如何生成两个值为lis的字典键

Question

Write1 [1.0]
Write1 [12.0]
Write1 ['C:/Users/san/Modeler/']
Write1 ['png']
Write1 ['rgb']
Write1 [True]
Write2 [1.0]
Write2 [1.0]
Write2 ['']
Write2 [' ']
Write2 ['rgb']
Write2 [False]

我想这样措辞 {'Write1'：[1.0,12.0，'c:/path'，'png'，'rgb'，True]，'Write2'：[1.0,1.0，'，''，''，'，'rgb'，False]} 列表中的值来自核弹.toNode（wNode）[eachAttrib].value（） 这就是我想做的

attributes=[]
wnodeData={}
for wNode in writeNodes:
    for eachAttrib in ['first','last','file','file_type','channels','use_limit']:
        wnodeData[wNode]=attributes
        attributes.append(nuke.toNode(wNode)[eachAttrib].value())

Answer 1

比如：

text="""Write1 [1.0]
Write1 [12.0]
Write1 ['C:/Users/san/Modeler/']
Write1 ['png']
Write1 ['rgb']
Write1 [True]
Write2 [1.0]
Write2 [1.0]
Write2 ['']
Write2 [' ']
Write2 ['rgb']
Write2 [False]"""

from ast import literal_eval

d = {}
lines = text.splitlines()
for line in lines:
    k, v = line.split(' ', 1)
    val = literal_eval(v)[0]
    d.setdefault(k, []).append(val)

print d

# {'Write1': [1.0, 12.0, 'C:/Users/san/Modeler/', 'png', 'rgb', True], 'Write2': [1.0, 1.0, '', ' ', 'rgb', False]}

Answer 2

import collections

s = '''Write1 [1.0]
Write1 [12.0]
Write1 ['C:/Users/san/Modeler/']
Write1 ['png']
Write1 ['rgb']
Write1 [True]
Write2 [1.0]
Write2 [1.0]
Write2 ['']
Write2 [' ']
Write2 ['rgb']
Write2 [False]'''

d = collections.defaultdict(list)
for line in s.split('\n'):
    print(line)
    key, val = line.split(' ', 1)
    d[key].append(val[1:-1])
print(d)

显然，如果你有一个文本文件而不是一个字符串，你应该用以下内容替换for循环：

with open('mystuff.txt', 'r') as f:
    for line in f:

如果您想实际计算方括号内的值（因此您得到的是1.0的浮点值而不是字符串'1.0'，字符串'png'而不是字符串"'png'"，等等），您需要知道它们所用的语言并对其进行解析。看起来它们是Python的一个子集，每个元素要么是没有转义符的单引号字符串文本，要么是非指数格式的浮点文本，要么是布尔文本，但我不会假设在不知道值来自何处的情况下（当然我也不会在不知道值来自何处的情况下调用eval，因为这样有人就可以在文本文件中插入os.system('rm -rf /')，然后你就可以对其进行评估。你知道吗

Answer 3

write1 = {'write1':[1,2,3,4] }

应该有用。。。你知道吗