Thread: ecpg compile error in regression tests

Hi!

Still working on the ecpg-regression-tests-on-msvc. Update to follow
later today I hope. But for now:

I get the following error when trying to build the sql/parser.pgc test:   c:\prog\pgbin\pgsql\bin\ecpg --regression  -o
parser.cparser.pgc   parser.pgc:26: ERROR: syntax error at or near "NULLS"
 


Any pointers for where to look? (Kinda seems it could be the ecpg binary
not being up-to-date, but I did a make clean and rebuild of it)

//Magnus

On Mon, Jun 11, 2007 at 07:03:05PM +0200, Magnus Hagander wrote:
> I get the following error when trying to build the sql/parser.pgc test:
>     c:\prog\pgbin\pgsql\bin\ecpg --regression  -o parser.c parser.pgc
>     parser.pgc:26: ERROR: syntax error at or near "NULLS"
> 
> Any pointers for where to look? (Kinda seems it could be the ecpg binary
> not being up-to-date, but I did a make clean and rebuild of it)

I get the same error IF I try using an 8.2 parser, be it in the backend
or in ecpg. With 8.3 however it works nicely.

Could it be that you have some older versions around? Yes, you did a
rebuild, but still.

Michael
-- 
Michael Meskes
Email: Michael at Fam-Meskes dot De, Michael at Meskes dot (De|Com|Net|Org)
ICQ: 179140304, AIM/Yahoo: michaelmeskes, Jabber: meskes@jabber.org
Go SF 49ers! Go Rhein Fire! Use Debian GNU/Linux! Use PostgreSQL!

On Tue, Jun 12, 2007 at 09:57:17AM +0200, Michael Meskes wrote:
> On Mon, Jun 11, 2007 at 07:03:05PM +0200, Magnus Hagander wrote:
> > I get the following error when trying to build the sql/parser.pgc test:
> >     c:\prog\pgbin\pgsql\bin\ecpg --regression  -o parser.c parser.pgc
> >     parser.pgc:26: ERROR: syntax error at or near "NULLS"
> > 
> > Any pointers for where to look? (Kinda seems it could be the ecpg binary
> > not being up-to-date, but I did a make clean and rebuild of it)
> 
> I get the same error IF I try using an 8.2 parser, be it in the backend
> or in ecpg. With 8.3 however it works nicely.
> 
> Could it be that you have some older versions around? Yes, you did a
> rebuild, but still.

I'll try to blow away my tree and do a fresh cvs checkout later. I'll go
ahead with what I for now and get it in thoguh, don't want to throw my
changes out. And since the problem is not actually with the regression test
*engine*, it's with the tests themselves.. :-)

//Magnus