Almudena Ballester 

I would say, especially since they seem to be segmented nicely, that the best solution would be this:

1. filter on the numbers
2. copy source to target
3. lock them (I would not confirm them as you probably don't want them in your TM)
4. filter on unlocked
5. Job done

Thanks, Paul. That's the workaround we are using now, but, from your answer, should I conclude, there is no way to change the settings in TM to do not recognize those as numbers? There is no number with such format in Spanish or Catalan (one digit number-dot_one digit number-dot) - they shouldn't be marked as numbers anyway.

Pre-processing projects like that is not always an option, and the TM is online with autopropagation.

Thanks in advance,

Almudena

Almudena Ballester 

Almudena Ballester said:

should I conclude, there is no way to change the settings in TM to do not recognize those as numbers?

Well... I think it' a bit tricky to recognise these separators as thousands and decimals for numbers which are clearly not real numbers.  What would 5.3.1 really be for example.  So I would say no, it's not possible.  What you could do is use the Terminjector (appstore) perhaps to recognise this pattern and substitute the TM result with the pattern you want when you pre-translate.

Almudena Ballester said:

Pre-processing projects like that is not always an option

Why not?  You could use the SDLXLIFF Toolkit for example and then do this pre- or post- pretranslation, but before you make the files available to your translators.  You can also do this for multiple files in a project or even multiple projects.

Almudena Ballester 

Almudena Ballester said:

should I conclude, there is no way to change the settings in TM to do not recognize those as numbers?

Well... I think it' a bit tricky to recognise these separators as thousands and decimals for numbers which are clearly not real numbers.  What would 5.3.1 really be for example.  So I would say no, it's not possible.  What you could do is use the Terminjector (appstore) perhaps to recognise this pattern and substitute the TM result with the pattern you want when you pre-translate.

Almudena Ballester said:

Pre-processing projects like that is not always an option

Why not?  You could use the SDLXLIFF Toolkit for example and then do this pre- or post- pretranslation, but before you make the files available to your translators.  You can also do this for multiple files in a project or even multiple projects.

Thanks again, Paul. I will explore the SDLXLIFF toolkit option - until now I was using the advanced filter with search expression ^\d to spot all the segments beginning with a number.

Regards,

Almudena

Hi Almudena Ballester 

Almudena Ballester said:

until now I was using the advanced filter with search expression ^\d

Me too!

BTW: This regex in Source would give you better matches:

^\d+\.

or this one:

^[0-9.]+$

Thanks for the tip, Jesús

I was thinking about an specific training for our project managers, for they to get most out of this advanced filter when it comes to detecting problematic segments. I dare to ask you as well-versed ;), Can you advise me to find a page with these resources on regexp well explained, in particular those used in Trados?

Cheers,

Almudena

Hi Almudena Ballester ,

This is what comes to my mind:

• I remember that Nora Díaz  published some posts in her blog (noradiaz.blogspot.com) about regex.
• Our colleague Anthony Rudd  published a couple of books about regex, one of them specifically for Trados. I can see these books in the big store with the smiling logo. Anthony call tell us more about his books.
• In my own blog (https://www.gonduana.com/etiqueta/regex/) I cover some basic regexes.

I'd recommend knowing regex. It's true that mastering all the Regular Expressions can take a lot of time, but I believe translators can get a lot just for investing a couple of hours in learning regex.

Almudena Ballester Jesús Prieto 

And just to add to it ;-)

https://multifarious.filkin.com/tag/regex-tutorial/

Many thanks to both of you

Almudena

Thanks for the shoutout, Jesús Prieto ! Almudena Ballester, all my regex info is now compiled and is part of The Translator's Tool Box book (www.thetranslatorstoolbox.com), in case you're interested!