Since how long CO_NESTED does nothing?

2.2

Oh. If CO_NESTED is useless since Python 2.2 which was released 19 years ago, it's maybe time to remove it.

Does this removal will cover disallowing from __future__ import nested_functions (and probably generators, which was from 2.1)? I can convert this patch to that way

Does this removal will cover disallowing from future import nested_functions

I don't see any benefit from removing this import: it doesn't reduce the Python maintenance burden, it just breaks existing applications for free.

Most __future__ imports simply do nothing. Extract of Python/future.c:

        } else if (strcmp(feature, FUTURE_DIVISION) == 0) {
            continue;
        } else if (strcmp(feature, FUTURE_ABSOLUTE_IMPORT) == 0) {
            continue;
        } else if (strcmp(feature, FUTURE_WITH_STATEMENT) == 0) {
            continue;
        } else if (strcmp(feature, FUTURE_PRINT_FUNCTION) == 0) {
            continue;
        } else if (strcmp(feature, FUTURE_UNICODE_LITERALS) == 0) {
            continue;

@isidentical Could you resolve the conflicts?

FWIW, removal of CO_NESTED was discussed in #96811 and rejected, since it is still useful (to detect nested functions) and used in third-party code, and setting it doesn't cause any problems.

FWIW, removal of CO_NESTED was discussed in #96811 and rejected, since it is still useful (to detect nested functions) and used in third-party code, and setting it doesn't cause any problems.

In that case, I think we can close the PR. @isidentical What are your thoughts on this?