Charles Kerr

> @ckerr Is this PR still desired if the author resolves the merge conflicts? Yes IMO this should get landed if the merge conflicts are resolved before 4.1.0. It's kind...

This isn't a 4897 change. Most likely candidate is #4877, _possibly_ 4909. @GaryElshaw completely remove the build directory and start a build from scratch. If that solves it, go ahead...

> Similarly, on Windows 10. > > ``` > C:/Users/Gary/Documents/GitHub/transmission/libtransmission/tr-macros.h:60:40: note: in definition of macro 'TR_LIKELY' > 60 | #define TR_LIKELY(x) __builtin_expect((x) ? 1L : 0L, 1L) > | ^...

> I'm completely mystified why this has been in limbo for so long. I've been sleeping on it because of the `needs ui review` label. :man_shrugging: Looks like @mikedld did...

I see you've already pinged Mike in that ticket, so I'll defer to him.

I like the idea of moving some the setup work out of `Application`, e.g. `Application` needs to know about `Prefs` but doesn't need to know about `tr_getopt`. We could have...

Notes: Improved error logging.

> but I'm getting different hashes between `main` and the content in this PR, so not sure it's a proper method. Interesting! I _believe_ these compressors are lossless but I...

> but I'm getting different hashes between main and the content in this PR, so not sure it's a proper method. I still have some work to do on this,...

I never did get back to this, and it's been 9 months