Re: [BUGS] BUG #4203: perform dblink() in begin/exception returns wrong SQLSTATE code
Tom Lane <[email protected]>
| Newsgroups | gmane.comp.db.postgresql.devel.patches |
|---|---|
| Message-ID | <[email protected]> |
Joe Conway <[email protected]> writes: > [ improved patch ] Still a couple quibbles: > + ereport(level, > + (errcode(sqlstate), > + errmsg(message_primary), This *must* be errmsg("%s", message_primary), else you have big problems with % in the text. Also, I think it's at least theoretically possible for message_primary to be null, in which case you'd better substitute "unknown error" or some such. You could avoid the ugly cast-away-const by making dblink_context_conname be const char *, no? Since dblink_res_error isn't going to return if fail = true, seems like you could skip the "if (!fail)" tests occurring after calls to it. regards, tom lane -- Sent via pgsql-patches mailing list ([email protected]) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-patches