Hi,
I have noticed that in the last few days ELAN “cuts” part of the target annotation when making searches in multiple eaf files and having them displayed in concordance view. This wouldn’t happen in the past. I thought I could be the font size, the context size or also my display resolution but changing these parameters hasn’t solved it.
I attach an example in which the segment umalang ten minati is reduced to …alang ten min…
Is there any solution to this? Thank you
Hi,
That looks strange; if I’m correct the text is sometimes even cut off in the middle of the target annotation? (Or are “kaya … alang gul…” on the third row not part of the same annotation?)
And this doesn’t even change with a context size of 0?
The things you already tried would be my suggestions for a possible solution. I’m intrigued by your “in the last few days”: does this mean you didn’t use this functionality for a long time, or did the alignment work well in the same version of ELAN until recently and now not anymore? In the latter case I would say that the reason for the change must be outside of ELAN (system update or setting update?). Even then the question remains what can be done about it.
PS which version of ELAN is this? I can actually see a similar effect in the search window in 5.9, but I don’t see it in 6.2 (or 6.1).
Many thanks for your response, Han. That version was 6.2. I tried uninstalling and reinstalling it and also reduced context size and font to the minimum, though still with the same result (see attached screenshot).
I can’t tell precisely when this started to happen, very roughly I’d say it at started some point in the last couple of months. I wish I could be more helpful with this.
The only workaround I have so far (in order to be able to quickly browse annotations after a search in multiple eaf files) is switching to alignment view. Any other solutions would be appreciated.
Thank you!
Hi, it remains strange then, that this effect occurs in your 6.2 version. I assume you also tried different fonts for the hits (so apart from the font size), with the same result?
Looking at the code, the only reason that I can think of for this effect, is that the graphics system that calculates and reports the width of a string based on the selected font, reports a width that is off by a few pixels. But I’m not sure if this is what actually happens, so I also don’t know at the moment what can be done about this.
I have tried manipulating font and font sizes and in some cases the results are displayed without being cut. I wasn’t able to find a clear pattern though. The search for which I attached the screenshot a few days ago is displayed correctly in Arial at 24 pt.
When using less simple regex, the display varies. I attach a screenshot in which all words starting with the prefix men- are displayed correclty. When trying to display words starting with the prefix me’e- however, some of the letters are substituted by dots (see screenshot two).
I am not sure this helps, but thank you again for looking into this!
