Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Optional/Nullable support to Python Cluster Objects #11515

Merged
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
78 changes: 48 additions & 30 deletions src/app/zap-templates/templates/app/helper.js
Original file line number Diff line number Diff line change
Expand Up @@ -415,50 +415,50 @@ function zapTypeToDecodableClusterObjectType(type, options)
return zapTypeToClusterObjectType.call(this, type, true, options)
}

function zapTypeToPythonClusterObjectType(type, options)
async function _zapTypeToPythonClusterObjectType(type, options)
{
if (StringHelper.isCharString(type)) {
return 'str';
}

if (StringHelper.isOctetString(type)) {
return 'bytes';
}

if ([ 'single', 'double' ].includes(type.toLowerCase())) {
return 'float';
}

if (type.toLowerCase() == 'boolean') {
return 'bool'
}

// #10748: asUnderlyingZclType will emit wrong types for int{48|56|64}(u), so we process all int values here.
if (type.toLowerCase().match(/^int\d+$/)) {
return 'int'
}

if (type.toLowerCase().match(/^int\d+u$/)) {
return 'uint'
}

async function fn(pkgId)
{
const ns = asUpperCamelCase(options.hash.ns);
const ns = options.hash.ns;
const typeChecker = async (method) => zclHelper[method](this.global.db, type, pkgId).then(zclType => zclType != 'unknown');

if (await typeChecker('isEnum')) {
return ns + '.Enums.' + type;
}

if (await typeChecker('isBitmap')) {
return 'int';
return 'uint';
}

if (await typeChecker('isStruct')) {
return ns + '.Structs.' + type;
}

if (StringHelper.isCharString(type)) {
return 'str';
}

if (StringHelper.isOctetString(type)) {
return 'bytes';
}

if ([ 'single', 'double' ].includes(type.toLowerCase())) {
return 'float';
}

if (type.toLowerCase() == 'boolean') {
return 'bool'
}

// #10748: asUnderlyingZclType will emit wrong types for int{48|56|64}(u), so we process all int values here.
if (type.toLowerCase().match(/^int\d+$/)) {
return 'int'
}

if (type.toLowerCase().match(/^int\d+u$/)) {
return 'uint'
}

resolvedType = await zclHelper.asUnderlyingZclType.call({ global : this.global }, type, options);
{
basicType = ChipTypesHelper.asBasicType(resolvedType);
Expand All @@ -469,14 +469,32 @@ function zapTypeToPythonClusterObjectType(type, options)
return 'uint'
}
}
}

throw "Unhandled type " + resolvedType + " (from " + type + ")"
let promise = templateUtil.ensureZclPackageId(this).then(fn.bind(this));
if ((this.isList || this.isArray || this.entryType) && !options.hash.forceNotList) {
promise = promise.then(typeStr => `typing.List[${typeStr}]`);
}

const isNull = (this.isNullable && !options.hash.forceNotNullable);
const isOptional = (this.isOptional && !options.hash.forceNotOptional);

if (isNull && isOptional) {
promise = promise.then(typeStr => `typing.Union[None, Nullable, ${typeStr}]`);
} else if (isNull) {
promise = promise.then(typeStr => `typing.Union[Nullable, ${typeStr}]`);
} else if (isOptional) {
promise = promise.then(typeStr => `typing.Optional[${typeStr}]`);
}

const promise = templateUtil.ensureZclPackageId(this).then(fn.bind(this));
return templateUtil.templatePromise(this.global, promise)
}

function zapTypeToPythonClusterObjectType(type, options)
{
return _zapTypeToPythonClusterObjectType.call(this, type, options)
}

async function getResponseCommandName(responseRef, options)
{
let pkgId = await templateUtil.ensureZclPackageId(this);
Expand Down
1 change: 1 addition & 0 deletions src/controller/python/BUILD.gn
Original file line number Diff line number Diff line change
Expand Up @@ -118,6 +118,7 @@ pw_python_action("python") {
"chip/clusters/Command.py",
"chip/clusters/Objects.py",
"chip/clusters/TestObjects.py",
"chip/clusters/Types.py",
"chip/clusters/__init__.py",
"chip/configuration/__init__.py",
"chip/discovery/__init__.py",
Expand Down
1 change: 1 addition & 0 deletions src/controller/python/build-chip-wheel.py
Original file line number Diff line number Diff line change
Expand Up @@ -118,6 +118,7 @@ def finalize_options(self):
'construct',
'ipython',
'dacite',
'rich',
]

if platform.system() == 'Darwin':
Expand Down
127 changes: 96 additions & 31 deletions src/controller/python/chip/clusters/ClusterObjects.py
Original file line number Diff line number Diff line change
Expand Up @@ -17,44 +17,92 @@

from dataclasses import dataclass, asdict, field, make_dataclass
from typing import ClassVar, List, Dict, Any, Mapping, Type, Union, ClassVar
import enum
import typing
from chip import tlv, ChipUtility
from chip.clusters.Types import Nullable, NullValue
from dacite import from_dict


def GetUnionUnderlyingType(typeToCheck, matchingType=None):
''' This retrieves the underlying types behind a unioned type by appropriately
passing in the required matching type in the matchingType input argument.

If that is 'None' (not to be confused with NoneType), then it will retrieve
the 'real' type behind the union, i.e not Nullable && not None
'''
if (not(typing.get_origin(typeToCheck) == typing.Union)):
return None

for t in typing.get_args(typeToCheck):
if (matchingType is None):
if (t != type(None) and t != Nullable):
return t
else:
if (t == matchingType):
return t

return None


@dataclass
class ClusterObjectFieldDescriptor:
Label: str = ''
Tag: int = None
Type: Type = None
IsArray: bool = False

def _PutSingleElementToTLV(self, tag, val, writer: tlv.TLVWriter, debugPath: str = '?'):
if issubclass(self.Type, ClusterObject):
def _PutSingleElementToTLV(self, tag, val, elementType, writer: tlv.TLVWriter, debugPath: str = '?'):
if issubclass(elementType, ClusterObject):
if not isinstance(val, dict):
raise ValueError(
f"Field {debugPath}.{self.Label} expected a struct, but got {type(val)}")
self.Type.descriptor.DictToTLVWithWriter(
elementType.descriptor.DictToTLVWithWriter(
f'{debugPath}.{self.Label}', tag, val, writer)
return

try:
val = self.Type(val)
val = elementType(val)
except Exception:
raise ValueError(
f"Field {debugPath}.{self.Label} expected {self.Type}, but got {type(val)}")
f"Field {debugPath}.{self.Label} expected {elementType}, but got {type(val)}")
writer.put(tag, val)

def PutFieldToTLV(self, tag, val, writer: tlv.TLVWriter, debugPath: str = '?'):
if not self.IsArray:
self._PutSingleElementToTLV(tag, val, writer, debugPath)
return
if not isinstance(val, List):
raise ValueError(
f"Field {debugPath}.{self.Label} expected List[{self.Type}], but got {type(val)}")
writer.startArray(tag)
for i, v in enumerate(val):
self._PutSingleElementToTLV(
None, v, writer, debugPath + f'[{i}]')
writer.endContainer()
if (val == NullValue):
if (GetUnionUnderlyingType(self.Type, Nullable) is None):
raise ValueError(
f"Field {debugPath}.{self.Label} was not nullable, but got a null")

writer.put(tag, None)
elif (val is None):
if (GetUnionUnderlyingType(self.Type, type(None)) is None):
raise ValueError(
f"Field {debugPath}.{self.Label} was not optional, but encountered None")
else:
#
# If it is an optional or nullable type, it's going to be a union.
# So, let's get at the 'real' type within that union before proceeding,
# since at this point, we're guarenteed to not get None or Null as values.
#
elementType = GetUnionUnderlyingType(self.Type)
if (elementType is None):
elementType = self.Type

if not isinstance(val, List):
self._PutSingleElementToTLV(
tag, val, elementType, writer, debugPath)
return

writer.startArray(tag)

# Get the type of the list. This is a generic, which has its sub-type information of the list element
# inside its type argument.
(elementType, ) = typing.get_args(elementType)

for i, v in enumerate(val):
self._PutSingleElementToTLV(
None, v, elementType, writer, debugPath + f'[{i}]')
writer.endContainer()


@dataclass
Expand All @@ -73,16 +121,19 @@ def GetFieldByLabel(self, label: str) -> ClusterObjectFieldDescriptor:
return field
return None

def _ConvertNonArray(self, debugPath: str, descriptor: ClusterObjectFieldDescriptor, value: Any) -> Any:
if not issubclass(descriptor.Type, ClusterObject):
if not isinstance(value, descriptor.Type):
def _ConvertNonArray(self, debugPath: str, elementType, value: Any) -> Any:
if not issubclass(elementType, ClusterObject):
if (issubclass(elementType, enum.Enum)):
value = elementType(value)

if not isinstance(value, elementType):
raise ValueError(
f"Failed to decode field {debugPath}, expected type {descriptor.Type}, got {type(value)}")
f"Failed to decode field {debugPath}, expected type {elementType}, got {type(value)}")
return value
if not isinstance(value, Mapping):
raise ValueError(
f"Failed to decode field {debugPath}, struct expected.")
return descriptor.Type.descriptor.TagDictToLabelDict(debugPath, value)
return elementType.descriptor.TagDictToLabelDict(debugPath, value)

def TagDictToLabelDict(self, debugPath: str, tlvData: Dict[int, Any]) -> Dict[str, Any]:
ret = {}
Expand All @@ -92,13 +143,30 @@ def TagDictToLabelDict(self, debugPath: str, tlvData: Dict[int, Any]) -> Dict[st
# We do not have enough infomation for this field.
ret[tag] = value
continue
if descriptor.IsArray:

if (value is None):
ret[descriptor.Label] = NullValue
continue

if (typing.get_origin(descriptor.Type) == typing.Union):
realType = GetUnionUnderlyingType(descriptor.Type)
if (realType is None):
raise ValueError(
f"Field {debugPath}.{self.Label} has no valid underlying data model type")

valueType = realType
else:
valueType = descriptor.Type

if (typing.get_origin(valueType) == list):
listElementType = typing.get_args(valueType)[0]
ret[descriptor.Label] = [
self._ConvertNonArray(f'{debugPath}[{i}]', descriptor, v)
self._ConvertNonArray(
f'{debugPath}[{i}]', listElementType, v)
for i, v in enumerate(value)]
continue
ret[descriptor.Label] = self._ConvertNonArray(
f'{debugPath}.{descriptor.Label}', descriptor, value)
f'{debugPath}.{descriptor.Label}', valueType, value)
return ret

def TLVToDict(self, tlvBuf: bytes) -> Dict[str, Any]:
Expand All @@ -109,9 +177,6 @@ def DictToTLVWithWriter(self, debugPath: str, tag, data: Mapping, writer: tlv.TL
writer.startStructure(tag)
for field in self.Fields:
val = data.get(field.Label, None)
if val is None:
raise ValueError(
f"Field {debugPath}.{field.Label} is missing in the given dict")
field.PutFieldToTLV(field.Tag, val, writer,
debugPath + f'.{field.Label}')
writer.endContainer()
Expand Down Expand Up @@ -198,13 +263,13 @@ def attribute_type(cls) -> ClusterObjectFieldDescriptor:
def _cluster_object(cls) -> ClusterObject:
return make_dataclass('InternalClass',
[
('Value', List[cls.attribute_type.Type]
if cls.attribute_type.IsArray else cls.attribute_type.Type, field(default=None)),
('Value', cls.attribute_type.Type,
field(default=None)),
('descriptor', ClassVar[ClusterObjectDescriptor],
field(
default=ClusterObjectDescriptor(
Fields=[ClusterObjectFieldDescriptor(
Label='Value', Tag=0, Type=cls.attribute_type.Type, IsArray=cls.attribute_type.IsArray)]
Label='Value', Tag=0, Type=cls.attribute_type.Type)]
)
)
)
Expand Down
Loading