MDMP Non-Unicode Conversion to Unicode

[Former Member]

Hello,

We are doing combined unicode conversion and technical upgrade from R3 4.7 to ECC6 EHP6. Our system is MDMP (with ZH, EN and DE) but we only use EN and ZH. After running all the scans (consistency check, tables w/o lang information, tables w/ ambiguous lang info, INDX Analysis, Tables w/ Lang Info), I have come up with 350000+ entries in the vocabulary. However, I have found words in different language (NW, PT, CS, FR, DA, etc.). What should I assign to these words? I can only see the 3 active language and unknown.

In addition, I have run the Reprocess worker jobs and there were tables with warning (Reason 1 and D). IN the reprocess logs, it says that I need to assign a language. But instead of one word per row, the words are combined in 1 line. There are some lines that has garbage texts and English words.. I logged in to ZH to check the characters. it was chinese with letters in between and english words.... I checked the table (sample table TSTCP - with only 2 columns) but there's no chinese entries specially in the key fields reported by reprocess log. How come it appears garbage/chinese text in the reprocess logs? What language should I assign to this?

Thanks in advance.

Regards,

Eco

[Former Member]

Hi Eco,

These are (NW, PT, CS, FR, DA) Western European language. Did you run the report RSCPINST before starting of scan. Please run this report and check all active / installed language and adjust zcsa/installed_languages profile parameter.Also check SMLT.

Regards,

Dipam Phukan

[Former Member]

Hello Dipam,

Yes, I ran this report and we have only ZH, DE and EN. Same result in SMLT. When checking these entries in the tables, they are pre-defined by SAP and not used by the customer. However, I'm still confused what language to assign.

Thank you,

Eco

[nils_buerckel]

Hi Eco,

in SAP-Standard there is no language NW ...

PT,FR,DA belong to code page 1100 (Western Europe) - hence you should use a supported language in your system which is assigned to code page 1100. This should be the case for DE, but in the standard MDMP case, EN is also assigned to 1100.

CS is an Eastern European language, which uses code page 1401 in Non-Unicode. As far as I understand, this code page is not supported in your system. Hence I would recommend to assign e.g. EN and repair the data in SUMG if necessary. In many cases (e.g. if you do not plan a rollout into Eastern Europe) this SAP delivered data is not needed in the Unicode system.

However for table TSTCP please implement SAP note 897002 - then there should be no standard entries in the vocab anymore .

Best regards,

Nils Buerckel