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 believe these issues could be resolved by appending the policies by a default alphanumerical sorting (there should also be a possibility to reverse this). It should, if possible, be language-dependent as collation order depends on the language of the literal. E.g., 'ä' is sorted before 'b' in English but not in Finnish.
The text was updated successfully, but these errors were encountered:
The preferred value policy selectors are not deterministic. That's bad - they should be.
You propose a solution which relies on locale-aware alphanumerical sorting. While that's one possible way to solve the issue, I think there are simpler ways that could be used instead. I'll comment on the PR next...
I noticed (due to NatLibFi/Finto-data@26a2d87#diff-b7776f22f1355c1d1fbc86095078db1711cc8e6dea017ad6524fbf3d595617ac) that the preferred value policy selectors are not, by default, deterministic.
I believe these issues could be resolved by appending the policies by a default alphanumerical sorting (there should also be a possibility to reverse this). It should, if possible, be language-dependent as collation order depends on the language of the literal. E.g., 'ä' is sorted before 'b' in English but not in Finnish.
The text was updated successfully, but these errors were encountered: