Yael Reinhold

As usual a sample file would be more helpful... but based on your explanation alone you might have some success by editing the segmentation rules to prevent segmentation where you don't want it.

Even though the file has been translated already?

I've sent you an email with the actual file - email from Fouad Habash.

I've already manually merged many segments, but I want to merge all mergeable segments.

Please check it and let me know

Thanks

Fouad

Yael Reinhold

You said:

Yael Reinhold said:

We had to export it from Studio as a bi-lingual file and copy the content to an Excel file for translation in SDL Studio.

So I assumed you could just run the Excel through again.

Yael Reinhold said:

I've sent you an email with the actual file

However, the PO file itself really needs to go back to your client... it's clearly crap.  It opens in PO Edit, Studio (out of the box and the appstore filetype) and none of them will be able to do anything with this.  How do you even know what you're supposed to translate?

Hi Paul,

I've clarified y request by email to make it easier.

I have 2 xlsx.sdlxliff files. The first one is with the regular filetype setting, and the second one is a bilingual Excel file.

I need to merge all mergeable segments from the regular setting filetype Excel file in order to pre-translate the untranslated segments from the bilingual setting filetype Excel.

Thanks

Fouad

Yael Reinhold

Yael Reinhold said:

I need to merge all mergeable segments from the regular setting filetype Excel file in order to pre-translate the untranslated segments from the bilingual setting filetype Excel.

In this case I think I understood you correctly the first time wrt the process you should have followed.  But I also see your problem now you have done this and are where you are.  So, this is what I did:

1. Took your translated bilingual Excel SDLXLIFF and converted it to a TMX.
2. Created a new SDLTM based on paragraph segmentation.  This will ensure it will ignore all normal segmentation rules when you open the Excel source file and segment based on the cell only.
3. Imported the TMX into the new SDLTM
4. Opened the source Excel (I saved it from the SDLXLIFF you provided) and created a project with the new paragraph segmented SDLTM
5. Pre-translated the file... all done.

I'll email you the files I used... hopefully this is what you were looking for.

Yael Reinhold

And if I got that the wrong way around then this is what you need to do:

1. save the source and the target Excel from your SDLXLIFF based on the monolingual filetype
2. align them ensuring that you set the language resources of your TM to use paragraph based segmentation for both languages
3. complete the alignment and update into your new TM
4. create a project with the Excel source file and use the new TM you just created to ensure the segmentation matches
5. Turn off the alignment penalty
6. Pre-translate the file to get what you need.... OR

1. save the source and the target Excel from your SDLXLIFF based on the monolingual filetype
2. align them ensuring that you set the language resources of your TM to use paragraph based segmentation for both languages
3. complete the alignment and update into your new TM
4. use this TM in the project for your Bilingual Excel SDLXLIFF
5. Pre-translate the file to get what you need

Hopefully one of these is what you're after.  I think I covered all bases depending on which of your files is actually the right one as I'm not 100% sure!

Thanks, Paul

It worked

What about the last PO file that I've sent you, it has a different structure and Studio can't manage to handle it. Anything can be done in the settings?

Yael Reinhold

You only sent me one PO file and the content is complete nonsense.  In any PO file the format should be this:

msgid "this should be the source content"

msgstr "this should be the translation"

https://www.gnu.org/software/gettext/manual/html_node/PO-Files.html

I would not have a clue where to start with your PO file.  Studio opens it without a problem but the information extracted is meaningless of course.

Exactly, same problem on my side as well.

I assume Passolo will deal with it the same - right?

Of course... any tool that handles PO correctly will produce the same crap.  The source file you have been provided with is the problem.

Thanks Paul,

I'll share the info with the client

Thanks Paul,

I'll share the info with the client