Thread: Re: [SQL] Uniqueness of rule, constraint, and trigger names
"Dan Langille" <dan@langille.org> writes:
> On 4 Mar 2002 at 14:24, Tom Lane wrote:
>> ... but I'm a little worried about the possibility of errors
>> in loading schemas from existing databases, where there might be
>> non-unique constraint names.
> Create a tool to generate unique constraint names during a dump.
And then all we need is a time machine, so we can make existing
instances of pg_dump contain the tool? It's not that easy ...
I am not sure that there's really a problem here, because I don't
think duplicate constraint names will be generated during plain
CREATE operations. However, an ALTER TABLE might leave you with
a problem. Hard to tell if this is critical enough to worry about.
regards, tom lane
If pgupgrade can be fixed to work, perhaps it could set off warnings on items that need to be corrected in a 'schema upgradability test' which will ensure that the user can upgrade it properly -- it shouldn't upgrade if it can't guarentee an upgrade will succeed. This should include full schema test (that whole bad schema entry stuff that pg_dump is supposed to work around) too. Something I'm thinking about digging into. -- Rod Taylor This message represents the official view of the voices in my head ----- Original Message ----- From: "Tom Lane" <tgl@sss.pgh.pa.us> To: <dan@langille.org> Cc: <pgsql-hackers@postgreSQL.org>; <pgsql-sql@postgreSQL.org> Sent: Monday, March 04, 2002 3:38 PM Subject: Re: [HACKERS] [SQL] Uniqueness of rule, constraint, and trigger names > "Dan Langille" <dan@langille.org> writes: > > On 4 Mar 2002 at 14:24, Tom Lane wrote: > >> ... but I'm a little worried about the possibility of errors > >> in loading schemas from existing databases, where there might be > >> non-unique constraint names. > > > Create a tool to generate unique constraint names during a dump. > > And then all we need is a time machine, so we can make existing > instances of pg_dump contain the tool? It's not that easy ... > > I am not sure that there's really a problem here, because I don't > think duplicate constraint names will be generated during plain > CREATE operations. However, an ALTER TABLE might leave you with > a problem. Hard to tell if this is critical enough to worry about. > > regards, tom lane > > ---------------------------(end of broadcast)--------------------------- > TIP 3: if posting/reading through Usenet, please send an appropriate > subscribe-nomail command to majordomo@postgresql.org so that your > message can get through to the mailing list cleanly >
On 4 Mar 2002 at 15:38, Tom Lane wrote: > "Dan Langille" <dan@langille.org> writes: > > On 4 Mar 2002 at 14:24, Tom Lane wrote: > >> ... but I'm a little worried about the possibility of errors > >> in loading schemas from existing databases, where there might be > >> non-unique constraint names. > > > Create a tool to generate unique constraint names during a dump. > > And then all we need is a time machine, so we can make existing > instances of pg_dump contain the tool? It's not that easy ... *sigh* You don't have to modify all previous versions of pg_dump. They use this tool if they want to upgrade to this version. There is no perfect solution. Even a script to check for uniqueness might help. I'm just trying to help the best way I can; by providing suggestions. As requested. > I am not sure that there's really a problem here, because I don't > think duplicate constraint names will be generated during plain > CREATE operations. However, an ALTER TABLE might leave you with > a problem. Hard to tell if this is critical enough to worry about. It would be easily detected at load time. Then you direct the user to the tool mentioned above. -- Dan Langille The FreeBSD Diary - http://freebsddiary.org/ - practical examples