Luka Lukić

Two bugs confirmed and logged with the development team:

CRQ-18036
Adding terms from Studio for Serbian termbase will add both source and target term to the same language

CRQ-18037
Term recognition does not find terms from Serbian termbase

I can't give you timescales for the fix but it will be resolved in a future update.

Luka Lukić

Paul said:

CRQ-18037
Term recognition does not find terms from Serbian termbase

The development team have done some good research on this one and have concluded that this isn't actually a bug in Studio/MultiTerm.  I tested this again this morning and can confirm that I can now make this work.  I'll give you the full explanation:

===========================

The example uses Serbian (Latin, Bosnia and Herzegovina) but if you look at the text you will notice it is Cyrillic, not Latin. Unlike our Cloud-Based Terminology, which is probably unicode throughout, MultiTerm still converts text using windows code pages - 1250 for the Serbian Latin and 1251 for Cyrillic.

Term recognition works as expected using Serbian (Cyrillic, Bosnia and Herzegovina) for the document and termbase.

=============================

In your example you are also using Latin and not Cyrillic.  I don't know how we missed this the first time!

Paul said:

CRQ-18036
Adding terms from Studio for Serbian termbase will add both source and target term to the same language

This one is still unresolved.  But if you use the correct languages it seems to work so I'm not sure this needs any work to solve the problem you have reported here.  Probably not.

Hi Paul,

unfortunately it seems that this issue isn't solved. Because I still have the same problem. In your example there you have set Cyrillic, but I'm using Latin in TB and in the project. See the following screen shoots.

Here are the TB settings. After them the project settings and the last screen shot is what happens when I use the function F2. I didn't have this problem when I use the language combination otherwise. Must say that I'm really disapointed regarding this and it's very annoying when you buy a tool and it does not work as intended.

Luka Lukić

Luka Lukić said:

unfortunately it seems that this issue isn't solved. Because I still have the same problem. In your example there you have set Cyrillic, but I'm using Latin in TB and in the project. See the following screen shoots.

Technically this is not the same problem you originally reported.  Please can you provide the following for me?

• your termbase
• your original source file

You can send these to pfilkin@sdl.com if this is acceptable for you?

Yes, it is the same problem, because when I mark the term in source and term in target and use the function CTRL+SHIFT+F2 it add both terms to German instead one term to German and one to Serbian. You can see it on the last screen shot in the Term Base viewer, there is only German available and the serbian and german term is undo the german term. On this screen shot you don't see the full source text, because of the TB viewer.

Will e-mail it to you right now.

Yes, it is the same problem, because when I mark the term in source and term in target and use the function CTRL+SHIFT+F2 it add both terms to German instead one term to German and one to Serbian. You can see it on the last screen shot in the Term Base viewer, there is only German available and the serbian and german term is undo the german term. On this screen shot you don't see the full source text, because of the TB viewer.

Will e-mail it to you right now.

Thanks Luka Lukić

I probably wasn't very clear:

Paul said:

Paul said:

CRQ-18036
Adding terms from Studio for Serbian termbase will add both source and target term to the same language

This one is still unresolved.  But if you use the correct languages it seems to work so I'm not sure this needs any work to solve the problem you have reported here.  Probably not.

So, if I set up a termbase like this:

Then it works.

In practice this should not be needed as it means you are forced to use specific languages rather than just the main group, but in all honesty I don't think this is anything we will ever fix in MultiTerm because the effort is too great for a product that is this old.

If you use "SDL Language Cloud Terminology" you should find this works without any problems, and I believe you can also use the Excel Terminology Provider from the appstore which may be a good workaround if you don't have complex definitions for your termbases.

So you have three workarounds, one still using MultiTerm and two others.  Hopefully this is going to help you?

Hi Paul,

the easiest way to explain it, is to check the option "Show sublanguages" by creating the TB. Thank you for the assistence.