Re: Commit 4f3a8eba3a doesn't compile

[Charlie Jiang]

> Interesting.  Maybe a buglet in Qt then?  My old openSUSE distribution
> uses version 5.12.7...  Within the next weeks I have to eventually
> upgrade my laptop to a new openSUSE version; will then test again.
>     
    Mine Kubuntu 22.04 uses 5.15.3. Tests on a newer version would be
    great.
> Again, no pressure!  You have a lot of time to work on that :-)
>

It's relieveing to hear that :D. However, what do you prefer for me to prioritize?

• The Hinting Comperator Tab (= `ftdiff`), a new tab parallel to singular grid view and continuous view
  
• The long, long list of TODO features, and the most wanted features seem to be the zoom/scroll/drag operations and click-for-detail features for continuous view. It seems that Alexei is more into this.
  

Personally both are OK to me, so I'd like to hear your opinion.

> However, if I press 'continue' in gdb, it stops after one second again
> at this place.  And again, and again...  Is this normal?  How shall I
> ever test a change?
>

Very abnormal. Is your `msyh.ttc` a symbolic link? It's pretty evident that it's FS watcher part triggering font reload operation, thus resetting the charmap option. Because the file was recognized as a symbolic link, and Qt's fs watcher can't handle this (as is mentioned in the comment), a timer is used to periodically reload the font instead of listening on fs events.

However, reloading of font should not reset the charmap option either if the charmaps didn't change during the reloading (e.g. when the font isn't changed actually). Line 116, 120~124 and 144~145 are exactly intended to avoid unexpected resetting, but it don't seem to work. I will look this tomorrow (7 hrs later) :P.

Cheers,
Charlie Jiang