Thread: Oracle Porting, Compiere
For those interested in porting from Oracle to PG, refer http://www.compiere.org/technology/pg/index.html Compiere is an Open Source accounting system designed around Oracle. It uses Oracle features extensively. The desire to port this to PG shows some of the present limitations in PG. I seem to recall that someone has a patch for hierarchal Queries. For your interest, john
Yes, we do. It is in my mailbox and we are hoping someone will make it standards-compliant via SQL99 recursive queries. --------------------------------------------------------------------------- John Huttley wrote: > For those interested in porting from Oracle to PG, refer > > http://www.compiere.org/technology/pg/index.html > > Compiere is an Open Source accounting system designed around Oracle. > It uses Oracle features extensively. The desire to port this to > PG shows some of the present limitations in PG. > > I seem to recall that someone has a patch for hierarchal Queries. > > For your interest, > > john > > > ---------------------------(end of broadcast)--------------------------- > TIP 6: Have you searched our list archives? > > http://archives.postgresql.org > -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 359-1001+ If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square, Pennsylvania19073
Bruce Momjian <pgman@candle.pha.pa.us> writes:
> Yes, we do. It is in my mailbox and we are hoping someone will make it
> standards-compliant via SQL99 recursive queries.
Andrew Overholt of Red Hat is in fact looking at SQL99 recursive
queries. I will not promise that he'll have it done in time for 7.4,
but something should emerge by and by.
regards, tom lane
John Huttley <john@mwk.co.nz> writes: > For those interested in porting from Oracle to PG, refer > http://www.compiere.org/technology/pg/index.html I can't see any indication of the date this was written --- but I get the feeling it dates back a bit. At least some of the issues mentioned have been solved in recent PG releases. regards, tom lane
This isn't more that a month or two old. The previous version worried about nested transactions. I know Bruce has been thinking about that for a while now, so possibly that was crossed off as 'fixed'. The 7.4 Docs cover the unlimited precision "NUMERIC" type, so perhaps that also should be crossed off. The use of 'out' parameters sounds hairy though. Regards john On Tue, 2003-04-01 at 15:48, Tom Lane wrote: > John Huttley <john@mwk.co.nz> writes: > > For those interested in porting from Oracle to PG, refer > > http://www.compiere.org/technology/pg/index.html > > I can't see any indication of the date this was written --- but I get > the feeling it dates back a bit. At least some of the issues mentioned > have been solved in recent PG releases. > > regards, tom lane > > > ---------------------------(end of broadcast)--------------------------- > TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql.org >
John Huttley <john@mwk.co.nz> writes:
> This isn't more that a month or two old.
Okay, it wasn't real clear...
> The previous version worried about nested transactions.
It still should. I see no prospect that nested xacts will be in 7.4,
and I wouldn't care to bet they'll be in the release or two after ...
On a more positive note:
* NUMERIC does indeed do the unspecified-precision thing.
* If you can't be bothered to rename datatypes, CREATE DOMAIN would serve to provide Oracle-compatible names for
Postgresdatatypes. There is some parse-time overhead associated though.
* CREATE OR REPLACE does exist for some object types now --- views and functions for sure, don't recall anything else.
* Postgres *does* do sub-SELECT-in-FROM, which is what I think is meant by the references to embedded views.
* On the outer join syntax issue: PG supports the SQL-standard syntax, which I believe is also supported by recent
Oraclereleases. Consider migrating away from Oracle's proprietary syntax.
regards, tom lane
Tom Lane wrote: > John Huttley <john@mwk.co.nz> writes: > > This isn't more that a month or two old. > > Okay, it wasn't real clear... > > > The previous version worried about nested transactions. > > It still should. I see no prospect that nested xacts will be in 7.4, > and I wouldn't care to bet they'll be in the release or two after ... Actually, Alvaro Herrera is working on it as a college project. I don't know the completion date, but I am sure it will be in 7.5, if not in 7.4. I am CC'ing him for a comment. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 359-1001+ If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square, Pennsylvania19073
On Tue, 1 Apr 2003 15:25:52 -0400, Alvaro Herrera <alvherre@dcc.uchile.cl> wrote: > pg_clog and >pg_subtrans stuff. Manfred Koizar said he wanted to work on that, but I >dunno if he is. Yes, he is. Although there's some discussion whether my proposal is the best way to do it ... ServusManfred
On Tue, Apr 01, 2003 at 12:33:52AM -0500, Bruce Momjian wrote: > Tom Lane wrote: > > It still should. I see no prospect that nested xacts will be in 7.4, > > and I wouldn't care to bet they'll be in the release or two after ... > > Actually, Alvaro Herrera is working on it as a college project. I don't > know the completion date, but I am sure it will be in 7.5, if not in > 7.4. I am CC'ing him for a comment. Yes, there's some progress in my private tree. As soon as I can make it bootstrap I'll post a patch for comment. I expect to have it for 7.4, althought I'm not sure if I will be able to do the pg_clog and pg_subtrans stuff. Manfred Koizar said he wanted to work on that, but I dunno if he is. I am CC'ing him for a comment :-) -- Alvaro Herrera (<alvherre[a]dcc.uchile.cl>) "No hay ausente sin culpa ni presente sin disculpa" (Prov. frances)