CafeTran

HI all,

I have a question on how partial matches from the TM are displayed.

It used to be the case that the TM window showed partial matches as the current source, the partial match from the TM and the translation from the TM.

Now it shows some horrible bastard hybrid of the current source and partial match with all sorts of colouring and crossings out. Is there some setting to restore sanity? (I couldn't find one.)

Also a minor bug to report on this Igor:
The colouring function can't cope with duplicate words.
I recently had a sentence roughly of the form: die Berliner, die die Berliner essen, CT crossed out one of the 'die's and added in a new one in yellow. That's not right.

The horrible bastard hybrid representation is Igor's answer to user requests of Michael B. and others. Perhaps you can make it more intrusive by changing the colours for added and deleted text? For example, you could decide to make deleted text white (white text on a white background). But I doubt that this is what you want. Currently, it is not possible to return to the old view.

Thanks for your reply Hans. I did come across this post when I was looking for a solution, and did wonder if this was some horrible solution to a non problem.
I can see that this could be useful where there are simple additions or subtractions, but for anything else it's completely unusable.

This is what I get from a typical partial match (blurred for confidentiality):

I'm working on an update of a previous translation, and there are lots of segments like this with sometimes minor changes, for which this is completely unhelpful.
Please put it back the way it was Igor, or introduce a high match % threshold for using this view. This is rubbish.

HI Igor,

I just realised that part of the problem is just that the matches aren't being displayed properly. I have a segment which is broadly identical to a memory segment, but it's not displayed as such, because of the way CT tries to match up bits of the segment.

Can I PM you a screenshot (can't post it here for confidentiality reasons) to illustrate the problem?

Jeremy