Page 1 of 1
Diacritics bizzare behaviour
Posted: 2021-07-11 22:32:40
by mfrance
If I type U+006D U+0325 (latin letter m with combining ring below), U+006E U+0325 (latin letter n with combining ring below), U+006D U+0310 (latin letter m with combining candrabindu), save the file, close and reopen it, what I get is one or more base character without diacritic followed by the base character correctly combined with the diacritic. In other word, what I get is, respectively: mm̥, nn̥, mm̐ (sometimes mmmm̥, nnnn̥, mmmm̐). Other word processors (e.g. Word) do not have this issue. Any suggestion?
Many thanks in advance,
Marco
Re: Diacritics bizzare behaviour
Posted: 2021-07-12 03:01:35
by Hamid
I can reproduce the bug.
Everytime the file is reopened, there is an extra m before m̥ and an extra n before n̥.
If I copy m̥ and paste it, I get mm̥; same with n̥ which turns to nn̥.
Re: Diacritics bizzare behaviour
Posted: 2021-07-12 03:54:31
by adryan
G'day, Marco, Hamid et al
I was not able to reproduce this behavior.
I copied the characters (one at a time) from the Forum and pasted them into a new Nisus document before saving and closing the document. Reopening the document showed no additional characters. Same behavior whether I did a plain Paste or a Paste Text Only. Formatting Examiner showed the correct character codes. I might add that in the Displayed Formatting section of the Formatting Examiner the Language is English (U.S.).
I wonder whether the aberrant behavior you are experiencing is a result of using a different operating system from me.
Cheers,
Adrian
Re: Diacritics bizzare behaviour
Posted: 2021-07-12 04:31:39
by Hamid
I am on macOS 11.4 with NWP 3.2.1.
Re: Diacritics bizzare behaviour
Posted: 2021-07-12 04:32:01
by xiamenese
I too can confirm it happening. M1 MacBook Air, MacOS 11.4. My iMac is now running Big Sur too, so I can no longer check under 10.15.
Mark
Re: Diacritics bizzare behaviour
Posted: 2021-07-12 08:15:39
by martin
This appears to be an issue that's new to macOS Big Sur, at least for some users of Nisus Writer Pro. It does not always occur. We're not sure what exactly triggers the bug on some systems but not others. The fonts installed on your system may be a factor, but that's not conclusive.
We're looking into getting this problem fixed. If you're interested in receiving early access to a private beta version that fixes the problem please contact us at:
Sorry for the trouble! If you have documents that have incorrectly doubled characters you can hopefully revert a clean copy of your file using the
document versions feature.
Re: Diacritics bizzare behaviour
Posted: 2021-07-12 08:44:35
by Þorvarður
adryan wrote: ↑2021-07-12 03:54:31
I was not able to reproduce this behavior.
Same here. Reopening the document shows no additional characters.
Must have something to do with the new operating system(s).
macOS High Sierra 10.13.6
MacBook Pro (mid-2010)
Nisus Writer Pro 3.2.1
Re: Diacritics bizarre behaviour
Posted: 2021-07-17 07:26:39
by mfrance
martin wrote: ↑2021-07-12 08:15:39
This appears to be an issue that's new to macOS Big Sur, at least for some users of Nisus Writer Pro. It does not always occur. We're not sure what exactly triggers the bug on some systems but not others. The fonts installed on your system may be a factor, but that's not conclusive.
We're looking into getting this problem fixed. If you're interested in receiving early access to a private beta version that fixes the problem please contact us at:
Sorry for the trouble! If you have documents that have incorrectly doubled characters you can hopefully revert a clean copy of your file using the
document versions feature.
I am running Nisus Writer Pro 3.2.1 on Apple M1, macOS Big Sur 11.4.
In my case, the issue involves the base characters (n, m, r, l,) combined with U+0325 (combining ring below) and m plus U+0310 (combining candrabindu).
It also involves base characters (h, m, t, d, n, s) plus U+0323 (combining dot below), but
only if they were already there in an "old" file, i.e.: the bug does not show up if I type a base character plus U+0323 now.
I teach Sanskrit, I have thousands of files with such diacritics and I can't dare to open none of them, otherwise everything is messed up, ergo: I am totally stuck. Please, have pity on me! Repair the bug!
Marco
Re: Diacritics bizzare behaviour
Posted: 2021-07-17 08:24:19
by adryan
G'day, Marco et al
It looks as though Martin is working on a fix for you. If you can't wait, it might be worth seeing whether Find & Replace can correct the situation for you. Obviously, you would test this on a copy of a file rather than on the original. You would probably have to do the replacement (perhaps via a Macro) each time you opened the file, but it might at least give you a workaround. I can't test this approach myself, as I have a different setup.
Cheers,
Adrian
Re: Diacritics bizzare behaviour
Posted: 2021-07-19 08:18:41
by martin
adryan wrote: ↑2021-07-17 08:24:19
It looks as though Martin is working on a fix for you. If you can't wait, it might be worth seeing whether Find & Replace can correct the situation for you.
We are indeed working on a fix for affected users. I'm sorry for the heartache in the meantime. As I mentioned earlier, please contact us if you'd like early access to a private beta version that remedies this problem.
Adryan's suggestion to use Find & Replace can help. Here's a basic macro (which I'll also attach) that uses Replace All to fix duplicated letters that occur before a combining mark:
Code: Select all
Find and Replace @Text<(\X)\1+(\p{CombiningMark})>, @Text<\1\2>, 'Ea'
A few caveats:
1. The macro can't recognize situations where repeated characters are intentional. If your document has any such text then you'd need to be careful.
2. The incorrectly duplicated letters will reappear after saving and reopening your file.
Re: Diacritics bizzare behaviour
Posted: 2021-07-19 11:10:18
by mfrance
Many thanks to Adrian and Martin for their replies. I am running a polishing macro (similar to the one kindly suggested by Martin) each time a open a file. It works, although it becomes a somewhat cumbersome procedure in the long run.
Best,
Marco
Re: Diacritics bizzare behaviour
Posted: 2021-07-19 13:30:36
by adryan
G'day, Marco et al
A couple of other thoughts….
Depending on the number of possible rogue character combinations, it may not be too onerous to incorporate them into the spell checker system and thus have the corrections attended to automatically.
If you have a file that you simply want to view frequently, without further editing, you could correct the aberrations once and then save the file as a PDF document. This will freeze the corrections in place and allow the file to be viewed, printed and distributed without further drama.
Cheers,
Adrian
Re: Diacritics bizzare behaviour
Posted: 2021-09-30 09:24:41
by martin
This issue has been resolved in Nisus Writer Pro version 3.2.2. That software update is available everywhere: inside the app, the Mac App Store, and
direct download.