Adam

Takase looked into this at using Harfbuzz at one point, and it was unfortunately significantly more complex than first expected. That said, it is something we could look more deeply...

Odd. This seems erroneous in general; I don't think you should be able to have an "empty" split view; it should revert to a single view outright. So yeah we'll...

As this is in master, I'm going to close this.

Hrm. Yeah. We should probably avoid changing process state, and just use internal state in the application.

Ah yeah; that's a MinGW header. We can look into seeing what might make more sense there.

Hrm. Interesting. I'll look at it, and see if I can reproduce on Linux, and then, on Windows. > I suspect that we could have a lot of regressions in...

OK; makes sense, the windows implementation is significantly different from those two (but should actually be more straightforward). I'll go and fix it.

Aha; found the issue. Seems like if your buffer is too small, windows doesn't like that, and refuses to give events individually. dmon's buffer size is `64512`, so I'll just...

OK; fixed that, and a few other minor details. Seems to be working now. I'm also going to do a bit more testing with scanning; I noticed that there was...

> The function CancelIoEx is not declared apparently so I had to add to declare explicitly: This is odd. I am also on the MinGW 64-bit toolchain, and that declaration...