Trados Business Manager
Speech to Text
Managed Translation - Enterprise
Translation Management Connectors
Machine Translation Cloud
Machine Translation Connectors
Machine Translation Edge
Tridion Docs Developers
RWS User Experience
RWS Community Internal Group
RWS Access Customer Portal
RWS Professional Services
RWS Training & Certification
Language Technology Partner Group
RWS Enterprise Technology Partners
Trados Academic Partners
Trados Approved Trainers
ETUG (European Trados User Group) Public Information
Machine Translation User Group
Nordic Tridion Docs User Group
Tridion Docs Europe & APAC User Group
Tridion UK Meetup
Tridion User Group Benelux
Tridion User Group New England
Tridion User Group Ohio Valley
Tridion West Coast User Group
WorldServer User Group
Trados GroupShare Ideas
Trados Studio Ideas
Language Weaver Cloud Ideas
Language Weaver Edge Ideas
RWS Language Cloud TMS Ideas
RWS Language Cloud Terminology Ideas
RWS Language Cloud Online Editor Ideas
Managed Translation - Enterprise Ideas
Tridion Docs Ideas
Tridion Sites Ideas
LiveContent S1000D Ideas
Events & Webinars
To RWS Documentation
To RWS Support
Detecting language please wait for.......
No matter how powerful your computer is, there will be a huge translation file which can be nerve-racking during a busy day. I am talking about huge translation files which contain more than 100,000 words with only 1,000 new words to translate. Normally you would finish this amount of work in two hours or less but if the file is big, your computer will slow down and you will spend more time.
There is a tool which can help you slice these 1,000 new words into a separate file and it is called SDLXLIFF Toolkit. I can see that around 3500 people have downloaded this tool but I do not know if everyone uses its slicing feature. I thought it might be useful to talk about it.
We just need to use the slice button right after specifying the segments to be sliced and exported into a new .sdlxliff file. We can extract fuzzy segments, new segments, 100% matches or only unlocked strings. I will slice not translated strings in my file with below options.
The slicing time can increase depending on the file size but it is worth waiting. The resulting file will look like this.
Once you translate these segments you might of course update your TM with them and use pre-translate function to apply your translations to the original file. I assume you know how to do that. In this post, I will use Perfect Match function (only available in Pro and higher versions of Studio) which is a safer method. With perfect match, you can be sure that only the related segments are updated and the status of those segments are in your control. For example, I set the newly translated strings’ status to signed-off and I want to see it like this in the original file.
We just right click original file and choose Apply Perfect Match under batch tasks. We match it with the extracted and translated file and remember to choose “Use the original translation origin and status” option to make sure that the status of the related segments are the same.
The result will look like this.
I wish the Perfect Match feature were available in all versions of Studio. It would help freelance translators in cases like these. However, I am sure this guide will also help PJMs and translators working in Pro versions of Studio. Of course the freelancers can go with pre-translate function and review them afterwards to make sure everything is in its place. Feel free to reach out for questions and comments.
Evzen Polenka said:Of course it's necessary to have the ability to reconstruct the original file from the individual pieces. Because the client of course wants the original file, not the pieces.
Well that's not exactly the process we are discussing here Evzen, of course nobody is going to send back pieces! I merely asked whether having this option to split in the toolkit would be useful. The toolkit method is different and pretty simple when you use Perfect match as described in the original post.
I'm not discussing the problems with split and merge... try and stick to the point ;-)