db-torque-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Bogdan Vatkov" <bvat...@globaltech-bg.com>
Subject Re: [DB Torque Wiki] Updated: PostgreSQLFAQ
Date Thu, 22 Apr 2004 14:02:18 GMT

I have a working Torque-gen source for PostgreSQL 7.3 - i mean the SERIAL
I will publish the fixes that have to be done.
I was extremly carefull while patching ..I have tried to do as minimal as
possible changes to the source to get
PostgreSQL 7.3 working correctly.
How to say ..SERIAL issue was not a big issue.
The multi schema projects was the hardest one to fix.

I have already fixed the most of issues to have it all working except one

When a multiple schema projects is launched then
the shema.sql is generated in the way (for each file)

CREATE TABLE statements...
ALTER TABLE ...constraint add statements...

and when you have interschema relations this is a real problem
You execute all the CREATE statements of one schema ..and after that
the ALTERs could fail ..as you refer to a different schema whose CREATE
statements are not
executed yet (as they are in another file)
I have to separate the CREATE from ALTERS in some way ...and i am working on
it now
I am not so fast 'cause I have to check all the source ..and to investigate
all the time what and how to modify
to have a smart diffs ..and I get no help on these issues that i am
targeting from the creators (or at least the guys that know the source)

I have spent my last 2 days of fixing the 7.3 issue. Maybe I will sent the
diffs before patching the multi schema problems, cause i have another work
to do also. But anyway ..these fixes will be available soon.


----- Original Message -----
From: "Scott Eade" <seade@backstagetech.com.au>
To: "Apache Torque Developers List" <torque-dev@db.apache.org>
Sent: Thursday, April 22, 2004 3:22 PM
Subject: Re: [DB Torque Wiki] Updated: PostgreSQLFAQ

> Shevek wrote:
> >On Thu, 22 Apr 2004, Scott Eade wrote:
> >
> >
> >>Shevek wrote:
> >>
> >>
> >>>So now if I don't set the id-method-parameter seqName, then postgres
> >>>because it's trying to select currval from a sequence which noone has
> >>>selected nextval from.
> >>>
> >>>And if I do set the id-method-parameter seqName, then postgres fails
> >>>because it tries to create tables with a primary key sequence that
> >>>already created.
> >>>
> >>>It looks rather like Hobson's choice, and it's documented as such at
> >>>
> >>>
> >>>
> >>7.3.4 - not cutting edge, but by no means old.  If you go to the page I
> >>reference above and read the FAQ you will save yourself a bunch of time.
> >>
> >>
> >
> >That FAQ page says under item 2 that it DOESN'T work for postgres 7.3.
> >It's quite right, it doesn't. It can be made to work by hacking the
> >generated output several times, but that isn't a solution. Please update
> >the code so that it does work.
> >
> >
> Please ask in a nicer way - e.g. by submitting a patch :-)
> >
> >
> >>>I won't say "You were right" because out of the box, it doesn't work,
> >>>can't be made to work without hacking the generated code. However, I
> >>>happily admit that it wasn't the bug I first thought it was.
> >>>
> >>>
> >>>
> >>>
> >>I am not looking for you to acknowledge that I am right, just attempting
> >>to jump to the conclusion (that it works) without covering the same old
> >>stuff.  Yes, there are a few small issues with PostgreSQL support in
> >>Torque.  I am reasonably comfortable that most of the issues are
> >>documented on the FAQ page and that none of these are show stoppers.
> >>
> >>
> >
> >It fundamentally doesn't work unless you modify the generated SQL by
> >
> >The wiki says so:
> >
> >``The situation with PostgreSQL 7.3 would appear to be quite different.
> >appears that in SQL it is enough to simply declare a column as "serial"
> >... This would mean that Tourque could be updated to not generate the
> >that creates and drops the sequence -- ScottEade 2003-02-06''
> >
> >``So this is basically a bug in Torque - i.e. it should generate
> >"tablename_columnname_seq" rather than "tablename_seq" and perhaps still
> >allow for the <id-method-parameter> element to allow for versions of
> >PostgreSQL that truncate the sequence name to 26 characters. -- ScottEade
> >2003-02-07''
> >
> >
> Wow!  So the problem you have been examining is fully documented - that
> is excellent!  If you can contribute the other issues you have
> identified to the Wiki then this would be really helpful to the next
> person or perhaps even to someone that wants to implement a solution/fix.
> >I suspect that if you do these fixes, then a lot of people will become
> >much happier, especially since postgres 7.2 isn't available any more.
> >
> >
> I will be one of them.
> Regards,
> Scott
> --
> Scott Eade
> Backstage Technologies Pty. Ltd.
> http://www.backstagetech.com.au
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: torque-dev-unsubscribe@db.apache.org
> For additional commands, e-mail: torque-dev-help@db.apache.org

To unsubscribe, e-mail: torque-dev-unsubscribe@db.apache.org
For additional commands, e-mail: torque-dev-help@db.apache.org

View raw message