Tech @ languidnights - config

KDE Applications under Sway

2022-07-11T00:00:00-04:00

Problem Statement
What doesn't work
What does work

Problem Statement

I'm using Sway as my "window manager" in my daily workflow. I happen to like Wayland and am used to tiling window managers, so it's a good fit. I also think the KDE project makes excellent software, which I like using in my day-to-day computing. There, however, is not a well-documented way to make qt or KDE applications use their KDE style outside of KDE itself.

What doesn't work

qt5ct (or qt6ct). Don't get me wrong, they both do what they advertise, but they don't really let you use your beautiful breeze(-dark) theme in non-KDE environments.
LXQT. My favorite desktop environment has its own styling/themeing support which is not the same as KDE's. It allows you to create quite beautiful qt setups, but if I want dolphin or okular to follow breeze(-dark), I'm out of luck. This is what I used for a long time to work around my lack of KDE style.
XDG_DESKTOP_ENVIRONMENT=kde. Ooo, now we're getting close! This technically does what I asked it to do, but it has unfortunate side effects. Various things start to break as xdg-desktop-portals get confused as to which variants they should be using. As I'm using Sway, this needs to be XDG_DESKTOP_ENVIRONMENT=wlr since the compositor and portal base is not kwin but rather wlroots.

What does work

I can not find this referenced anywhere with my meagre ddg-foo, so I'm setting it here for future wanderers to find. After exhausting what I could find online, I went back to the qt5ct setup. I wondered, is it possible to set QT_QPA_PLATFORMTHEME to something else sensible? Yes! It just so happens you can. I discovered, by trial-and-erroring my way though it (breeze, Breeze, Fusion, and plasma all fail), that you can set it to kde, allowing your qt and KDE apps to follow the themeing you set up in ksytemsettings. Lovely good time!

While it wasn't my answer, I couldn't have found it without qt5ct and the reddit community's guides.

Here's hoping someone else can pull out a little less of their hair than did I. I didn't have enough to start with ;-)

libvirt lessons

2022-03-19T00:00:00-04:00

Contents

Background
Automagic Configs
- networking
- qemu command line
Storage
Video Acceleration
FreeDOS Specific

Background

On my powerhouse machine, I use virtual machines to contain experiments and code I don't trust into little self-contained sandboxes. I also separate some concerns into their own machines. Here, we'll go over things I've learned in the course of my setups.

First off, I'm using linux, so my base layer is qemu + kvm. The perfect tool to manage your virtualization is libvirt. For my graphical management, I'm using the absolutely amazing virt-manager.

Automagic Configs

There are a few things that, if you use libvirt and virt-manager, you no longer have to worry about configuring. Advancements is the tooling and automation have come far since most tutorials.

networking

A lot of documentation asks you to create a bridge interface and add it to your /etc/network/interfaces or create a systemd service for it. This is no longer necessary. Simply right-click your qemu+kvm connection details, navigate to the Virtual Networks tab, and click the run button at the bottom. Boom! Network setup. For the individual users, this is all you'll have cause to user; you're not an enterprise sysadmin, you don't need to learn networking managements details for a simple project.

qemu command line

I don't know about you, but I never could type the qemu command line properly, so I ended up writing a script and triple checking where I'd modified an example to do the setup, and still get it wrong one time out of three. This is messy and error-prone, and kind of silly to still be doing manually. Luckily, if you use virt-manager, it'll write out a pretty standard startup and connection string for you, as well as the libvirt xml to match it. I'm a software engineer by training, not an enterprise sysadmin, I'm really grateful that I don't have to anymore.

Storage

Almost any virtual machine needs a backing store. The default way for virt-manager is to use a qcow2 file in /var/lib/libvirt/images/. I am already using lvm (see here) for managing my host system, so may well use it for virtual systems too. This is a fairly easy setup if you've already set up your volume group. In the 'add storage' dialog, the left pane in your storage pools. Click the add button on the lower left. Then, under the Type dropdown, select logical: LVM Volume Group. Your volume groups should be populated in the Volgroup Name: dropdown. Select the proper volume group, and you can then create a new logical volume using default settings (no need to call lvcreate --size xx --name volume_name volume_group_name yourself).

Video Acceleration

In the before times, to get gpu acceleration in a virtual machine you had to pass through an unused phsycial graphics card. For Windows VMs, you probably still have to. These days, if your host and guest graphics stack uses Mesa, you have the ability to use the hosts gpu acceleration on the guest's window. I originally had to build various bits myself, but after I came across Ryan Himmelwright's blog I discovered the stuff was already built in. Just set the graphics type to virtio, tick the 3d Acceleration box, and set your Display to Spice and tick OpenGL. Yes, these two settings will complain until they match. That's fine, just match them :-)

FreeDOS Specific

I installed a working FreeDOS install. Everything worked out of the box, except for the audio device. I manually changed the audio setting to sb16 by manually editing the xml snippet to sb16.

I then had to add a soundblaster driver in FreeDOS itself. I got the Win3.11.zip archive from this URL from this YouTube video (warning, possible audio issues). I pulled the soundblaster directory and passed it through to FreeDOS, then ran the executable and it set up autoexec.bat and everything else for me. (Don't be surprised if that link goes away, the video author owns it, and it appears to be a personal website). The oemdrivers site may be a good backup. Now I can hear that glorious (?) crunchy sound of Doomguy shooting his pistol.

Home Directory in Git

2022-03-18T00:00:00-04:00

Keeping my $HOME in git.

Table of Contents

Living in Git, or Why are we Here?
Other Implementations
Conceptual Design
Implementation
I want it!

Living in Git, or Why are we Here?

Over time I've accumulated a massive amount of built-up configurations and other cruft in my home directory, and whenever I need to bring a new computer online I have to copy my desktop configuration to it, like a game of software config telephone. (this happens to me more often that you'd think, as I have a number of Raspberry Pi boards I use as desktops until I have a need for it in an embedded solution). I figured there had to be a better way, using the tools I picked up as a developer. Turns out there is :-)

I remembered A Long Time Ago™ reading about Joey Hess writing etckeeper to keep the configuration files in /etc/ under, ahem, control. I also remember vaguely seeing similar software for home directories, but those software projects are lost to years of imperfect memory, and so I couldn't find them. Hence, my writing a new one.

Other Implementations

There are probably many, many, many implementations to keeping your home directory in git. Most of them would probably even work for my personal setup, but NIH and all. So I wrote my own! I'm actually kind of proud of it, but if the gods are kind it stays pretty much the way it is for the foreseeable future, as I've settled somewhat in my 'config' churn and am running Debian Stable anyway so my software shouldn't be changing its config syntax weekly.

Conceptual Design

My requirements are the following:

the configuration files should be in version control
the configuration files must be able to be kept separate from the live config files
the script must be able to be executed on any system I might call home (this means POSIX compliant shells, such as DASH, or perl)
ideally, allows symlinking so as to make changes easily pushed upstream

Implementation

To achieve this, I went with git for version control. This allows me to utilize the git setup framework I already had installed for my software projects (it's here if you're interested). It's also quite well documented and recipes exist for almost every scenario on either Server Fault or reddit. I initialized a new git repository on my server for this to live in, calling it l-homedir after a bit of self-serving names :-)

In my shiny new git repo copy, I moved (as in mv) the interesting and non-private configurations into it. I maintained the directory structure you would find in my home directory (for example, my .zshrc file is in $REPO_ROOT and my nvim config lives in $REPO_ROOT/.config/nvim). This allows me to separate the files I care about from those that I do not care if get lost, or those that contain private information (ssh private keys and such).

To accomplish the copying/linking files (linking in this case) I wrote a mildly complicated shell file, and a corresponding config file. linked_files contains a list of files I want to maintain here---in case something accidentally ends up in the repo. make_links is the shell script. The script follows this basic flow

iterate over each line in linked_files
if it is blank or starts with a hash (#) ignore it
check if the file already exists?
if the file already exists, is it running in destructive mode?
if in destructive mode, rm the original file, otherwise skip it
if skipped, return to Loop Top
ln -s the file in the repository to the home directory
Return to Loop Top

I wrote a README and brought in a license file to top it off, and I have a shiny new home-in-git solution :-D

I want it!

Assuming you're not trolling me, get it right here