CafeTran

I'm sure I've read this somewhere, but I can't find where and I think it's an issue.

When translating from French into Spanish, the French "élision" is considered as a whole word, so termbases don't recognize it as a word (nor the QA). Example:

L’élément - L’élision - L’invention and so on are not recognized even if I record "élement", "élision", "invention". I usually use this (') as an apostrophe for French, but my project manager sends me the documents with this (’) instead (I guess it's due to Abbyy FineReader, as we work with PDFs converted into .docx files).

> I don't know if this also happens with Glossaries.

It does.

Hello,

I have just completed a project with the latest CafeTran version and I have noticed a few things that could be improved:

1. TM matches highlighted in the source segment editor seem to also highlight parts of words that contain the TM match.

So for example, if you ad "or" or "and" as a fragment, any word with these letters will have these highlighted, such as as "more" or "understand".

Is there a way to avoid this? Is it related to the latest prompter (auto-complete) improvements?

Screenshot: 

2. In the target segment, even small words are cut to the other line sometimes. This makes it difficult to read.

Example:

Here, the word is "simplement", the t appears in the next line. This happens for small words such as "qui" Which appears as:

qu

i

This may also be related to the prompter update, and suggestions in the middle of a word.

3. Finally, there has been an issue with an Excel file export which lost paragraph marks within one cell, where I have used the split feature, but I prefer to open a ticket for this as I need to attach the project files.

Jean

Someone will be happy here ;).

Yep. The 'bilingual import' has been fixed. Greatz!

Just out of curiosity. What changes are there to the MS Excel filter? Sometimes it did not work for me, when I wanted to do “Bilingual Excel”.

>> Sure, please turn it off in Preferences > Prompter tab.

>Uh, I did not have the last edition installed. Perhaps this should be deactivated by default. 

I agree. The current default behaviour has annoyed me too, this week. 

> Sure, please turn it off in Preferences > Prompter tab.

Uh, I did not have the last edition installed. Perhaps this should be deactivated by default. 

Or at least consider commas, full stops and parenthesis as not part of a word, to avoid their deletion (this would perhaps also be a plus for all these frog legs eaters that do not have composite words). 

IMHO "autocompletion in the middle of words" is one of the best features added in the last monhs.

However it still has some really important flaws.

• When used with composite words (yeah, German!) it deletes the rest of the following word. This is not always desired (e.g. when changing undercapacity to overcapacity (having marked the first five letters "under"), AC proposes "over" first and then deletes the rest of the word ("capacity").
  
• Same with following commas or closing parenthesis. They are simply overwritten/deleted. This is even worse.
  
• Same with following words (sometimes), this should not happen at all. The last two cases can perhaps be seen in conjunction with the first one – CT does not recognize spaces, commas or parenthesis as word limits. 
  

Could there be some more finetuning for the new upper/lower case AC feature.

When typing MAss (second letter uppercase by mistake), AC completes to MAssACHUSETTS.

Could there please be an AC to Massachusetts or MASSACHUSETTS only (assuming that a mix a upper/lower case is not very probable and only accepting it when already present in TM/target segments/TB)?