从模板字典和键生成字典列表

Question

I'm trying to automate a series of tests, and I need to have a loop where I change the parameters.

mydictionary={'a':10,'b':100,'c':30}

def swapRules(d,rule):
   "clear dict, set to 100 the rule that match the string"
   print d, rule
   if not d.has_key(rule): raise Exception("wrong string")
   d=resetDict(d)
   d[rule]=100
   return d

def resetDict(d):
   '''clear the dict '''
   for i in d.keys():
       d[i]=0
   return d

def tests(d):
   from itertools import starmap, repeat, izip
   keys=d.keys()
   paramsDictionaries=list(starmap(swapRules, izip(repeat(d),keys)))
   print(paramsDictionaries)

I'cannot understand why when I run test(mydictionary) the output always contains the same value. Seems that the issue is not in a wrong use of the itertools: as the REPL shows by substituting with a simple list comprehension:

In [9]: keys=mydictionary.keys()
In [10]: [tr.swapRules(mydictionary,jj) for jj in keys]
{'a': 0, 'c': 0, 'b': 100} a
{'a': 100, 'c': 0, 'b': 0} c
{'a': 0, 'c': 100, 'b': 0} b
Out[10]:
[{'a': 0, 'b': 100, 'c': 0},
 {'a': 0, 'b': 100, 'c': 0},
 {'a': 0, 'b': 100, 'c': 0}]

I'm really puzzled since when the swapRules function is evoked alone, produces the expected result, as shown by the print statements... any idea on what I'm doing wrong? is it by any chance caching something?

Answer 1

Answering my own question since I've found that this work:

def swapRules2(d,rule):
    '''clear dict, return new with only rule set'''
    keys=d.keys()
    if rule not in keys: raise Exception("wrong key")
    outd=dict.fromkeys(keys,0)
    outd[rule]=100
    return outd

and in the list comprehension it returns the expected output:

 In [16]: [tr.swapRules2(mydict,jj) for jj in keys]
Out[16]:
[{'a': 100, 'b': 0, 'c': 0},
 {'a': 0, 'b': 0, 'c': 100},
 {'a': 0, 'b': 100, 'c': 0}]

However, I still do not understand why the previous approach did not.