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
{{ message }}
Repository navigation
Could a 0.3.2 carry the zero-length magnet fix (#435)? #469
We use pyAML with lattices whose correctors are zero-length elements. On 0.3.1 setting their strength
divides by zero (abstract_impl.py, the poly = s / (length * sign) write), which #360 reported and #435 fixed on main on 2026-09-24.
Since 0.3.1 was tagged on 2026-09-16, the fix is not on PyPI. Would you consider a 0.3.2 with #435, or
is the next release?
@thellert Yes, we can make a release with some of the recent bug fixes. I'll put it up for discussion at the maintainer meeting if there's no release until then. There's also a bug on serialized magnets that is awaiting a fix. I think a release after it is necessary.
Activity
gubaidulinvadim commentedon Oct 12, 2026
@thellert Yes, we can make a release with some of the recent bug fixes. I'll put it up for discussion at the maintainer meeting if there's no release until then. There's also a bug on serialized magnets that is awaiting a fix. I think a release after it is necessary.