Re: SQLl5011 error again

[Alan Cassidy <cfuture@xxxxxxxxxxx>]

Sorry for all the bother. Looks like the precompiler doesn't like the 
LIKEREC() keyword.

I changed to LIKEDS(CS) where CS was an EXTNAME(<filename>) data 
structure. That did it.

Argh

aec

On 6/9/2021 3:36 PM, Alan Cassidy wrote:
> By the way, I did compile with RPGPPOPT(*LVL2) and I'm getting this 
> anyway still...
>
> SQL5011  30    4088  Position 70 Host structure array MT not defined 
> or not
>                      usable. Reason: The host variable is not defined 
> or is
>                      not a valid structure. A valid host structure 
> must have
>                      at least one valid member.
>
>
> On 6/9/2021 3:33 PM, Alan Cassidy wrote:
> > Okay. The scoop now is, different.
> >
> > When I compile the SQLRPGLE, whether it's in RDi or from the 5250, 
> > I'm getting this:
> >
> > SQL0511. "Host structure not defined or not usable. Reason: *The 
> > datetime format of a member of the host structure array does not 
> > match the SQL datetime format.*"
> >
> > Okay, that was the only compile error remaining.
> >
> > So I went to QSQLSRC and compiled the precompiler output source for 
> > the program and it compiled! Hunh?
> >
> > My plan is this (unless I hear some alternative here):
> >
> > Test it, run it through a few lines to make sure the fetch is getting 
> > data into the data structure array. I might have to use it this way, 
> > but I don't like it, it's inelegant, uncouth. If it compiles the 
> > precompiler output source, then why not straight from the SQLRPGLE?
> >
> > --aec
> >
> >
> >
> > On 6/9/2021 11:18 AM, Alan Cassidy wrote:
> > > Spoke too soon. That was a different source that compiled, a 
> > > previous version of the same program.
> > >
> > > so back again to :
> > >
> > > SQL0511.
> > > "Host structure not defined or not usable. Reason: *The datetime 
> > > format of a member of the host structure array does not match the 
> > > SQL datetime format.*"
> > >
> > > I have put DATFMT(*ISO) into the CTL-OPT parameters  in RPG, and 
> > > added _EXEC SQL SET OPTION DATFRM(*ISO)_ to the SQL options.
> > >
> > > How about it?
> > >
> > > --Alan
> > >
> > >
> > > On 6/9/2021 9:16 AM, Alan Cassidy wrote:
> > > > I finally thought about putting DATFMT(*ISO) into the CTL-OPT 
> > > > parameters and using EXEC SQL SET OPTION DATFRM(*ISO).
> > > >
> > > > Okay, now it compiled.
> > > >
> > > > The proof of the pudding will be in testing the program.
> > > >
> > > > Thanks to everybody!!
> > > >
> > > > Alan Cassidy
> > > >
> > > >
> > > > On 6/9/2021 8:40 AM, Alan Cassidy wrote:
> > > > > I noted that this issue was resolved before,  and the way it got 
> > > > > resolved was I compiled the QSQLSRC output of the program and saw 
> > > > > what was wrong.
> > > > >
> > > > > I inadvertently overlaid that code from the previous version of it 
> > > > > (from Production) and now I'm getting a different error message 
> > > > > Reason with SQL5011.:
> > > > >
> > > > > "Host structure not defined or not usable. Reason: *The datetime 
> > > > > format of a member of the host structure array does not match the 
> > > > > SQL datetime format.*"
> > > > >
> > > > > Is there an SQL option or an RPG pre-compiler value that can help 
> > > > > for this one. I compiled the SQLRPGLE with RPGPPOPT(*LVL2) and 
> > > > > then compiled the QSQLSRC-RPG output but that didn't help either, it
> > > > >
> > > > > Is there a solution to handling the "datetime /*format*/ matching" 
> > > > > issue here? It's happening at the cursor open statement, not the 
> > > > > fetch.
> > > > >
> > > > > I /think/ it had been working but I'm not sure now... Compiling 
> > > > > the QSQLSRC output, searching IBM's web pages about it, searching 
> > > > > midrange list archives, isn't helping.... :(
> > > > >
> > > > > --Alan (aec)