How R3load interacts with SPUMG and SUMG
The tool which performs the database export and import and thus the Unicode conversion is R3load. Usually, R3load runs with the installation tool SAPinst. R3load uses the information prepared in SPUMG to convert the data during the export phase, and creates the R3load log containing the data that could not be converted properly. The R3load log and the Reprocess log (which is not used by R3load, see picture below) are imported to the Unicode database to be available for postprocessing steps in transaction SUMG.
The most important SPUMG and R3load interfaces that are used during the export of the database are:
As the database of the source system cannot be accessed when the database is imported, the Unicode conversion can only be carried out during the export of the data.
The picture below shows the basic database conversion steps of an MDMP system:
1. You start the export of the data from non-UC database with SAPinst (internally R3load).
2. While the export of the non-Unicode database is running R3load converts the data and writes R3load logs.
3. After the installation of new database on the target system with SAPinst, R3load executes the import of the converted data into the UC database.
4. After the import, you perform the necessary postprocessing steps in transaction SUMG (only required for MDMP systems).
Export Control Table
The Export Control Table combines the key information of a table for R3load. This includes the table category (TABCAT1/2 or 3), information about language fields or a "character set for all entries", that is, a character set that is used to convert all the data records. The Export Control Table is filled automatically by the SPUMG scans and should only be changed manually in exceptional cases.
The Language List contains the assignments between languages and character sets. This assignment is used for all data records in TABCAT1 tables. The Language List is not shown in the picture above.
Every data record of the texts with special characters outside the Common Character Set has words in the Vocabulary that have been assigned a language. This language information is used during the export to determine the source character set for the conversion to Unicode.
.log files: Not required for postprocessing in SUMG but important for error analysis. (R3load Log Files)
.xml files: Are written only for Unicode Conversions. Contain all information required for MDMP postprocessing in SUMG.