robertmassaioli(at)gmail(dot)com writes:
> Now, the issue is that when you do \d and view the table that has been
> created the CHECK block is missing parens that show that it really will
> behave as an XOR. I think this is a bug that ould be great to fix. Please
> let me know if you need more information. Cheers!
It's not a bug that we print check constraints without unnecessary parens,
any more than it is that we don't reproduce whitespace exactly.
Those parens are unnecessary, because per the SQL standard AND binds more
tightly than OR.
regards, tom lane