lhchavez

> The indexer is called to parse the packfile as it comes in from the network. As such, there is no pre-existing file to mmap. The reason the mmap windows...

here we see `append_to_pack` only taking ~1.16% of the total execution time (cloning the [git](https://github.com/git/git.git) repository instead since the rust one was _too_ slow):  [source](https://gist.github.com/lhchavez/1bede1a8d73902f73ce4d136c4f0cf9f).

I agree with the spirit of https://dave.cheney.net/2016/04/27/dont-just-check-errors-handle-them-gracefully 's "Assert errors for behaviour, not type" section. but in practice, creating an interface for each category of errors is a bit too...

This is required to make go-langserver work with [prabirshrestha/vim-lsp](https://github.com/prabirshrestha/vim-lsp) when invoked with nonstandard flags.

This could in theory be achieved through the [Pointer Lock API](https://developer.mozilla.org/en-US/docs/Web/API/Pointer_Lock_API) ([demo](https://mdn.github.io/dom-examples/pointer-lock/)). Are patches welcome for this? If so, I can maybe cook something up.

> @lhchavez your timing on working on this is hilarious, because I am in the middle of trying to add game support to [this project](https://github.com/tinyzimmer/kvdi) right now and just jumped...

> Thanks. This would be nice to get working. I do have some reservations though: > > * Cursor isn't shown when this is active. This seems to be how...

> > I couldn't find a way to detect when this was requested by the server :( Otherwise, the [QEMU Pointer Motion Change](https://github.com/rfbproto/rfbproto/blob/master/rfbproto.rst#qemu-pointer-motion-change-pseudo-encoding) might have been a better solution for...

> Right, so that's when we would show something for the user to click. Like how the browsers pop up something saying that this page requires feature X. E.g. a...

> > Let me look into this. (Pointers as to how to achieve that are welcome). > > You probably need to dig around in `core/util.cursor.js`. It already handles touch...