Hu Jialun

I am also running into this issue as well trying to use the [revealjs](https://github.com/rstudio/revealjs) package which is a Rmd wrapper around the [reveal.js](https://revealjs.com) presentation framework. The output is rendered as...

Hi @ElianHugh, Thanks for the reply. I tried to diff the `pandoc` command lines and here is what I think makes the difference, ```diff --to -html4 +revealjs --from -markdown+autolink_bare_uris+tex_math_single_backslash +markdown+autolink_bare_uris+tex_math_single_backslash-implicit_figures...

The relevant snippet seems to be https://github.com/rstudio/revealjs/blob/e1893aed9df79d26db48538ace56cc0d55ae2f0c/R/revealjs_presentation.R#L319. Is it that the `pandoc` options not respected for `render_site`?

This was fixed upstream in https://github.com/stackgl/headless-gl/pull/243, but gpu.js is stuck at https://github.com/gpujs/gpu.js/blob/6d7dde32679f56f3dc99013cfdb5aa8accc159da/package.json#L22. I assume it is non-trivial to move to a new major version of `gl`.

I wonder how this additional parameter can be implemented if the D-Bus API just cannot stand this without disconnecting. Are we changing to a native backend like bluepy does?

@chrisgzf maybe it is feasible to reevaluate & take in some user inputs on this ER now that it is ModReg time again? 😸 > this feature is only useful...

That looks really amazing! Looking forward to the code release :)

@tmbinc Current `HEAD` (3de438b) still does not work with my Nexus 4 DDR board with exactly the same symptoms, and the patch fixes it. ```diff --- a/src/io_ftdi.c +++ b/src/io_ftdi.c @@...

FWIW `iso` build artifact attempts (and of course fails) to mount host `hardware-configuration.nix` partition UUIDs.

To avoid the `readlink` error, `touch aur.db.tar.gz` and soft link `aur.db` to it instead. According to [Arch wiki](https://wiki.archlinux.org/title/Pacman/Tips_and_tricks#Custom_local_repository), > A package database is a tar file, optionally compressed. Valid extensions...