Python RuntimeError: dictionary changed size during iteration

Fix RuntimeError when mutating a dict or set while iterating it, and choose between iterating a copy, over list(d), or deferred deletion.

beginnerdebuggingpythoniteration

Symptoms

RuntimeError: dictionary changed size during iteration
RuntimeError: Set changed size during iteration
The error appears only when a filter branch deletes a key

Possible causes

Code deletes or adds keys inside a for key in my_dict loop
A callback invoked during iteration mutates the same mapping
set.discard / set.add called while iterating the set
A defaultdict grows lazily when a missing key is read during iteration

Diagnosis steps

Locate the loop and the mutation

Interpretation

Look for del, pop, assignment, or update inside a for loop over the same object.

Reproduce minimally

python -c "d={1:1,2:2}; [d.pop(k) for k in d]"

Interpretation

Confirming the pattern reproduces the error shows the mutation is the cause.

Check for implicit growth

Interpretation

Reading a missing key from a defaultdict inserts it, which also changes size mid-loop.

Root cause

Python dict and set iterators are live views. Changing the container size during iteration invalidates the iterator and raises RuntimeError to prevent undefined traversal. Even an implicit insert from a defaultdict or a callback can trigger it.

Fix

Iterate over a snapshot: `for k in list(d):` or `for k, v in d.copy().items():`. Or collect keys to delete and remove them after the loop with a dict comprehension. Use `d.pop(k, None)` outside the loop, and avoid defaultdict reads inside iteration.

Verify the fix

Re-run the routine; no RuntimeError occurs and the container has the expected final contents, verified by an assertion on its size and keys.

Symptoms

A loop over a dict or set raises RuntimeError: dictionary changed size during iteration or its Set variant, usually only on the branch that mutates the container.

Diagnosis

Find the mutation inside the loop:

for key in my_dict:
    if not wanted(key):
        del my_dict[key]   # raises RuntimeError

Minimal reproductions with pop confirm the pattern, and reads from a defaultdict can insert keys implicitly.

Fix

Iterate a snapshot, or defer deletions:

for key in list(my_dict):
    if not wanted(key):
        del my_dict[key]

# or build a new dict
my_dict = {k: v for k, v in my_dict.items() if wanted(k)}

Verification

Re-run and assert the final keys and size. The error disappears and the container reflects the intended deletions.

Patterns That Hide the Mutation

The mutation is not always a literal del. A helper called inside the loop that pops from the container, a defaultdict read that inserts a default, or a generator that yields while another coroutine mutates the same dict all trigger the same RuntimeError. When the loop body is short but the error persists, search the call graph of every function invoked inside the loop for any write to the container, then move those writes outside the iteration or snapshot the keys first.