python variable contents changed by function when

问题:

Here is a sample of code where a function is run repeatedly with new information for most of the input variables except one, good_ens. The input variable good_ens that should never be changed, gets changed. What is going on here? This defies my understanding of scope.

def doFile(infileName, outfileName, goodens, timetype, flen):

    print('infilename = %s' % infileName)
    print('outfilename = %s' % outfileName)
    print('goodens at input are from %d to %d' % (goodens[0],goodens[1]))
    print('timetype is %s' % timetype)

    maxens = flen # fake file length
    print('%s time variable has %d ensembles' % (infileName,maxens))

    # TODO - goodens[1] has the file size from the previous file run when multiple files are processed!
    if goodens[1] < 0:
        goodens[1] = maxens

    print('goodens adjusted for input file length are from %d to %d' % (goodens[0],goodens[1]))

    nens = goodens[1]-goodens[0]
    print('creating new netCDF file %s with %d records (should match input file)' % (outfileName, nens))


# user settings
datapath = ""

datafiles = ['file0.nc',\
             'file1.nc',\
             'file2.nc',\
             'file3.nc']
# fake file lengths for this demonstration
datalengths = [357056, 357086, 357060, 199866]
outfileroot = 'outfile'
attFile = datapath + 'attfile.txt'
# this gets changed!  It should never be changed!
# ask for all ensembles in the file
good_ens = [0,-1]

# --------------  beyond here the user should not need to change things
for filenum in range(len(datafiles)):

    print('\n--------------\n')
    print('Input Parameters before function call')
    print(good_ens)
    inputFile = datapath + datafiles[filenum]
    print(inputFile)
    l = datalengths[filenum]
    print(l)
    outputFile = datapath + ('%s%03d.cdf' % (outfileroot,filenum))
    print(outputFile)

    print('Converting from %s to %s' % (inputFile,outputFile))
    # the variable good_ens gets changed by this calling function, and should not be
    doFile(inputFile, outputFile, good_ens, 'CF', l)
    # this works, but will not work for me in using this function
    #doNortekRawFile(inputFile, outputFile, [0,-1], 'CF', l)

Output for the first two iterations of the for loop is below. Note good_ens gets changed from [0, -1] to the value of goodens that is inside the function. Why? Never mind the difference in variable names, they don't even share the same scope.

--------------

Input Parameters before function call
[0, -1]
file0.nc
357056
outfile000.cdf
Converting from file0.nc to outfile000.cdf
infilename = file0.nc
outfilename = outfile000.cdf
goodens at input are from 0 to -1
timetype is CF
file0.nc time variable has 357056 ensembles
goodens adjusted for input file length are from 0 to 357056
creating new netCDF file outfile000.cdf with 357056 records (should match input file)

--------------

Input Parameters before function call
[0, 357056]
file1.nc
357086
outfile001.cdf
Converting from file1.nc to outfile001.cdf
infilename = file1.nc
outfilename = outfile001.cdf
goodens at input are from 0 to 357056
timetype is CF
file1.nc time variable has 357086 ensembles
goodens adjusted for input file length are from 0 to 357056
creating new netCDF file outfile001.cdf with 357056 records (should match input file)

--------------

There is a similar question here:
Python issue value of property changes when falling out of loop scope

However I do not want to embed the variable good_ens down in a for loop. I want its value to be set by the user once at the head of a script, then used in the for loop.

回答1:

When you call doFile try this instead:

doFile(inputFile, outputFile, list(good_ens), 'CF', l)

I think of it this way: A list is a thing which points to the value of each element within the list. When you pass a list into a function, the thing that does the pointing gets copied, but the values of the things pointed to do not get copied.

doing list(good_ens) actually makes copies in memory of the elements of the list, and will keep the original values from getting changed. See below:

>>> def change_list(the_list):
...     the_list[0] = 77
...     return
...
>>> a=[0,1,2,3,4]
>>> change_list(a)
>>> a
[77, 1, 2, 3, 4]
>>>
>>> a=[0,1,2,3,4]
>>> change_list(list(a))
>>> a
[0, 1, 2, 3, 4]

EDIT: The reasoning for this is that, as the other answers have indicated, list is a mutable data type in python. Mutable data types can be changed, whereas immutable data types cannot be changed but rather return new objects when attempting to update.

回答2:

In python, lists are mutable.

>>> a = [1,2,3]
>>> def fun(d):
...  d.append(55)
...
>>> a
[1, 2, 3]
>>> fun(a)
>>> a
[1, 2, 3, 55]
>>> fun(a[:])
>>> a
[1, 2, 3, 55]
>>> fun(a)
>>> a
[1, 2, 3, 55, 55]

If you want them to be immutable consider using a tuple.

回答3:

The other answers cover the idea that lists are mutable. Below is a possible refactoring that gets around this issue in what I think is a sensible way.

def doFile(infileName, outfileName, goodens, timetype, flen):
    # ...
    min_ens, max_ens = goodens

    if max_ens < 0:
        max_ens = flen

    nens = max_ens - min_ens
    # ...

This way, you can still call the function as you have been, your variables within the function are named more aptly, and you never mutate the provided list object.