Java <JCo> SAP XI <RFC> SAP R/3 6.C ... Unicode issue

[Former Member]

Hello,

My scenario is very simple. I'd like to connect a Java program to a SAP R/3 (MDMP) for RFC calls (in both direction), using SAP XI (some of the RFC calls will redirected to some other systems). I use SAP JCo v2.1.8 and tried with Java 1.4.2_09 and Java 6.0.

It works well when the Java program is calling a function module, but when SAP R/3 is doing it, the returned texts looks like: S#A#P# #J#C#o# #v#2#.#1#.#8# #(#2#0#0#6#-#1#2#-#1#1#)# (the text should be SAP JCo v2.1.8).

It looks like an issue with Unicode. Although I didn't check the Unicode checkbox in the RFC sender adapter (the RFC receiver one doesn't offer the possibility to specify unicode or not).

Do anyone have a solution ? Is it a code in the Java program or something to set in SAP XI / SAP R/3 ?

Thanks in advance for your help.

[Former Member]

Finally, the issue is linked to the fact that string in abap are fixed sized and in java are variable. By padding the spaces at the end, this issue is solved.

[Former Member]

This is a workaround solution. In the case of this prototype it is ok, but the final plan is to migrate a web based application. It was directly connected to one SAP R/3, but now we would like to connect it to SAP XI because a few RFC calls will be re-directed to some other systems, but the majority of them will we'll still be sent to the SAP R/3. We don't want to make development on this part for cost reason (and time). Plus on some text there are the "#" character that is legitimate.

[Former Member]

I you can't figure a way out of this, just use a UDF in mapping and use regex and replace # with null, there's probably a better way though.