Re: SQL_SUPPRESS_WARNINGS

On 2/14/2011 10:12 AM, rob@xxxxxxxxx wrote:

According to
http://www-912.ibm.com/n_dir/nas4apar.nsf/c79815e083182fec862564c00079d117/c5040ccf8ee78d22862577cb003c8e67?OpenDocument
when running 8.1 of LX or above it is no longer required to turn on
SQL_SUPPRESS_WARNINGS in the QAQQINI file.

One developer is concerned that if we turn that off then custom code may
fail.
Apparently the big concern is if you did something like
'Select * from myfile into :myfileDS' and the number of fields in the
table has changed but the DS has not been recompiled to take advantage of
this then a hard error can ensue on V5R1 or above. Infor has weeded this
error out of their code but our developer is concerned about our code.

Is there any advantage to turning this off? Hate to turn it off and
potentially have errors if there is no potential benefit.

Not to be pedantic, but the benefit is that you're not suppressing errors. If you're counting on code like the above (select * into DS) and you haven't recompiled, then you're courting disaster anyway. Any sort of changes to the file, including expansion of fields, could trigger the same hard error.

I would think you would want to know about such problems sooner rather than later, and then get into the habit of recompiling programs that are at risk.

On another subject, I try to avoid SELECT * in production. It removes a lot of the benefits of SQL provides in the first place.

Joe