Commit 3513358
committed
Issue 2440: remove the guard around the handling of case 'n' in getargs.c's convertsimple() such that we always treat it as an index type, regardless of whether or not sizeof(size_t) == sizeof(long). Fix the test_args2.Signed_TestCase.test_n() such that it tests for adherence to PEP 357 (don't try and coerce objects that don't have nb_index slots but do have nb_int slots (i.e. floats) into indexes 'just because we can'). Three other commits are related to this one: r62269 and r62279, which were changes to PyNumber_Index (among other things) to check for nb_int slots when we lack nb_index slots -- and r62292, which is when I reverted these changes after various people pointed out that the test was in fact wrong, not the code.
1 parent b9e2304 commit 3513358
2 files changed
Lines changed: 3 additions & 6 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
187 | 187 | | |
188 | 188 | | |
189 | 189 | | |
190 | | - | |
191 | | - | |
| 190 | + | |
| 191 | + | |
192 | 192 | | |
193 | 193 | | |
194 | 194 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
663 | 663 | | |
664 | 664 | | |
665 | 665 | | |
666 | | - | |
667 | 666 | | |
668 | 667 | | |
669 | 668 | | |
| |||
672 | 671 | | |
673 | 672 | | |
674 | 673 | | |
675 | | - | |
| 674 | + | |
676 | 675 | | |
677 | 676 | | |
678 | 677 | | |
679 | 678 | | |
680 | 679 | | |
681 | | - | |
682 | | - | |
683 | 680 | | |
684 | 681 | | |
685 | 682 | | |
| |||
0 commit comments