A couple of questions:
- why doesn't the new column have a foreign key referring to the bug table?
- what's the test for?
Also, if I understand this correctly, this patch denormalizes the database -- information about whether the bug is a dupe is already stored somewhere. I guess this change to the model allows you to support crazy things like the example described above, but why is this desirable?
If we do keep this patch, then I think (but am not sure) that we need a uniqueness constraint such that any given bugaffecstperson.bug only has one master_affects.
Hey Karl,
A couple of questions:
- why doesn't the new column have a foreign key referring to the bug table?
- what's the test for?
Also, if I understand this correctly, this patch denormalizes the database -- information about whether the bug is a dupe is already stored somewhere. I guess this change to the model allows you to support crazy things like the example described above, but why is this desirable?
If we do keep this patch, then I think (but am not sure) that we need a uniqueness constraint such that any given bugaffecstperso n.bug only has one master_affects.
Thanks,
jml