解包参数时修改列表

Question

I've recently come accross a problem where I needed to find an elegant way to itterate through a list while having readable code. I discovered the 'unpacking'.

Assume I have a list : l = [ ['param_1', 1] ['param_2',2] ]

I used to iterate through this list as follows :

for param in l:
  if param[0] == 'param_1':
    param[1] = 27

Now I am doing :

for param_name, param_value in l:
  if param_name == 'param_1':
    param_value = 27

Though it is much more readable, the fact that I assign param_value to 27 doesn't change my list as it used to with the old method. Any idea why ? Couldn't find some answer on the internet.

Answer 1

This is what dictionaries are for; you're trying to update the value associated with the string param_1 .

l = [ ['param_1', 1], ['param_2',2] ]
d = dict(l)
# {'param_1': 1, 'param_2': 2}

Instead of iterating over the whole list, you can now set the value directly:

d['param_1'] = 27
# {'param_1': 27, 'param_2': 2}

Note that this will only work properly if your parameter names are all unique; identical names will all get collapsed into the same dictionary key.

Answer 2

You said "Couldn't find some answer on the internet" , the answer is in between the lines, indeed. In the official documentation, though, the Data Model has all the basic stones to understand what's going on. This is related to objects' mutability and variables name binding .

In the first case,

for param in l:
    ...

, param is a list. Lists are mutable objects, containers which values can change.

When you change one of param 's values the way you do,

    param[1] = 27

, you're telling Python to modify the second value of the list param is pointing to.

On the other hand, in the second case,

for param_name, param_value in l:
    ...

, you're unpacking those lists inside l into the corresponding pair of (scalar) values, param_name and param_value . The values they are pointing to are immutable types (string, number), and you also "lost" the pointer to the containet/list they were in (ie, param in the first case).

Now, when you do param_value = 27 , Python is actually creating a new space in memory to allocate 27 and param_value is now referring to that value, but this is a different variable -- same name param_value but different space in memory.

---

References:

• https://docs.python.org/3/reference/datamodel.html
• https://docs.python.org/3/reference/executionmodel.html