docs: Update fill_value representation for fixed-point types

This commit updates the 'Fill value representation' section in all
fixed-point data type READMEs to clarify how fill values should be
handled.

The changes specify that:
- The `fill_value` SHOULD be a JSON number.
- For exact bit-level precision, it MAY be represented as a
  hexadecimal string (e.g., 0x0000).
- `NaN` and `Infinity` values are not applicable to these types.

Author: mkitti

data-types/fixed-point/N0f16/README.md

@@ -20,8 +20,9 @@ The stored `UInt16` value `i` is interpreted as `i / (2^16 - 1)`. This is equiva
 
 ## Fill value representation
 
-The `fill_value` for this data type should be represented as a floating-point number in the JSON metadata.
-
+The `fill_value` for this data type SHOULD be represented as a JSON number with the value to be represented.
+To represent the underlying integer bits exactly, the `fill_value` MAY be provided as a hexadecimal string representing the underlying integer (e.g., "0x0000" for a fill value of 0).
+There are no `NaN` or `Infinity` values for fixed-point types.
 ## Codec compatibility
 
 This data type is stored as an `uint16`. It is expected to be compatible with any codec that can handle the `uint16` data type.

data-types/fixed-point/N0f32/README.md

@@ -20,8 +20,9 @@ The stored `UInt32` value `i` is interpreted as `i / (2^32 - 1)`. This is equiva
 
 ## Fill value representation
 
-The `fill_value` for this data type should be represented as a floating-point number in the JSON metadata.
-
+The `fill_value` for this data type SHOULD be represented as a JSON number with the value to be represented.
+To represent the underlying integer bits exactly, the `fill_value` MAY be provided as a hexadecimal string representing the underlying integer (e.g., "0x00000000" for a fill value of 0).
+There are no `NaN` or `Infinity` values for fixed-point types.
 ## Codec compatibility
 
 This data type is stored as an `uint32`. It is expected to be compatible with any codec that can handle the `uint32` data type.

data-types/fixed-point/N0f64/README.md

@@ -20,8 +20,9 @@ The stored `UInt64` value `i` is interpreted as `i / (2^64 - 1)`. This is equiva
 
 ## Fill value representation
 
-The `fill_value` for this data type should be represented as a floating-point number in the JSON metadata.
-
+The `fill_value` for this data type SHOULD be represented as a JSON number with the value to be represented.
+To represent the underlying integer bits exactly, the `fill_value` MAY be provided as a hexadecimal string representing the underlying integer (e.g., "0x0000000000000000" for a fill value of 0).
+There are no `NaN` or `Infinity` values for fixed-point types.
 ## Codec compatibility
 
 This data type is stored as an `uint64`. It is expected to be compatible with any codec that can handle the `uint64` data type.

data-types/fixed-point/N0f8/README.md

@@ -20,8 +20,9 @@ The stored `UInt8` value `i` is interpreted as `i / (2^8 - 1)`. This is equivale
 
 ## Fill value representation
 
-The `fill_value` for this data type should be represented as a floating-point number in the JSON metadata. For example: `"fill_value": 0.5`.
-
+The `fill_value` for this data type SHOULD be represented as a JSON number with the value to be represented.
+To represent the underlying integer bits exactly, the `fill_value` MAY be provided as a hexadecimal string representing the underlying integer (e.g., "0x00" for a fill value of 0).
+There are no `NaN` or `Infinity` values for fixed-point types.
 ## Codec compatibility
 
 This data type is stored as a `uint8`. It is expected to be compatible with any codec that can handle the `uint8` data type.

data-types/fixed-point/N10f22/README.md

@@ -20,8 +20,9 @@ The stored `UInt32` value `i` is interpreted as `i / (2^22 - 1)`. This is equiva
 
 ## Fill value representation
 
-The `fill_value` for this data type should be represented as a floating-point number in the JSON metadata.
-
+The `fill_value` for this data type SHOULD be represented as a JSON number with the value to be represented.
+To represent the underlying integer bits exactly, the `fill_value` MAY be provided as a hexadecimal string representing the underlying integer (e.g., "0x00000000" for a fill value of 0).
+There are no `NaN` or `Infinity` values for fixed-point types.
 ## Codec compatibility
 
 This data type is stored as an `uint32`. It is expected to be compatible with any codec that can handle the `uint32` data type.

data-types/fixed-point/N10f54/README.md

@@ -20,8 +20,9 @@ The stored `UInt64` value `i` is interpreted as `i / (2^54 - 1)`. This is equiva
 
 ## Fill value representation
 
-The `fill_value` for this data type should be represented as a floating-point number in the JSON metadata.
-
+The `fill_value` for this data type SHOULD be represented as a JSON number with the value to be represented.
+To represent the underlying integer bits exactly, the `fill_value` MAY be provided as a hexadecimal string representing the underlying integer (e.g., "0x0000000000000000" for a fill value of 0).
+There are no `NaN` or `Infinity` values for fixed-point types.
 ## Codec compatibility
 
 This data type is stored as an `uint64`. It is expected to be compatible with any codec that can handle the `uint64` data type.

data-types/fixed-point/N10f6/README.md

@@ -20,8 +20,9 @@ The stored `UInt16` value `i` is interpreted as `i / (2^6 - 1)`. This is equival
 
 ## Fill value representation
 
-The `fill_value` for this data type should be represented as a floating-point number in the JSON metadata.
-
+The `fill_value` for this data type SHOULD be represented as a JSON number with the value to be represented.
+To represent the underlying integer bits exactly, the `fill_value` MAY be provided as a hexadecimal string representing the underlying integer (e.g., "0x0000" for a fill value of 0).
+There are no `NaN` or `Infinity` values for fixed-point types.
 ## Codec compatibility
 
 This data type is stored as an `uint16`. It is expected to be compatible with any codec that can handle the `uint16` data type.

data-types/fixed-point/N11f21/README.md

@@ -20,8 +20,9 @@ The stored `UInt32` value `i` is interpreted as `i / (2^21 - 1)`. This is equiva
 
 ## Fill value representation
 
-The `fill_value` for this data type should be represented as a floating-point number in the JSON metadata.
-
+The `fill_value` for this data type SHOULD be represented as a JSON number with the value to be represented.
+To represent the underlying integer bits exactly, the `fill_value` MAY be provided as a hexadecimal string representing the underlying integer (e.g., "0x00000000" for a fill value of 0).
+There are no `NaN` or `Infinity` values for fixed-point types.
 ## Codec compatibility
 
 This data type is stored as an `uint32`. It is expected to be compatible with any codec that can handle the `uint32` data type.

data-types/fixed-point/N11f5/README.md

@@ -20,8 +20,9 @@ The stored `UInt16` value `i` is interpreted as `i / (2^5 - 1)`. This is equival
 
 ## Fill value representation
 
-The `fill_value` for this data type should be represented as a floating-point number in the JSON metadata.
-
+The `fill_value` for this data type SHOULD be represented as a JSON number with the value to be represented.
+To represent the underlying integer bits exactly, the `fill_value` MAY be provided as a hexadecimal string representing the underlying integer (e.g., "0x0000" for a fill value of 0).
+There are no `NaN` or `Infinity` values for fixed-point types.
 ## Codec compatibility
 
 This data type is stored as an `uint16`. It is expected to be compatible with any codec that can handle the `uint16` data type.

data-types/fixed-point/N11f53/README.md

@@ -20,8 +20,9 @@ The stored `UInt64` value `i` is interpreted as `i / (2^53 - 1)`. This is equiva
 
 ## Fill value representation
 
-The `fill_value` for this data type should be represented as a floating-point number in the JSON metadata.
-
+The `fill_value` for this data type SHOULD be represented as a JSON number with the value to be represented.
+To represent the underlying integer bits exactly, the `fill_value` MAY be provided as a hexadecimal string representing the underlying integer (e.g., "0x0000000000000000" for a fill value of 0).
+There are no `NaN` or `Infinity` values for fixed-point types.
 ## Codec compatibility
 
 This data type is stored as an `uint64`. It is expected to be compatible with any codec that can handle the `uint64` data type.