cancel
Showing results for 
Search instead for 
Did you mean: 

BPM 7.2, analytics, multilingual support

Former Member
0 Kudos

Good day!

We are creating reports by using new BI content and we have a questions about multilingual support.

As we can see the task definition name of a BPM task is the physical name of the task (I mean it is the name of the file task_name.task). Of course we can use user texts, but they appear as a task instance name, not the task definition. Instance task name always have variables so it's not possible to group task properly.

The problem appears with the process definition also - it shows us the name of the process model...

Are there any options to have process and task definition names be translated?

Accepted Solutions (0)

Answers (2)

Answers (2)

Former Member
0 Kudos

Hello,

We are facing issues with the process analytics part and I hope you can help us out. With CE7.2 and BI 7.03, we are trying to get the process analytics data for a simple BPM process we deployed in the CE server. The help document says that once the process is deployed the data sources in BW would be created. Are we missing on something over here??

Former Member
0 Kudos

Hello,

The only data sources defined in BW are the default ones which are part of BI content. Any custom reporting objects you define in your BPM process will have to be created manually in BW.

http://help.sap.com/saphelp_nwce72/helpdata/en/a3/92f601cc1d49c292527841a9ca76a8/content.htm

O.

0 Kudos

Dear Kirill,

You can translate processes and tasks texts. Here the procedure for translating process texts:

1. Go into src/bpmn and duplicate the XLF file.

2. Rename the duplicated file with for example the locale code at the end (for example: myprocess.bpmn_en_US.xlf).

3. Double-click on the file to edit it and choose the source language in the drop-down list (for example American English).

4. Click then on the Resource Text tab and translate the texts.

5. Build and deploy the DC.

6. Run at least one instance of the new process version
It is required to run at least once the process in order that the texts are visible for hte BPEM/BI connector.

You need then to extract again the texts from the process text datasource, execute the data transfer processes and may be to activate finally the master data content. You will then be able to see the process texts with your user locale in the Query Designer.

For tasks, the procedure is almost the same. The XLF files are located under src/tasks.

Finally, please notice that some issues were found in 7.20 regarding the texts (text bundles are not packaged and are therefore not deployed). Fixes and patches are in preparation (see note 1412368).

Best regards,

-Alexandre

PS: When the process or task texts are translated, the default texts - the one defined when the process or the task was created - are no more extracted. It's therefore recommended to create an additional XLF file with the same locale than the DC locale.

Edited by: Alexandre Gueriot on Feb 23, 2010 2:15 PM

Former Member
0 Kudos

Hi Alexandre!

Thank you for your answer, it gives me a clue.

But that doesn't work for me in a right manner...

I found the name of the task in xlf file (it was the same as the file name without an extension), changed it to required. New xlf files isn't included to sda, as you have mentioned, so I changed existing xlf files. Deployed new process, started it, update the BI/BPM characteristics, but the was not any changes.

For example I have a task with filename FillBusinessTripTask.task. I changed FillBusinessTripTask in the xlf file to desired.

When I reload characteristics it nevertheless shows me FillBusinessTripTask (I look on it in PSA 0GLX_DEF_TEXT). Also there is en_US locale only, but my default DC locale is ru_RU. Is it an error?

It's important, it's not instance texts, it's definition texts!

Because of an error with new xlf files I can't check whether the new xlf file with fillbusinesstriptask.task_ru.xlf (explicitly locale) will help.

But for the future am I right in may steps to have a correct characteristic texts in my analytics?

0 Kudos

Dear Kirill,

There is actually here two different issues. The first one is that the translated texts get not deployed on the system. Including manually the XLF files in the SDA is not sufficient unfortunately. Some other files need some changes. Basically, it works when you can find your texts in the table GLX_RRH_TEXTS. The BPEM/BI connector extracts the texts from this table.

The second issue limitation is that the locale of the DC is lost when the default texts are transfered to BI. When a process is deployed, the default texts are installed in the GLX_RRH_TEXTS table without a specific locale (value is '_'). The reason is that the default texts should be used as fallback when no translation is available. When the BPEM/BI connector extracts the texts, it assigns the Java system locale to the texts to replace '_'. In your case, I can imagine that the Java system locale is en_US. We are currently working on this issue (a note will come soon) and for the moment, a workaround would be to create a new XLF file with the locale ru_RU so that the texts be imported with the locale ru_RU.

Best regards,

-Alexandre

Former Member
0 Kudos

Hi Alexandre,

I think I understand the way it has to work:).

So I have to wait the release of the note which repairs the translation?

BTW, where is the table GLX_RRH_TEXTS? I failed to find it neither in backend, nor CEP DB.

0 Kudos

Dear Kirill,

The table GLX_RRH_TEXTS is located in the CE database (normally SAP<SID>DB). Depending on the DBMS you're using, you might need to add a namespace.

I created a note for wrong locale issue in BI for BPEM process & task texts:1442372. The note will be released in the next days. You can track there the progresses of the fix.

Best regards,

-Alexandre