Fix validate_statement to not rely on hardcoded library

[julesyan]

Changes

Fixes #549, the validate statement component was using a hardcoded library in the source. We have removed the qualifier so it uses the default library of the RUNSQLSTM job (which is already set properly)

How to test this PR

1. Connect to a system where the temp library is not ILEDITOR
2. Once connected edit an SQL file to trigger the syntax checker

Checklist

• have tested my change
• Remove any/all console.logs I added
• have added myself to the contributors' list in CONTRIBUTING.md

[github-actions]

👋 A new build is available for this PR based on f587afd.

• Download here.
• Read more about how to test

[SanjulaGanepola]

it uses the default library of the RUNSQLSTM job (which is already set properly)

If we don't qualify it, my understanding would be that it looks it up in the caller's library list or am I wrong about that?

[julesyan]

I tested this with a new library where the only thing in my library list was QGPL and QTEMP, so it looks like its grabbing the correct function @forstie might be able to help more on this

[sebjulliand]

I thinks it's going to be a fun one.
My assumption is it will depend on the naming option of the job.

• *SYS: it will look in the library list
• *SQL: it will look in the SQL PATH

Not entirely sure but from my past experience, it was something like that.
And we can't assume what the job configuration is going to be.

For now, it doesn't work (it says "CHKSTMNT0001 of type *N not found in ILEDITORX"):

Even though it exists:

[SanjulaGanepola]

Why not make the library a parameter on the SQL function we are creating so when we calculate the hash value, it is not dependent on any hardcoded library name?

[sebjulliand]

No, but I'm not sure that would work (assuming IN_LIBRARY is a parameter):

call IN_LIBRARY.CHKSTMNT(...)

Worth a try though!

[julesyan]

Yea setting it as a param does not work, i get [SQL0204] CHKSTMNT0001 in LIB type *N not found.

[julesyan]

Ok @sebjulliand if you could test now with your case. This worked on my local but im not sure if im missing a case

[sebjulliand]

Great job @julesyan , it works like a charm now!
Tested with *SYS and *SQL naming. No issues.

I just had to delete the function because of my previous test, but that won't happen to the users.