attribute 'Unicode checks active'

[Former Member]

Hi,

we just made an upgrade from 4.6 to ECC 6.0.

The unicode conversion check and the unicode conversion itself is done.

Each Abap program that has the attribute 'Unicode checks active' not checked ( off ), is causing a syntax error in the new system. Up till now we manually adjusted each attribute of each program, just to get the basic business transactions running.

My question is: what went wrong, and what is the best way of solving all this errors at once ?

Thanks in advance,

Rolf

[Former Member]

Hi,

You can see syntax errors of all objects in the transaction code UCCHECK.

Thanks,

Satya

[Former Member]

Satya:

There are a lot of Z programs AND standard SAP programs that do not appear in the uccheck, and that DO generate the syntax error "The program ... is not Unicode-compatible, according to its program attributes."

After i change the program attribute to checked, the syntax error disappears.

The question remains: what went wrong here ? Did we forget something during the conversion ?

[Former Member]

hi...

there is nothing wrong with your conversion, your programs will work just fine without the Unicode flag checked.

Unicode is just a different way of storing data...

You need to check all data definitions...some data types like RAW or at times P are not supported by Unicode.

All these data types need to be changed to C or string...also you might have to change the way some statements are written, maybe I would be able to help more if you send me an example of the error and the corresponding statement in the code.

[Peter_Inotai]

You can ignore the SAP programs, unless there is a user exit is activated within them. SAP didn't convert all the old programs to Unicode. It's normal.

For the Z-programs, you have to make sure:

- they don't give any syntax errors (eg: in Code Inspector)

- they work as expected

The selection scrren of UCCHECK is quite tricky, because by defulat it excludes quite a lot of objects. I'd suggest to double check the selection screen. maybe pick one example, and enter only that one to the selection screen.

best regards

Peter

[Former Member]

Hi rof,

just go to transction UCCHECK GIVE THE PROGRAM NAME AND CHECK IOT OURT AND SEE THE ERROR. THE ERRORS AR MINOR ONLY

SIMPLE ERROR IS

OPEN DATASET FILE FOR INPUT IN TEXT MODE

TO FIX THAT ERROR

OPEN DATASET FILE FOR INPUT IN TEXT MODE ENCODING DEFAULT.

[Former Member]

In the transaction UCCHECK there is a possibility to set the Unicode attribute for all programs that are checked and corrected.

Doing so, is the solution to the problem.

Thanks to all.

Rolf.

Message was edited by: Rolf Schorpion

Message was edited by: Rolf Schorpion

[Former Member]

Rolf,

I'm using ECC 6.0. I tried to do what you suggested for setting unicode flag active for a bunch of programs. Unfortunatelly, I'm bumping in the following problem:

I run UCCHECK and I get "green light" for some objects that don't have unicode flag active yet. So I click on the correspondent line and after on "Set Unicode Attribute" button. I get a pop-up saying that unicode is set active and program is written into a transport request... i click on OK. Another pop-up comes up. "Set Unicode Flag?", I click on Yes. Then, another pop-up comes up saying "Set Flag for Error-Free Programs Only?". I tried Yes, and I get this information message "The current system is not the original system". I tried also NO, but same message is shown.

Any idea on how to get it right?

Thanks.