Why does removing items while looping over a list skip some values?
Quick answer
Removing an item changes the list’s positions while the loop is moving through them. After a removal, the next item can shift into a position the loop has already passed, so it is skipped. For filtering, build a new list with a comprehension, such as [x for x in items if keep(x)]. If you must change the original list, choose an approach based on that requirement rather than assuming one method fits every case. [S1]
The question, the details, the next step.
I’m looping over a Python list and removing values that match a condition, but some matching values remain. Why?
The loop is progressing through positions in the list. Removing an item shifts later items toward the start; the loop then advances, which can take it past the item that just shifted into the removed item’s position. The Python tutorial cautions that changing a list while looping can be tempting, but creating a new list is often simpler and safer. [S1]
For example, suppose you want to remove every even number:
numbers = [2, 4, 5, 6]
for number in numbers:
if number % 2 == 0:
numbers.remove(number)
print(numbers)After 2 is removed, 4 shifts into its position. The loop advances to the next position, so it sees 5 rather than checking 4. The result is [4, 5], even though 4 also matches the condition.
What should I write instead if I just want a list with the even numbers removed?
Create a filtered list with a comprehension. It checks each value and includes it only when the condition is true. To keep the odd numbers:
numbers = [2, 4, 5, 6]
remaining = [number for number in numbers if number % 2 != 0]
print(remaining) # [5]This makes the filtering decision without removing elements from the list being traversed. The original numbers list remains [2, 4, 5, 6]; remaining is the new filtered list. The tutorial illustrates filtering into a separate list and describes a comprehension as an expression of that pattern. [S1]
Can I replace numbers with the filtered result so later code uses the shorter list?
Yes. If you do not need to preserve the original values under the name numbers, assign the new list to that name:
numbers = [2, 4, 5, 6]
numbers = [number for number in numbers if number % 2 != 0]
print(numbers) # [5]The right-hand side is built from the current list, then the name refers to the filtered result. This is still a new-list filtering approach; it does not remove values one by one during iteration. If other parts of your program depend on changing the existing list object itself, first establish that requirement. The tutorial does not prescribe one universal in-place strategy; it says the right approach depends on the desired mutation. [S1]
What if I really do need to change the original list rather than make a replacement?
Treat that as a separate requirement from filtering. Ask whether other parts of your program need to observe the same list object changing, or whether it is enough for the variable to refer to a filtered list. If a replacement is acceptable, the comprehension is the straightforward option. If the original object must be mutated, choose a method designed for that situation and check its behavior against your condition; there is no single in-place strategy that fits every case. [S1]
Before settling on an approach, try a small example that includes matching values next to each other. Adjacent matches are useful for revealing skipped values, as in [2, 4, 5, 6]. Compare the result with the intended output, not just with an example where matching values are separated.
A practical checklist
Write down which values should remain after filtering.
Avoid removing matching values from the same list you are traversing.
Use
[value for value in items if condition]to build a filtered list.Decide whether replacing the variable is enough or whether the original list object must change.
Test adjacent matching values so a skipped item is easy to spot.
A few more useful answers
Does a list comprehension change the original list?
No. It creates a new list from the values that meet the condition. You can assign that result to the same variable name if you want later code to use the filtered list.
Why are adjacent matches a useful test?
When an item is removed during traversal, a neighboring item can shift into a position the loop has already passed. Adjacent matches make that skipped-value problem easier to notice. [S1]
Is there one best way to mutate a list in place?
Not for every situation. The suitable approach depends on whether you need the existing list object to change and on the filtering task. The Python tutorial does not prescribe one universal in-place strategy. [S1]
Sources
References used for this guide.
- [S1] Python Tutorial: Data Structures docs.python.org · Checked