Upload translation values in LSMW

There was a debate among several of the consultants I work with regarding how many values you could store in a translation table, so I decided to check it out since I had already posted something about this topic earlier.My colleagues were debating whether there was a limitation to how many entries the table could hold.  I started to think back about my previous projects where I had used translation tables.  I could not remember having used more than a few hundred, maybe even up to a thousand.  Another colleague said he had loaded nearly ten thousand, so I thought there might not be a stated limit.  I mocked up a file with over one hundred thousand entries.  Although it took a while, the big file loaded up without a problem.

This brings up an important point that I overlooked in my initial post.  I did not describe how you can upload data from your local PC into these tables.  It’s very simple.  Remember that there are two different types of tables: 1:1 and interval translations.  Also, keep in mind that these translations are specific to a single project, but they can be copied to other projects via LSMW Administration.

Refer to the example in the screenshot below for the format to load data into the 1:1 translation.  It requires just two columns of data – one for the old value and one for the new value – separated by a tab.

The example below shows the format for loading data into the interval translation.  It requires three columns: a “from” column and a “to” column for the old values followed by a column for the new value.

There are a few caveats you should be aware of:

• There is no way to identify entries as not “OK” – that is, unchecked – in your resulting translation table.  The import program only recognizes the old and new columns in your upload file.
• Do not add column headers to your upload file.  This will add another entry to the translation table which may produce unexpected or unwanted results when you use the table in your LSMW.
• Be aware of the defined lengths of the old and new values.  If the values in your upload file does not adhere to these limits, the resulting entries may be truncated giving you unexpected or unwanted results.
• Check your upload file for duplicates in the old column.  The upload program does not perform any validation or provide any warnings about duplicates.  It will simply take the last entry read from your upload file.  Duplicates in the new column is not an issue and, in fact, is quite commonplace.
• It appears that you can upload 1:1 values that fall within the interval values and no errors or warnings are raised.  The control sequence you define in the table dictates which one takes precedence.

Select the tab of the appropriate table type.  At the far right of the toolbar, you will see a button like the one indicated in the screenshot below.

There is a popup that appears to confirm whether you want to overwrite existing entries or not.

Next, you’ll be presented with a popup to select your file.  The default in the underlying program points to “C:\” so that’s where it will start.  Find your file and open.

Above is a screenshot of some uploaded 1:1 translation values.

Above is a screenshot of some uploaded interval translation values.

See?  I told you it was easy.  Now follow the instructions in this earlier post to use this uploaded data appropriately in your next LSMW!