Detecting language please wait for.......
I have a project template for a customer with TM and all. This TM is big. As I am not their only translator, I sometimes download their TM again and import it in the TM in the project template.
I might be wrong, but it seems that Trados reads all the segments and then imports the new ones as they come along. Maybe that is the only way to do it, but when it has to read 300 000 segments and import 26 new ones, this takes time. Is there any workaround, filtering or other best practises here?
Thanks for any input!
Knut Atle Klausen said:I might be wrong, but it seems that Trados reads all the segments and then imports the new ones as they come along.
Knut Atle Klausen
Would you be happy to have duplicates?
No, I do not. But some systems, like Google search, do have ways of ignoring non relevant content. I was just thinking, there must be a way of "reading" the bulk of existing segments, like 300 000, with a special variable, let's say a before date filter, so it takes lesser time to figure out what to import. For me it can take an hour importing five segments. There must be a better way.
Knut Atle Klausen said: take an hour
That is just... Silly
300,000 TU is a quite tiny TMShould be done less then a Minute
And, I guess (or promise), that is quite possible-ah, I mean not in your process, but in my process (not showed you yet though..)
First, how do you think the "non relevant content" is being classified, other than by actually reading it and comparing to the existing content?
Second, there is rather something weird in your environment, as 300k segments is really nothing...Are you perhaps using some cloud storage like OneDrive and the likes? If yes, then this is most probably the root cause.
Hi Kelly and Evzen,
Yes, 300 000 was a bad example. Sorry about that. I meant 30 million or whatever takes a looong time. My computer specifications are about average.
Evzen Polenka said:First, how do you think the "non relevant content" is being classified, other than by actually reading it and comparing to the existing content?
Well, that is what I ask for. It could read let's say a category with dates earlier then the new segments. But, again, I am not the developer. If there are no solution I accept that.
Knut Atle Klausen said:It could read let's say a category with dates earlier then the new segments.
Then you have such functionality already - use the filter during import.
Thanks, Evzen, I will try to make a date filter next time I import. That is what I am looking for. I will just import segments with a date new then my last import.
this procedure is far from the tedious "Export/Import" nonsense.so, I guess (did not make any bench test though), should be quite fast, enough...