Python在使用zip（）遍历列表时不会修改值，但是在使用enumerate（）时会修改值

Question

Why Python works differently when iterating using index numbers? For example with the zip function (that don't has indexes when used in the for loop) a value isn't stored in a variable. While with the enumerate() function (that allow me to have an "index" in the foor loop) it happens normally.

A solution that don't use index has different values for my "width" variable when I'm trying to print it inside and outside the loop, here the zip() function was used to work with the two lists at the same time without needing to deal with their indexes:

tableData = [['apples', 'oranges', 'cherries', 'banana'],
            ['Alice', 'Bob', 'Carol', 'David'],
            ['dogs', 'cats', 'moose', 'goose']]

def printTable(myListOfLists):
    colWidths = [0] * len(myListOfLists)
    for lists, width in zip(myListOfLists, colWidths):
        for item in lists:
            if width < len(item):
                width = len(item)
        print(width) #This prints my 'width' rightly
    print(colWidths) #This prints my 'width' wrongly

printTable(tableData)

The above code produces the following not desired output:

8
5
5
[0, 0, 0]

But if I use enumerate() (generating indexes to work with the two lists at the same time) the values modified inside my for loop will be properly saved in the width variable, see the code example:

tableData = [['apples', 'oranges', 'cherries', 'banana'],
            ['Alice', 'Bob', 'Carol', 'David'],
            ['dogs', 'cats', 'moose', 'goose']]

def printTable(myListOfLists):
    colWidths = [0] * len(myListOfLists)
    for index, lists in enumerate(myListOfLists):
        for item in lists:
            if colWidths[index] < len(item):
                colWidths[index] = len(item)
        print(colWidths[index]) #This prints my 'width' rightly
    print(colWidths) #This ALSO prints my 'width' rightly

printTable(tableData)

This second output is:

8
5
5
[8, 5, 5]

Then, please, how can I work with the zip() and have my "external" variable modified? What am I doing wrong? Thanks.

Answer 1

I believe, what you want to achieve is to get length of longest item in given sub-list from tableData . To do that you can get longest string from sub-list and then measure its length like this:

tableData = [['apples', 'oranges', 'cherries', 'banana'],
             ['Alice', 'Bob', 'Carol', 'David'],
             ['dogs', 'cats', 'moose', 'goose']]


def printTable(myListOfLists):
    colWidths = []
    for sub_list in myListOfLists:
        longest_item = max(sub_list, key=len)
        colWidths.append(len(longest_item))
    print(colWidths)


printTable(tableData)

edit with explanation:

By using for lists, width in zip(myListOfLists, colWidths):

on each cycle you're getting 1) single list with strings from tableData as lists . 2) integer, exactly 0 from colWidths as width

then, on iteration over each text like 'apples', 'oranges' etc.

what you're doing in here: width = len(item) is like saying "now width isn't 0 anymore, it's len(item) , by this you're not affecting list colWidths because you're not working with it, you're manipulating this 0 value, it's not related to place in colWidths list, it's just 0 which is replaced by len(item) under "width" name.

Answer 2

by this way you wont change the value of the list you need the index to change the real value.

example:

for e in [1, 2, 3]:
    e = 10  # this way is wrong 

data = [1, 2, 3]
for index, e in enumerate(data):
    e[index] = 50 # always it is needed an index  

a good way to solve your problem:

tableData = [['apples', 'oranges', 'cherries', 'banana'],
             ['Alice', 'Bob', 'Carol', 'David'],
             ['dogs', 'cats', 'moose', 'goose']]


data = [len(max(e, key=len)) for e in tableData]

print(data)

or

tableData = [['apples', 'oranges', 'cherries', 'banana'],
            ['Alice', 'Bob', 'Carol', 'David'],
            ['dogs', 'cats', 'moose', 'goose']]

def printTable(myListOfLists):
    colWidths = [0] * len(myListOfLists)
    for index, row in enumerate(myListOfLists):
        colWidths[index] = len(max(row, key=len))
    print(colWidths)

printTable(tableData)

on another hand you could try out.

for i, (a, b) in enumerate(zip(alist, blist)):
    print i, a, b

Answer 3

enumerate is a Pythonic way of generating indices (similar to, but more idiomatic, for i in range(len(...):

colWidths = [0] * len(myListOfLists)
for index, lists in enumerate(myListOfLists):
    for item in lists:
        if colWidths[index] < len(item):
            colWidths[index] = len(item)

Here lists and item come from myListOfLists , but you explicitly modifying colWidths list.

In the first case, you are modifying an iteration variable, but not the source list:

colWidths = [0] * len(myListOfLists)
for lists, width in zip(myListOfLists, colWidths):
    for item in lists:
        if width < len(item):
            width = len(item)

width is an iteration variable. But the width = len(item) line assigns a new value to the variable (name), breaking its link with the source array. On the next iteration, width is again set by the iterator. The source list, colWidths does not change.

This pattern is true for simple cases:

alist = [1,2,3,4]
for i in alist:
   i = 5     # does nothing to alist

for i in range(4):
    alist[i] = 5    # changes an element of alist

Changes to the iteration variable are tricky. You have to understand when it represents something 'mutable', and not. And understand the difference between assignment to a variable, and modifying an object (like a list).