如何将嵌套的namedtuple转换为dict？

Question

我试图将下面提到的嵌套namedtuple转换为dict

我正在使用Python 3.4.2和{}

TypeToolObjectSetting(version=1, xx=0.0, xy=-1.55729984301413, yx=1.6070307595731337, yy=0.0, tx=628.1016949152543, ty=516.5, text_version=50, descriptor1_version=16, text_data=Descriptor(name='', classID=b'TxLr', items=[(b'Txt ', String(value='34px')), (b'textGridding', Enum(type=b'textGridding', value=b'None')), (b'Ornt', Enum(type=b'Ornt', value=b'Hrzn')), (b'AntA', Enum(type=b'Annt', value=b'Anno')), (b'bounds', Descriptor(name='', classID=b'bounds', items=[(b'Left', UnitFloat(unit='POINTS', value=-10.0)), (b'Top ', UnitFloat(unit='POINTS', value=-6.908203125)), (b'Rght', UnitFloat(unit='POINTS', value=10.0)), (b'Btom', UnitFloat(unit='POINTS', value=2.42578125))])), (b'boundingBox', Descriptor(name='', classID=b'boundingBox', items=[(b'Left', UnitFloat(unit='POINTS', value=-9.34375)), (b'Top ', UnitFloat(unit='POINTS', value=-5.9375)), (b'Rght', UnitFloat(unit='POINTS', value=9.5)), (b'Btom', UnitFloat(unit='POINTS', value=1.609375))])), (b'TextIndex', Integer(value=0)), (b'EngineData', RawData(value=b'\n\n<<\n\t/EngineDict\n\t<<\n\t\t/Edito ... =8205'))]), warp_version=1, descriptor2_version=16, warp_data=Descriptor(name='', classID=b'warp', items=[(b'warpStyle', Enum(type=b'warpStyle', value=b'warpNone')), (b'warpValue', Double(value=0.0)), (b'warpPerspective', Double(value=0.0)), (b'warpPerspectiveOther', Double(value=0.0)), (b'warpRotate', Enum(type=b'Ornt', value=b'Hrzn'))]), left=0, top=0, right=0, bottom=0)

我尝试过_asdict()，但它没有用，因为数据正在流式传输，我不想更改库中的任何内容。在

Answer 1

通过递归地解压._asdict().items()，检查值类型的实例并相应地解包嵌套的值，可以取消namedtuples。instance checking of a namedtuple需要做一些额外的工作，以确保嵌套的namedtuples不会被解包程序视为纯元组。在

def isnamedtupleinstance(x):
    _type = type(x)
    bases = _type.__bases__
    if len(bases) != 1 or bases[0] != tuple:
        return False
    fields = getattr(_type, '_fields', None)
    if not isinstance(fields, tuple):
        return False
    return all(type(i)==str for i in fields)

def unpack(obj):
    if isinstance(obj, dict):
        return {key: unpack(value) for key, value in obj.items()}
    elif isinstance(obj, list):
        return [unpack(value) for value in obj]
    elif isnamedtupleinstance(obj):
        return {key: unpack(value) for key, value in obj._asdict().items()}
    elif isinstance(obj, tuple):
        return tuple(unpack(value) for value in obj)
    else:
        return obj

# data = TypeToolObjectSetting(version=1, xx=0.0, ..
unpacked_data = unpack(data)