PS: my client is an agency working with many other translators and I am the only one encountering this problem.

Hello Anne-Laure Guichard

Here on the screenshot, in segment 608, could you please confirm where the match is coming from? 
It is already there from somewhere, but if you could not find it in the TM as an automatic lookup match, where does it come from? Was the file pre-translated by the agency?

Maybe the memories themselves do not contain these matches? If I understand correctly you do not receive neither automatic matches, nor concordance matches - could you please confirm?

Could you please run an analysis on the file and post here the results, similar to this?

Also, do you know maybe what version of Studio is the agency using?

Looking forward to hearing back from you.

Best wishes,
Greta

Dear Greta,

Thank you very much for your answer!

I am going to try to answer your questions.

I don't know exactly from which of the many client's TMs in the package the match in segment 608 is coming from, but I know I can find it through a contextual search when working with a single file and with the client's TMs sent separately with their names not beginning with "Request_...", which you can see in my parameters screen cap (I forgot to say this). 

So I would say the TMs with their names not beginning with "Request_..." do contain this matches. As for the ones beginning with "Request_", I don't know...

Nor do I know if one may say the client pre-translated the file, but the fact is that when I open it, matches are already displayed in the right column, for example for this segment with a 86% match.

I am not sure to understand the difference between automatic matches and concordance matches: if concordance matches mean what I get through a contextual search ("recherche contextuelle" in French), then I don't get anything when working from the package (only when working from a single file).

If automatic matches mean the text displayed in the right column, then I get them, they are displayed before I click in the target segment.

Here are the results of the analysis:

As for the agency's Studio version, it is 2019. But the other translators of packages who have 2015 like me don't have this problem.

I hope this helps you to understand?

Kind regards,

Anne-Laure

Dear Greta,

Thank you very much for your answer!

I am going to try to answer your questions.

I don't know exactly from which of the many client's TMs in the package the match in segment 608 is coming from, but I know I can find it through a contextual search when working with a single file and with the client's TMs sent separately with their names not beginning with "Request_...", which you can see in my parameters screen cap (I forgot to say this). 

So I would say the TMs with their names not beginning with "Request_..." do contain this matches. As for the ones beginning with "Request_", I don't know...

Nor do I know if one may say the client pre-translated the file, but the fact is that when I open it, matches are already displayed in the right column, for example for this segment with a 86% match.

I am not sure to understand the difference between automatic matches and concordance matches: if concordance matches mean what I get through a contextual search ("recherche contextuelle" in French), then I don't get anything when working from the package (only when working from a single file).

If automatic matches mean the text displayed in the right column, then I get them, they are displayed before I click in the target segment.

Here are the results of the analysis:

As for the agency's Studio version, it is 2019. But the other translators of packages who have 2015 like me don't have this problem.

I hope this helps you to understand?

Kind regards,

Anne-Laure

PS: Here a screen cap of my TM parameters when working with a single file and with the client's TMs sent separately with their names not beginning with "Request_...":

Here is what happens when clicking in target segment 608 (nothing):

And here is what happens when I make a contextual search on a part of the sentence:

All of these 3 caps are for working with the single fr-FR file without opening the package and with the client's TMs sent separately with their names not beginning with "Request_..."

Best regards,

Anne-Laure

More about my last screen cap:

- when I select the whole sentence the search result is zero (for an 86% match...);

- when working from the package, it is actually the same: when I make a contextual search on a part of the sentence I get something but when I select the whole sentence the search result is zero.

But in some cases, when working from the package and there is only one word in the source segment, then I can't do anything else but selecting the whole "sentence", and the search result is still zero.

Here an example from another project:

(Please excuse me for these many messages but all this is pretty confusing for me too!)

Kind regards,

Anne-Laure

Hello Anne-Laure Guichard

Apologies for the delay in my reply, I have not seen your messages.

For future reference, kindly tag people using @ and their name (without a space) so they would get notified.

There are quite a few things to check and discuss here, and might be easier with live voice and screensharing - I wonder if you would be available for a Skype call tomorrow afternoon, from 14:00 or from 16:00 CET?

Looking forward to hearing from you.

Best wishes,
Greta 

Dear Greta Magyar

Thank you for your reply, yes a Skype call would (hopefully) be useful!

Today I was not available and tomorrow (Thursday) I am not either, but I will have time on Friday afternoon: what about you?

Kind regards,

Anne-Laure