupported_format():