1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
|
"""@package docstring
Encoding and decoding python native data structures as
portable JData-spec annotated dict structure
Copyright (c) 2019 Qianqian Fang <q.fang at neu.edu>
"""
__all__ = ['encode','decode','jdtype','jsonfilter']
##====================================================================================
## dependent libraries
##====================================================================================
import numpy as np
import copy
import zlib
import base64
try:
import lzma
except ImportError:
from backports import lzma
##====================================================================================
## global variables
##====================================================================================
""" @brief Mapping Numpy data types to JData data types
complex-valued data are reflected in the doubled data size
"""
jdtype={'float32':'single','float64':'double','float_':'double',
'bool':'uint8','byte':'int8','short':'int16','ubyte':'uint8',
'ushort':'uint16','int_':'int32','uint':'uint32','complex_':'double','complex128':'double',
'longlong':'int64','ulonglong':'uint64','csingle':'single','cdouble':'double'};
_zipper=['zlib','gzip','lzma'];
##====================================================================================
## Python to JData encoding function
##====================================================================================
def encode(d, opt={}):
"""@brief Encoding a Python data structure to portable JData-annotated dict constructs
This function converts complex data types (usually not JSON-serializable) into
portable JData-annotated dict/list constructs that can be easily exported as JSON/JData
files
@param[in,out] d: an arbitrary Python data
@param[in] opt: options, can contain 'compression'=['zlib','lzma','gzip'] for data compression
"""
if isinstance(d, float):
if(np.isnan(d)):
return '_NaN_';
elif(np.isinf(d)):
return '_Inf_' if (d>0) else '-_Inf_';
return d;
elif isinstance(d, list) or isinstance(d, tuple) or isinstance(d, set) or isinstance(d, frozenset):
return encodelist(d,opt);
elif isinstance(d, dict):
return encodedict(d,opt);
elif isinstance(d, complex):
newobj={
'_ArrayType_': 'double',
'_ArraySize_': 1,
'_ArrayIsComplex_': True,
'_ArrayData_': [d.real, d.imag]
};
return newobj;
elif isinstance(d, np.ndarray):
newobj={};
newobj["_ArrayType_"]=jdtype[str(d.dtype)] if (str(d.dtype) in jdtype) else str(d.dtype);
newobj["_ArraySize_"]=list(d.shape);
if(d.dtype==np.complex64 or d.dtype==np.complex128 or d.dtype==np.csingle or d.dtype==np.cdouble):
newobj['_ArrayIsComplex_']=True;
newobj['_ArrayData_']=[list(d.flatten().real), list(d.flatten().imag)];
else:
newobj["_ArrayData_"]=list(d.flatten());
if('compression' in opt):
if(opt['compression'] not in _zipper):
raise Exception('JData', 'compression method is not supported')
newobj['_ArrayZipType_']=opt['compression'];
newobj['_ArrayZipSize_']=[1+int('_ArrayIsComplex_' in newobj), d.size];
newobj['_ArrayZipData_']=np.asarray(newobj['_ArrayData_'],dtype=d.dtype).tostring();
if(opt['compression']=='zlib'):
newobj['_ArrayZipData_']=zlib.compress(newobj['_ArrayZipData_']);
elif(opt['compression']=='gzip'):
newobj['_ArrayZipData_']=zlib.compress(newobj['_ArrayZipData_'],zlib.MAX_WBITS|32);
elif(opt['compression']=='lzma'):
newobj['_ArrayZipData_']=lzma.compress(newobj['_ArrayZipData_'],lzma.FORMAT_ALONE);
if(('base64' in opt) and (opt['base64'])):
newobj['_ArrayZipData_']=base64.b64encode(newobj['_ArrayZipData_']);
newobj.pop('_ArrayData_');
return newobj;
else:
return copy.deepcopy(d);
##====================================================================================
## JData to Python decoding function
##====================================================================================
def decode(d, opt={}):
"""@brief Decoding a JData-annotated dict construct into native Python data
This function converts portable JData-annotated dict/list constructs back to native Python
data structures
@param[in,out] d: an arbitrary Python data, any JData-encoded components will be decoded
@param[in] opt: options
"""
if (isinstance(d, str) or type(d)=='unicode') and len(d)<=6 and len(d)>4 and d[-1]=='_':
if(d=='_NaN_'):
return float('nan');
elif(d=='_Inf_'):
return float('inf');
elif(d=='-_Inf_'):
return float('-inf');
return d;
elif isinstance(d, list) or isinstance(d, tuple) or isinstance(d, set) or isinstance(d, frozenset):
return decodelist(d,opt);
elif isinstance(d, dict):
if('_ArrayType_' in d):
if(isinstance(d['_ArraySize_'],str)):
d['_ArraySize_']=np.array(bytearray(d['_ArraySize_']));
if('_ArrayZipData_' in d):
newobj=d['_ArrayZipData_']
if(('base64' in opt) and (opt['base64'])):
newobj=base64.b64decode(newobj)
if('_ArrayZipType_' in d and d['_ArrayZipType_'] not in _zipper):
raise Exception('JData', 'compression method is not supported')
if(d['_ArrayZipType_']=='zlib'):
newobj=zlib.decompress(bytes(newobj))
elif(d['_ArrayZipType_']=='gzip'):
newobj=zlib.decompress(bytes(newobj),zlib.MAX_WBITS|32)
elif(d['_ArrayZipType_']=='lzma'):
newobj=lzma.decompress(bytes(newobj),lzma.FORMAT_ALONE)
newobj=np.fromstring(newobj,dtype=np.dtype(d['_ArrayType_'])).reshape(d['_ArrayZipSize_']);
if('_ArrayIsComplex_' in d and newobj.shape[0]==2):
newobj=newobj[0]+1j*newobj[1];
newobj=newobj.reshape(list(d['_ArraySize_']));
return newobj;
elif('_ArrayData_' in d):
if(isinstance(d['_ArrayData_'],str)):
newobj=np.frombuffer(d['_ArrayData_'],dtype=np.dtype(d['_ArrayType_']));
else:
newobj=np.asarray(d['_ArrayData_'],dtype=np.dtype(d['_ArrayType_']));
if('_ArrayZipSize_' in d and newobj.shape[0]==1):
if(isinstance(d['_ArrayZipSize_'],str)):
d['_ArrayZipSize_']=np.array(bytearray(d['_ArrayZipSize_']));
newobj=newobj.reshape(d['_ArrayZipSize_']);
if('_ArrayIsComplex_' in d and newobj.shape[0]==2):
newobj=newobj[0]+1j*newobj[1];
newobj=newobj.reshape(d['_ArraySize_']);
return newobj;
else:
raise Exception('JData', 'one and only one of _ArrayData_ or _ArrayZipData_ is required')
return decodedict(d,opt);
else:
return copy.deepcopy(d);
##====================================================================================
## helper functions
##====================================================================================
def jsonfilter(obj):
if type(obj) == 'long':
return str(obj)
elif type(obj).__module__ == np.__name__:
if isinstance(obj, np.ndarray):
return obj.tolist()
else:
return obj.item()
elif isinstance(obj, float):
if(np.isnan(obj)):
return '_NaN_';
elif(np.isinf(obj)):
return '_Inf_' if (obj>0) else '-_Inf_';
def encodedict(d0, opt={}):
d=dict(d0);
for k, v in d0.items():
newkey=encode(k,opt)
d[newkey]=encode(v,opt);
if(k!=newkey):
d.pop(k)
return d;
def encodelist(d0, opt={}):
d=copy.deepcopy(d0)
for i, s in enumerate(d):
d[i] = encode(s,opt);
return d;
def decodedict(d0, opt={}):
d=dict(d0);
for k, v in d.items():
newkey=encode(k,opt)
d[newkey]=decode(v,opt);
if(k!=newkey):
d.pop(k)
return d;
def decodelist(d0, opt={}):
d=copy.deepcopy(d0)
for i, s in enumerate(d):
d[i] = decode(s,opt);
return d;
|
