You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I found a weird bug (or bugs) with shifting - I think I can replicate this first one with croaring alone though so I will raise an issue there to confirm. The failures happen with both the current croaring and the latest release (v2.1.0).
Another case that fails with a segfault, but only when copy_on_write is enabled (transcribed from the test case - hypothesis doesn't deal well with a segfault). The segfault happens in the final comparison.
frompyroaringimportBitMapcls=BitMap# Also fails with FrozenBitMapvalues=range(67335, 68618, 8)
offset=33554432cow=True# Doesn't fail when this is offbm_before=cls(values, copy_on_write=cow)
bm_copy=cls(bm_before)
bm_after=bm_before.shift(offset)
assert(bm_before==bm_copy)
expected=cls([val+offsetforvalinvaluesifval+offsetinrange(0, 2**32)], copy_on_write=cow)
assert(bm_after==expected)
The text was updated successfully, but these errors were encountered:
Example 1
I found a weird bug (or bugs) with shifting - I think I can replicate this first one with croaring alone though so I will raise an issue there to confirm. The failures happen with both the current croaring and the latest release (v2.1.0).
The output should be (64000, 64002, 64004, ...), but you don't even get a set:
Example 2
Another case that fails with a segfault, but only when copy_on_write is enabled (transcribed from the test case - hypothesis doesn't deal well with a segfault). The segfault happens in the final comparison.
The text was updated successfully, but these errors were encountered: