> if (needlabel) > appendStringInfo(buf, "::%s", > - > format_type_with_typemod(node->consttype, > - > node->consttypmod)); > + > format_type_be_qualified(node->consttype)); > Pondering more about this one, I think that we are going to need a new > routine in format_type.c to be able to call format_type_internal as > format_type_internal(type_oid, typemod, true/false, false, true). If > typemod is -1, then typemod_given (the third argument) is false, otherwise > typemod_given is true. That's close to what the C function format_type at > the SQL level can do except that we want it to be qualified. Regression > tests will need an update as well.
I don't know what's going on here, but please look at the patch posted by Alexander Shulgin in the thread about JSON DDL deparse today; there's some additional stuff in format_type.c there that is probably useful to share between these two patches.
Oh, OK. That's good to know. I'll have a look at it. I think that it may be possible to extract a single patch usable for both facilities. --
