Start a new topic

Suggestion on how to set preference for Japanese glossaries

In the sentence below, the red part is in a glossary and CTE finds it and displays it no problem:


よって本選考委員会は頭書の賞を贈ると共にその栄誉を称えここに表彰いたします


The next sentence consists of the same red part only:


選考委員長


However CTE doesn't find it in the glossary. Why? Is there a particular setting to change? In Preferences > Glossary,  "Prefix matching" and "Look up word stem" are selected and "Minimal prefix length" is set to "3".


The Glossary "Prefix matching" option hasn't affected the Japanese entries before. However, you can try it out in the recent CTE 10.2 Crossaint update where it has been made active for this language too. Thanks!  

OK, but why with a cell containing only 選考委員長 (which is in the glossary) CTE is unable to find it?

Because the red part in the second source segment is not exactly the same as the red part in your glossary - see the last red character is different in your first post.

Oh, what a shame! Fortunately I got it right in the translation. Sometimes it's easy to mix up Japanese logograms. Deep apologies indeed and I will shut up for one month!

Again on this topic: When translating from Japanese I set glossary minimal prefix length to "3" (I wish I could set it to "2" because many terms consist of two characters only, but this would show too many fuzzies).


But, why with this setting 2-character hits too are shown in the glossary pane?


Thank you for any suggestion, if any.

They are unaffected because CT does not 'touch' them if such glossary entries are shorter than the predefined minimal prefix length (e.g 3 characters as in your example). In other words, two-character glossary entries present in your glossary should be matched. 

But, even when the glossary minimal prefix length is set to "3", I can still see 2-character entries in the glossary pane.

Even 1-character only.

The 3-character prefix length means that your glossary entries longer than three character are matched based on their prefixes. However, the shorter  glossary entries of two- or one-character length are unaffected and still can be found in the given segment. That's why you can still see them.

OK, now I understood. But, being Japanese (that is, without separating spaces), doing so makes irrelevant hits showing in the glossary pane, thus wasting precious space. Would it technically possible to show only the hits whose number of characters matches the minimal prefix length count onward?

Then, it would make all the 2- or 1-character entries hide, even the correct ones.

Then, it would make all the 2- or 1-character entries hide, even the correct ones.


I think that, with Japanese and ant other languages without word separators a trade off should accepted some how.


Or, why not using percentage values instead of number of characters?

Login to post a comment