$ /snap/clion/473/bin/clion --version
CLion 2026.1.4
Build #CL-261.26222.59

$ /snap/clion/473/bin/cmake/linux/x64/bin/cmake --version
cmake version 4.2.2

CMake suite maintained and supported by Kitware (kitware.com/cmake).

$ /snap/clion/479/bin/clion --version
CLion 2026.2
Build #CL-262.8665.262


$ /snap/clion/479/bin/cmake/linux/x64/bin/cmake --version
cmake version 4.3.1

CMake suite maintained and supported by Kitware (kitware.com/cmake).

Thank you for breaking my builds ig

The solution is to edit the config files directly. For a Flatpack installation:

# access the flatpak shell
flatpak run --command=bash im.nheko.Nheko  

echo $XDG_CONFIG_HOME

The output will be in the form $HOME/.var/app/im.nheko.Nheko/config, navigate to it and open the nheko/nheko.conf file.

emoji_font_family=default
...
font_family=default

Restore the font family settings to default.

The latter is just a python script provided by the reserialize package with lot of bugs.

On later version I suggest rust-toml2json via apt.

My morally neutral research!

Source

Unfortunately the official docs don’t know which is your target architecture and resolve it by providing a “one size fits all” solution, with the expected drawbacks in storage.

In this guide I’ll demonstrate how to create a T2 topology Zephyr workspace for an RP2040 board on Ubuntu. It can be extended to other boards if necessary.

Dependencies

Source

sudo apt update
sudo apt upgrade
sudo apt-get install --no-install-recommends git cmake ninja-build gperf \
  ccache dfu-util device-tree-compiler wget \
  python3-dev python3-pip python3-setuptools python3-tk python3-wheel xz-utils file \
  make gcc gcc-multilib g++-multilib libsdl2-dev libmagic1 \
  libusb-1.0-0-dev   

Install Zephyr SDK

Source

ℹ️ Note

This targets zephyr-sdk-1.0.1, change it according to your needs.

Download

cd ~
wget https://github.com/zephyrproject-rtos/sdk-ng/releases/download/v1.0.1/zephyr-sdk-1.0.1_linux-x86_64_gnu.tar.xz

Extract

tar xvf zephyr-sdk-1.0.1_linux-x86_64_gnu.tar.xz

Install

Copy the extracted files to the install directory,

ℹ️ Note

This targets /usr/local

Change it depending on your needs to one of the suggested locations.

sudo mv zephyr-sdk-1.0.1 /usr/local

Register the package

cd /usr/local/zephyr-sdk-1.0.1
sudo ./setup.sh

Follow the instructions.

Create the workspace

ℹ️ Note

I will assume the workspace is in $HOME/projects/zephyr-ws, change it according to your preferences.

Navigate your working directory to the workspace

cd $HOME/projects/zephyr-ws

Initialize and activate the virtual environment

python3 -m venv .venv

Instantiate the virtual environment**

source .venv/bin/activate

! Important

Repeat this operation every time you open a new shell. To make sure everything is configured correctly check the (.venv) symbol at the start of your shell.

Install west

pip install west

Create manifest

Create a folder for your manifest

mkdir manifest-repo

Inside the manifest-repo folder create a west.yml with the following content:

manifest:
  remotes:
    - name: zephyrproject
      url-base: https://github.com/zephyrproject-rtos
  projects:
    - name: zephyr
      repo-path: zephyr
      remote: zephyrproject
      revision: "v4.4-branch"
      clone-depth: 1
      import:
        path-prefix: external
        name-allowlist:
          - cmsis_6
          - picolibc
          - hal_rpi_pico

You can select different modules depending on your board, check the official modules list (or alternativelly).

In general try to have at least the HAL for your board.

ℹ️ Note

This example targets Zephyr 4.4, change it depending on your needs

Initialize the workspace

Back in the workspace folder:

west init

This will create a .west folder, inside edit the config file to:

[manifest]
path = manifest-repo
file = west.yml

This will pull all the dependencies of your project

west update
west packages pip --install

First project

You can now test the installation by compiling an example project:

Copy the blinky example from zephyr_ws/external/samples/basic/blinky

cp zephyr_ws/external/samples/basic/blinky .
cd blinky

Compile for your board

west build -b=rpi_pico

If it compiles successfully then you’re done.

To start a new project just create a new folder in zephyr_ws/ with the necessary file structure.

You may experience it when using a laptop with a metal chassis: if you use it long enough and then you touch some metal objects connected to ground you’ll get zapped by the static electricity.

For soldering irons this means that your soldering tip will have a voltage differential of around 90V to ground. The current will be really small because the secondary/DC circuit is protected by a small capacitor, so it’s technically safe for humans. You can’t say the same to fragile electronic components, which will sometime fail due to the ESD. One simple precaution is to use a power bank when soldering delicate components.

There is another solution:

The doohickey

(Remember to use the name of the manufacturer if you don’t wanna end up reading about soviet tanks)¹

This is an option ig, with people really going through this route ²³. Other suggestions are the Apple USB-C chargers with a 3 pin plugs.

Causes

For more information on why this happens Wikipedia has a nice reference.

Demonstration

Video demonstrations, found thanks to this reddit comment.

A few years ago I mentioned staticman dot net in a blog post.

Some time ago the domain expired and it was squatted by someone (allegedly) illegally streaming NBA matches.

The NBA decided to send a takedown notice to everyone linking to that domain.

Other people complaining

You need to push some artifacts from a Github Actions runner into a folder in a server where you only have SSH enabled.

Create a new user

This is a basic safety measure. It limits the directories the SSH/SCP connection can access.

In this case we will create a user called dummy

sudo adduser dummy
# follow the commands       

Do not add it to the sudo group.

Test the new user account:

su dummy
# try the new password
cd   

Generate an SSH keypair

ssh-keygen
# change the defaults if necessary but keep it without a keyphrase

You’ll end up with two new files in ~/.ssh:

• id_{algorithm}: private key

• id_{algorithm}.pub: public key

For now add the public key to the authorized_keys file

cat id_{algorithm}.pub >> authorized_keys

Github actions

You can now add them to your repository secrets.

1. Horizontal navigation bar: “Settings”

2. Left Navigation bar: “Security” -> “Secrete and variables” -> “Actions”

3. “Secrets” -> “Repository secrets” -> “New Repository Secret”

4. Fill:

   • Name: SSH_PRIVATE_KEY
   • Secret: The content of id_{algorithm} obtained from before.
     bash

     cd ~/.ssh
     cat id_{algorithm}

5. “Add Secret”

In your YAML recipe:

- name: load priv key
  run: |
    mkdir $HOME/.ssh/
    touch $HOME/.ssh/known_hosts 
    ssh-keyscan -H "{ip_address_of_your_server}" > $HOME/.ssh/known_hosts 
    echo "${{ secrets.SSH_PRIVATE_KEY }}" > $HOME/.ssh/id_{algorithm} 
    chmod 400 $HOME/.ssh/id_{algorithm}

- name: do work
  run: rm -rf /
  
- name: upload to vps
  run: scp -i $HOME/.ssh/id_{algorithm} -r {artifact} dummy@${ ip_address_of_your_server }:{destination_path}

Remenber to change id_{algorithm} and {ip_address_of_your_server}

All convex sets are alike; each non-convex set is NP-Hard in its own way.

Leo Tolstoy

Source

Create a polygon layer

Open QGIS.

In the left side bar, right-click “SpatiaLite” and select “Create Database…”

You will be asked to save the database in a .sqlite file somewhere on your disk. In my case I called it tutorial.sqlite.

Right-click on the generated database and select “New Table…”

Create the following schema: []

Save with “OK”

Double click on the newly created table to add it to your layers.

Add other layers below (eg: OpenStreetMap, orthophoto providers, etc…).

In the top menu, go to “View” -> “Toolbars” and enable “Advanced Digitizing Toolbar”

It should look like:

How to draw

In the layer panel, select the project layer created previously and then press on the pencil icon in the toolbar

Now you can add your outer perimeter by pressing the “Add Polygon Feature” icon.

The polygon will obscure everything underneath making it hard to see the obstacles. We can fix this by changing the opacity of the layer.

Right click on the layer and select “Properties”

In the new window, use the navigation panel to select “Symbology”.

Open the “Later Rendering” menu at the bottom and change the opacity. Save with “OK”.

In the toolbars, press the “Select Feature” and click on your previosly created area. It will change color, usually into yellow.

Now we can add the obstacles: select the “Add Ring” from the toolbar and trace the shape of the obstacles.

After being done with it: exit the edit mode by pressing the pencil icon again and save your obect database by pressing the Floppy Disk icon 💾.

To export your file:

Right click on the object layer and go to “Export” -> “Save Feature As”

In the new window:

• Select GeoJson as the format
• Pick a destination for the exported file
• De-select the optional integer field
• De-select “Persist Layer Metadata”

Now that you have a “.json” file you can import in F2C with

F2CFields fields;
f2c::Parser::importJson(std::string(DATA_PATH) + "tutorial.geojson", fields);
 

Full example

Final result:

[] []

https://github.com/e271828e/Flight.jl

There are probably some fixes around ( --stop-screensaver maybe?) but it’s funny how XKCD 196 keeps haunting Linux

[STATUS: WONTFIX, Minor inconvenience] Source

They are all rebranded Hung Chang oscilloscopes, especially the “TELAV 09635” (a rebranded “Hung Chang OS-635”) and the “TELAY 09620” and “AWS MODEL 620C”² (both rebrands of the “Hung Chang OS-620”).

Source

The trick is to place a driver inside the fan capable of obtaining the necessary voltage/current.

Source

This exonerates the MoBo engineers from accounting for high current requirements when designing the PWM controller and avoids wasting energy driving up and down a potentially high current through the connectors and cables.

https://janismac.github.io/ControlChallenges/

https://quanserinc.box.com/shared/static/lhf0gilgy1zu5hedkokb9mi025gzcu19.zip

Source

# By default the ssh-agent service is disabled. Configure it to start automatically.
# Make sure you're running as an Administrator.
Get-Service ssh-agent | Set-Service -StartupType Automatic

# Start the service
Start-Service ssh-agent

# This should return a status of Running
Get-Service ssh-agent

# Now load your key files into ssh-agent
ssh-add $env:USERPROFILE\.ssh\id_ecdsa

Source

sudo raspi-config nonint enable_overlayfs

reboot

To disable it:

sudo raspi-config nonint disable_overlayfs

reboot

mkdir pdfout
soffice --headless --convert-to pdf --outdir ./pdfout *.doc

It also works with .docx, .ppt, .pptx etc

is a skew-symmetric matrix (i.e. ) so we have

With no demonstrations whatsoever.

[] []

Let’s demonstrate it once and for all.

Proof

With a scalar. Being a scalar we have:

2D Check

Fun fact: Hugo supports downloading JSON, CSV, XML, Images and is also capable of processing them at render time.

Movies

Like the original blog I also used letterboxd becuase it’s free. (Usually I use Trakt + Showly but it requires a subscription). We create a simple shortcode and then call it from the markdown.

{{< letterboxd >}}

{{ $data := dict }}
{{ $url := "https://letterboxd.com/yourusername/rss/" }}
{{ with resources.GetRemote $url }}
  {{ with .Err }}
    {{ errorf "%s" . }}
  {{ else }}
    {{ $data = . | transform.Unmarshal }}
  {{ end }}
{{ else }}
  {{ errorf "Unable to get remote resource %q" $url }}
{{ end }}

{{ with $data.channel.item }}
  <ul class="consumed-list">
    {{ range first 3 . }}
        {{ $title := .filmTitle }}
        {{ $year := .filmYear }}
        {{ $posterUrl := index (findRE `(https://.+\.jpg)` .description ) 0 }}
      <li class="consumed-element">
        <a href="https://www.themoviedb.org/movie/{{ .movieId }}" >
        {{ with $thumbnail := resources.GetRemote $posterUrl }}
        {{ with .Err }}
            {{ warnf "%s" . }}
            <img src="$posterUrl" 
            width="100" 
            loading="lazy" 
            alt="Poster of {{ $title }}">
        {{ else }} 
            {{ $filter := images.Process "resize 100x q90 webp" }}
            {{ with . | images.Filter $filter }}
            <img src="{{ .RelPermalink }}" 
            width="{{ .Width }}" 
            height="{{ .Height }}" 
            alt="Poster Image of {{ $title }}" 
            loading="lazy">
            {{ end }}
        {{end}}
        {{end}}
        </a>
        <p><a href="https://www.themoviedb.org/movie/{{ .movieId }}" >{{ $title }}</a> ({{$year}})</p>
    </li>
    {{ end }}
  </ul>
{{ end }}


{{/*  <pre>{{ debug.Dump $data }}</pre>  */}}

This will automatically download and parse the RSS, download the posters from TMDB and optimize them.

Books

This time instead I used OpenLibrary because it has a slightly more complete collection.

{{< openlibrary "currently-reading" >}}
{{< openlibrary "already-read" >}}

{{ $type := (index .Params 0) }}
{{/*  types: already-read currently-reading want-to-read https://openlibrary.org/dev/docs/api/mybooks  */}}
{{ $data := dict }}
{{ $url := printf "https://openlibrary.org/people/mekBot/books/%s.json"  $type }}
{{ with resources.GetRemote $url }}
  {{ with .Err }}
    {{ errorf "%s" . }}
  {{ else }}
    {{ $data = . | transform.Unmarshal }}
  {{ end }}
{{ else }}
  {{ errorf "Unable to get remote resource %q" $url }}
{{ end }}

{{/*  {{ $data.reading_log_entries }}  */}}

{{ with $data.reading_log_entries }}
<ul class="consumed-list">
    {{ range first 3 . }}
      {{ $date := index . "logged_date" }}
      {{ $title := index . "work" "title" }}
      {{ $url := index . "work" "key" }}
      {{ $year := index . "work" "first_publish_year" }}
      {{ $cover :=  index . "work" "cover_id" | string}}
      {{ $thumbnailUrl := printf "https://covers.openlibrary.org/b/id/%s-L.jpg" $cover }}
      {{ $authors := index . "work" "author_names" }}
      <li class="consumed-element">
        <a href="https://openlibrary.org{{ $url }}">
        {{ with $thumbnail := resources.GetRemote $thumbnailUrl }}
        {{ with .Err }}
            {{ warnf "%s" . }}
            <img 
            src="https://covers.openlibrary.org/b/id/{{ string  $cover }}-S.jpg" 
            width="100" 
            loading="lazy" 
            alt="Cover Image of {{$title}}">
        {{ else }} 
            {{ $filter := images.Process "resize 100x q90 webp" }}
            {{ with . | images.Filter $filter }}
                <img src="{{ .RelPermalink }}" 
                width="{{ .Width }}" 
                height="{{ .Height }}" 
                alt="Cover Image of {{$title}}" 
                loading="lazy">
            {{ end }}
        {{end}}
        {{end}}
        </a>
        <div>
            <a href="https://openlibrary.org{{ $url }}">{{ $title }}</a> ({{$year}})
            <p>written by <em>{{ delimit $authors ", " " and "}}</em></p>
        </div> 
    </li>

    {{ end }}
  </ul>
{{ end }}

Like before, it downloads the file, parses the JSON, looks up the cover and embeddeds it into the page.

It has some quirks like: the JSON orders the entries based the logged date when it was added in your “general” collection and not when it was moved in a specific collection, so when you change a book from “reading” to “finished reading” it won’t be in the first entry.

Games

I don’t play that many games tbh so I didn’t bother.

TV Shows

{{< serializd "yourusername" >}}

{{ $username := (index .Params 0) }}
{{ $refurl := printf "https://www.serializd.com/user/%s/currently_watching"  $username }}
{{ $requrl := printf "https://www.serializd.com/api/user/%s/currently_watching_page/1?sort_by=date_added_desc"  $username }}
{{ $data := dict }}
{{ $opts := dict
    "method" "GET"
    "body" `{"complete": true}` 
    "headers" (dict  
        "Accept" "application/json, text/plain, */*"
        "Accept-Encoding" "deflate, br, zstd"
        "Accept-Language" "en-US,en;q=0.9"
        "Dnt" "1"
        "Referer" $refurl
        "Sec-Ch-Ua" `"Chromium";v="123", "Not:A-Brand";v="8"`
        "Sec-Ch-Ua-Mobile" "?1"
        "Sec-Ch-Ua-Platform" `"Android"`
        "Sec-Fetch-Dest" "empty"
        "Sec-Fetch-Mode" "cors"
        "Sec-Fetch-Site" "same-origin"
        "user-agent" "Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Mobile Safari/537.36"
        "x-requested-with" "serializd_vercel"
    ) 
}}

{{ with resources.GetRemote $requrl $opts }}
  {{ with .Err }}
  {{debug.Dump $opts}}
  {{debug.Dump .}}
    {{ warnf "line 23 error: %s" . }}
    <p>(Error occured while retrieving the data)</p>
  {{ else }}
    {{ $data =  .Content | transform.Unmarshal }} 
  {{ end }}
{{ else }}
  {{ errorf " Line 28 Unable to get remote resource %q" $url }}
{{ end }}

{{ with $data.items }}
  <ul class="consumed-list">
    {{ range first 3 . }}
        {{ $title := index . "showName" }}
        {{ $showid := index . "showId" | string  }} 
        {{ $filename := index . "bannerImage"  }}
        {{ $posterUrl := printf "https://serializd-tmdb-images.b-cdn.net/t/p/w400%s" $filename }}
      <li class="consumed-element">
        <a href="https://www.serializd.com/show/{{ $showid }}" > 
        {{ with $thumbnail := resources.GetRemote $posterUrl }}
        {{ with .Err }}
            {{ warnf "%s" . }}
            <img src="$posterUrl" width="100" loading="lazy" alt="Poster of {{ $title }}">
        {{ else }} 
            {{ $filter := images.Process "resize 100x q90 webp" }}
            {{ with . | images.Filter $filter }}
                <img src="{{ .RelPermalink }}" width="{{ .Width }}" height="{{ .Height }}" alt="Poster Image of {{ $title }}" loading="lazy">
            {{ end }}
        {{end}}  
        {{end}} 
        </a>
        <p><a href="https://www.serializd.com/show/{{ $showid }}" >{{ $title }}</a></p>
    </li>
    {{ end }}
  </ul>
{{ end }}

Final result

My now page is visible here.

The list includes:

• Github Pages
• Render.com
• Cloudflare Pages
• Vercel
• Netlify

Scope: to install the environment needed for the labs without having to deal with Docker or dual booting in Ubuntu.

Target: PCs running Windows 10 or up supporting virtualization.

Basic Installation

First you need to set-up WSL in your machine. From a Windows command line run:

wsl --update
wsl --install Ubuntu-20.04

And fill the username and password as you please.

Now you should get a linux shell, in other case you can launch it from the start menu (simply search Ubuntu, it should also apper in the recently installed section) or from the dropdown menu of the terminal (my suggested method).

Let’s update some packages as a precaution and then continue.

sudo apt update
sudo apt upgrade -y

n.b: If you’re using the internal PoliMi Wifi you may get some errors, simply switching the Wifi from “Public Network” to “Private Network” should fix it. You may have to change it back once you’re done with this tutorial.

Now we start dealing with the core ROS installation. These commands are taken from the Docker repository

sudo apt-get install -q -y --no-install-recommends dirmngr gnupg2
sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv-keys C1CF6E31E6BADE8868B172B4F42ED6FBAB17C654
sudo  sh -c 'echo "deb http://packages.ros.org/ros/ubuntu focal main" >> /etc/apt/sources.list.d/ros1-latest.list'
sudo apt-get update 
sudo apt-get install -y --no-install-recommends ros-noetic-ros-core=1.5.0-1*
source "/opt/ros/noetic/setup.bash"
echo "source /opt/ros/noetic/setup.bash" >> ~/.bashrc
source ~/.bashrc

These are the commands for the “base” ROS installation taken from here

sudo apt-get install --no-install-recommends -y  build-essential  python3-rosdep  python3-rosinstall   python3-vcstools 
sudo rosdep init 
rosdep update --rosdistro noetic
sudo apt-get update 
sudo apt-get install -y --no-install-recommends ros-noetic-ros-base=1.5.0-1*

Now finally for the “perception” ROS package

sudo apt-get install -y --no-install-recommends   ros-noetic-perception=1.5.0-1* 

ℹ️ Note

Basically in the Dockerfile shared in class what we just did was simply the first section.

(This is not a command!)

dockerfile

# Start from the ROS Noetic perception image
FROM ros:noetic-perception

Now we continue with the other installation commands:

sudo apt-get install -y --no-install-recommends nano \
  vim tmux gedit ros-noetic-rqt* \
  net-tools iproute2 ros-noetic-plotjuggler* \
  ros-noetic-foxglove-bridge ros-noetic-turtlesim \
  ros-noetic-tf*

Update for lesson 7

sudo apt-get install -y --no-install-recommends  \
  ros-noetic-stage* ros-noetic-gmapping \
  ros-noetic-slam-toolbox* ros-noetic-move-base* \
  ros-noetic-amcl ros-noetic-global-planner \
  ros-noetic-dwa-local-planner ros-noetic-navfn \
  ros-noetic-turtlebot3-* ros-noetic-teleop-twist-keyboard \
  ros-noetic-pointcloud-to-laserscan 

You should already be in the home directory but just in case:

mkdir -p ~/catkin_ws/src
cd ~/catkin_ws/
ln -s /opt/ros/noetic/share/catkin/cmake/toplevel.cmake src/CMakeLists.txt
catkin_make
echo "source ~/catkin_ws/devel/setup.bash"  >> ~/.bashrc

To test it:

In three different tabs execute these commands in this order:

roscore
rosrun turtlesim turtlesim_node
rosrun turtlesim turtle_teleop_key

In case you wanted to run everything in one shot:

After opening the Linux shell:

sudo apt update
sudo apt upgrade -y
sudo apt-get install -q -y --no-install-recommends dirmngr gnupg2
sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv-keys C1CF6E31E6BADE8868B172B4F42ED6FBAB17C654
sudo  sh -c 'echo "deb http://packages.ros.org/ros/ubuntu focal main" >> /etc/apt/sources.list.d/ros1-latest.list'
sudo apt-get update 
sudo apt-get install -y --no-install-recommends ros-noetic-ros-core=1.5.0-1*
source "/opt/ros/noetic/setup.bash"
echo "source /opt/ros/noetic/setup.bash" >> ~/.bashrc
source ~/.bashrc
sudo apt-get install --no-install-recommends -y  build-essential  python3-rosdep  python3-rosinstall   python3-vcstools 
sudo rosdep init 
rosdep update --rosdistro noetic
sudo apt-get update 
sudo apt-get install -y --no-install-recommends ros-noetic-ros-base=1.5.0-1*
sudo apt-get install -y --no-install-recommends   ros-noetic-perception=1.5.0-1* 
sudo apt-get install -y --no-install-recommends nano vim tmux gedit ros-noetic-rqt* net-tools iproute2 ros-noetic-plotjuggler* ros-noetic-foxglove-bridge ros-noetic-turtlesim ros-noetic-tf*
sudo apt-get install -y --no-install-recommends  ros-noetic-stage* ros-noetic-gmapping ros-noetic-slam-toolbox* ros-noetic-move-base* ros-noetic-amcl ros-noetic-global-planner ros-noetic-dwa-local-planner ros-noetic-navfn ros-noetic-turtlebot3-* ros-noetic-teleop-twist-keyboard ros-noetic-pointcloud-to-laserscan 
mkdir -p ~/catkin_ws/src
cd ~/catkin_ws/
ln -s /opt/ros/noetic/share/catkin/cmake/toplevel.cmake src/CMakeLists.txt
catkin_make
echo "source ~/catkin_ws/devel/setup.bash"  >> ~/.bashrc

Quality of life improvements (you can skip this)

If you want an easier tmux experience with mouse support (like scrolling for output history, click to change pane, etc) and color:

cd $HOME
touch ~/.tmux.conf
echo "setw -g mouse on" >> ~/.tmux.conf
echo "setw -g default-terminal "tmux-256color"" >> ~/.tmux.conf

Cloning

Now your ROS developement environment is fully functioning, so we can just “clone” the distro to be able to interact in a multihost configuration. Write down your Ubuntu username, it’s just the name of the account that we setup in the first step. It also appears before the @ in your shell.

sudo nano /etc/wsl.conf

Now we have to modify the text file into this:

[boot]
systemd=true
[user]
default=UsernameThatIJustToldYouToWriteDOwn

And now just CTRL + O to save, same name,CTRL + X to exit.

Now create an empty folder in a place of your liking (possibly in a disk with at least 20Gb)

Open a Powershell terminal in that folder and type:

wsl --shutdown
wsl.exe --export Ubuntu-20.04 UbuntuRosBck.tar

Just wait….

If everything went correctly then you can type in the same terminal:

wsl.exe --import UbuntuRosSecondary ./filesystemsecondary UbuntuRosBck.tar

Wait, again. As soon as the process is complete we can just launch the instance with

wsl.exe --distribution UbuntuRosSecondary 

Now we have two identical WSL instances that we can launch from the start menu or from the terminal drop down.

Now, going back to the directory where we exported the backup at the start of this step we have two big files: UbuntuRosBck.tar and filesystem\ext4.vhdx

⚠️ Warning

Precaution when moving the generated files pt. 1

For filesystem\ext4.vhdx: DON’T TOUCH, MOVE OR DELETE IT. It contains all your data.

💡 Tip

Precaution when moving the generated files pt. 2

For UbuntuRosBck.tar instead you can either delete it or you can also share it with your friends.

Testing

With this configuration you don’t have to fiddle with Master IPs and such like for Docker. To test it just run like this:

roscore
rosrun turtlesim turtlesim_node
rosrun turtlesim turtle_teleop_key

Sharing is Caring

UbuntuRosBck.tar basically contain a full linux distro with all the ROS packages already installed. You can transfer it on their PC and then in an empty folder just run:

wsl.exe --import UbuntuRosPrimary ./filesystemprimary UbuntuRosBck.tar
wsl.exe --import UbuntuRosSecondary ./filesystemsecondary UbuntuRosBck.tar

Ideally this should just run™

Be mindful to not put anything personal or harmful inside and to give them the password you used in the first step. It’s technically a local VM so it shouldn’t bee too much of a risk from that point of view.

❗ Caution

Just in case it’s not clear:

Your WSL password, not your Windows password!

Also, like for the previous section: please never delete the filesystemprimary\ext4.vhdx and filesystemsecondary\ext4.vhdxfiles but you can delete UbuntuRosBck.tar file without causing problems.

Quick course in WSL

In case you never used WSL these are some nice tips that can become handy when developing with ROS.

VS Code Integration

If you have VS Code installed in Windows then you can open it in your current directory with just

code .

Don’t forget the point.

For code completion, in your catkin_ws:

From the linux shell.

cd ~/catkin_ws
mkdir .vscode
cd .vscode
touch c_cpp_properties.json
cd ..
code .

Install the catink_tools extension.

Then paste this in the .vscode/c_cpp_properties.json file.

{
    "configurations": [
        {
            "name": "Linux",
            "includePath": [
                "${workspaceFolder}/**",
                "/opt/ros/noetic/include"
            ],
            "defines": [],
            "compilerPath": "/usr/bin/gcc",
            "cStandard": "c17",
            "cppStandard": "gnu++14",
            "intelliSenseMode": "${default}",
            "configurationProvider": "b2.catkin_tools"
        }
    ],
    "version": 4
}

Accessing your files

At any moment you can launch your Windows file explorer in the current directory with just

explorer.exe .

Don’t forget the point.

It will also show up in the left bar of the built-in file explorer.

To access your ROS file simply go to ubuntu-20.04 -> home -> "your username" -> catkin_ws

Let’s try something simple: I’m going to put a Latex expression and see how Katex compares to “raw” MathML code generated by the coutless renderers you can find online:

Latex Code:

\vec x \equal 
\begin{bmatrix}

\begin{pmatrix}
1 & 2 & 3\\
a & b & c
\end{pmatrix}
&
\begin{pmatrix}
4 & 5 & 6\\
d & e & f
\end{pmatrix}
\\
\begin{pmatrix}
6 & 5 & 4\\
z & y & x
\end{pmatrix}
&
\begin{pmatrix}
3 & 2 & 1\\
r & s & t
\end{pmatrix}

\end{bmatrix}

\begin{bmatrix}
\pi \\ \pi/2 \\ \frac{\pi}{3} \\ 3\pi \\ \pdv{f}{t} \\ \alpha
\end{bmatrix}

+
\vec {z^T} \vec g
\begin{bmatrix}

\begin{pmatrix}
1 & 2 & 3\\
a & b & c
\end{pmatrix}
&
\begin{pmatrix}
4 & 5 & 6\\
d & e & f
\end{pmatrix}
\\
\begin{pmatrix}
6 & 5 & 4\\
z & y & x
\end{pmatrix}
&
\begin{pmatrix}
3 & 2 & 1\\
r & s & t
\end{pmatrix}

\end{bmatrix}

\begin{bmatrix}
\pi \\ \pi/2 \\ \frac{\pi}{3} \\ 3\pi \\ \pdv{f}{t} \\ \alpha
\end{bmatrix}

+ \sum_{i=\vec {x_0}}^{\vec {x_n}}g(\vec x_i)\Lambda_2

Katex Rendering

MathML with default font family

They should render pretty similar depending on your browser. For example on Windows 11 it works “fine” on both Chrome and Firefox while on Android I’ve encountered some errors in Chrome.

What if we wanted to change the font to have a better control and fix some of the shortcomings of default fonts?

My choice went on Asana, Firamath, GFSNeohellenicMath, Stix 2 and TEX Gyre Schola:

Font family: Asana

Font family: Firamath

Font family: FSNeohellenicMath

Font family: STIXTwo

Font family: TexGyreSchola

The parenthesis behaviour on the multiline parts is a bit heptic on Chrome, on the other hand Firefox works fine with all of them.

The problem with Firefox is that it doesn’t recognize the max-width=100% and overflow: auto hidden so it goes beyond the page layout in the MathML sections, while Katex works just fine. If you know if that’s a bug in Firefox or an error of mine please get in touch with me.

One solution might be to try to keep all the math parts as compact as possible while using one of the working fonts but that’s really limiting tbh.

For now I decided to reamain on client-side Katex with default output settings, despite the overhead of the JS library the user experience and mantainability is just better that way.

tl,dr:

💡 Tip

Final Formula Given a system of springs and bodies in degrees of freedom, we can calculate the static preload of our springs at equilibrium by:

with as the vector of our ordered solution, with

and

is our standard lagrangian matrix for the springs, instead is built by selecting the components of vertical displacement from (lagrangian matrix of the inertial terms) and as the list of masses in column form.

Proof 1-DOF case

Let’s start by defining our problem. is our free coordinate, so we have:

is the potential energy of our system and it can be divided into an elastic component and a gravitational component .

The equilibrium is defined as the point so that

To simplify our math let’s suppose that (it’s a simple change of coordinates that will help in the N-dof case).

The first step is to derive the components of the potential energy:

Elastic energy 1-dof

describes the energy stored in an ideal linear spring with Hooke’s coefficient subject to a deformation

We can ignore the index for now and express as , with as the total length of our spring and as the initial length of the spring at resting position, thus giving us the convention of positive sign for tension and negative sign for compression of our spring.

We can expand it as:

We are going to square and derive this expression, so every term of with power greater or equal than two is going to show up as a function of and when evaluated at they simply dissapear.

When putting it back into the sum we get:

Let’s evaluate the derivative in respect to :

is a constant and gets eliminated by the derivative, the other sqare term becomes and goes to zero when evaluated at , so only the mixed term survives and the derivative becomes:

We can rewrite this in vector form as:

Gravitational energy 1-dof

This expresses the total gravitational energy in our system supposing bodies with mass and height measured from an arbitrary system of coordinates.

Applying the same concepts as the elastic term we get:

Let’s focus on the lower order terms and we get:

In this context is the linearized displacement of the height at equilibrium.

Deriving it we get

Let’s put everything in vector form

Final expression

To find we just rearrange it as:

N-DOF case

We follow the same procedure but now we have additional dimensions for our :

So we need to evaluate the gradient of our potential energy funcion for the equilibrium:

Elastic energy

with

As usual we expand our near the equilibrium:

and we simplify it as:

Plugging it into the initial equation we get:

Like for our 1-DOF case the only terms that survive the gradient are the mixed ones:

ℹ️ Note

For and .

Then is just a scalar. So:

If is symmetric:

Gravitational Energy

This must be straightforward now if you understood the past passages but for completeness let’s write them down:

with

Expanding the height we get:

Picking just the linear and terms we get:

Transposing and putting it back in the original energy expression we get:

Deriving it we get

Final formula

The story¹² goes like: If you want to use videos instead of GIFs to save on bandwith for your users you can just use something like:

<video autoplay loop muted playsinline>
<source src="cat.webm" type="video/webm">
<source src="cat.mp4" type="video/mp4">
<img src="cat.gif"/>
</video>

The <img src="cat.gif"/> will behave like a fallback for browsers not supporting the video element or the provided formats. Just use ffmpeg to convert GIFs to MP4 and/or webm and you will have a couple of decades of video encoders developement to show animated cats in the fastest way possible.

But no, even if your noble intents to show animated cats to your visitors while saving their bandwith seem completed there is a problem:

With the code showed above all the browsers will download the video file AND the gif file, throwing outside the window your bandwith saving intents and making it worse because you’re forcing your visitors to download an additional file, no matter how small it is.

You can try it directly

As you see in the screen above the browser still downloads the “fallback” GIF.

This behaviour doesn’t seem widely reported in the “use videos instead of gifs” guides, staying relegated to some StackOverflow discussions ³⁴.

How should you resolve this?

1. Don’t provide an <img>, just link to the GIF and ask them to update the browser to something more “modern”.⁵

   html

   <video autoplay loop muted playsinline>
   <source src="cat.webm" type="video/webm">
   <source src="cat.mp4" type="video/mp4">
   <p>In case you browser doesn't support video files open the <a target="_blank" href="cat.gif">GIF</a> in another page</p>
   </video> 

2. Fix stuff server side, check the UA and provide the <video> instead of the <img> or viceversa depending on it. If your content is statically rendered then it won’t work.

3. The <!--[if lt IE 9]>⁶. This will work because the only main browsers not supporting the video elements are the Internet Explorer previous to the 9th and Opera Mini. Your code will just be

   html

   <video autoplay loop muted playsinline>
   <source src="cat.webm" type="video/webm">
   <source src="cat.mp4" type="video/mp4">
   <!--[if lt IE 9]><img src="cat.gif"/><![endif]-->
   </video> 

4. No GIF fallback. Really. At the time of writing those IE versions account for a 0,06% of the broswer share. Opera Mini is a bit more used (1,19%) but I think that CanIUse uses the UA to calculate the browser usage regardless of the type of data saving enabled, while for the tests it uses the aggressive server rendering option. In the worst case scenario the 1,25% of your visitors will have a degraded experience without the animated cats. Being more realistics I will say that even before the lack of animated cats those 1,25% will have a bad time in 2021 to reach your website between expired certificates, unsupported protocols, etc. (Try downloading an old Firefox version to see it by yourself).

It happened to me to buy a new laptop. This laptop comes with all the usual OEM stuff to manage security and updates. One of those software scans every downloaded file and marks its origin. To do that the software enables the GroupPolicyUsers and installs an extension.

Ok, cool but now whenever I want to change some Chrome (or even Firefox) settings I’m greeted with this message:

Since I wanted to use my NextDNS account I had to fix this.

First try was removing the User Policy by using a simble bash script found somewhere online

@echo off

IF NOT EXIST %WINDIR%\\System32\\GroupPolicy goto next

echo Deleting GroupPolicy folder...
RD /S /Q "%WINDIR%\\System32\\GroupPolicy" || goto error
echo.

:next
IF NOT EXIST %WINDIR%\\System32\\GroupPolicyUsers goto next2

echo Deleting GroupPolicyUsers folder...
RD /S /Q "%WINDIR%\\System32\\GroupPolicyUsers" || goto error
echo.

:next2
gpupdate /force

pause
exit

:error
echo.
echo An unexpected error has occurred. ¨Have opened the program as an administrator (right click, run as administrator)?
echo.
pause
exit

Look at all the delete folder commands so close to System32, that absolutely doesn’t make nervous.

But this resetted at every boot. A more robust soultion was needed.

Removing the OEM security software? I was afraid of breaking some weird software integration with the fingerprint and boot security software. 😅

Now the final solution: If you can’t remove the User Policy use it.

Looking up about the various admin policies on the Google Chrome website it’s possibile to find the ones useful for us: DnsOverHttpsMode and DnsOverHttpsTemplates.

Just set DnsOverHttpsMode to secure or automatic depending if you want it to not have an insecure fallback.

Then DnsOverHttpsTemplates you need to fill your DoH provider.

The Chrome Policy is a bit confusing but you just need to take the https URL and add /dns-query{?dns} after it (better explaination).

So if NextDNS provides you with https://dns.nextdns.io/123abc your URL to fill is https://dns.nextdns.io/123abc/dns-query{?dns}

Here is a list of various DoH providers with everything already filled in:

Bot how to enable it? You can edit manual through regedit or you can copy this code in a new file and save it with a .reg extension. You can directly insert your DNS of choice in place of the default Google one.

Windows Registry Editor Version 5.00

[HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Google\Chrome]
"DnsOverHttpsTemplates"="https://dns.google/dns-query{?dns}"
"DnsOverHttpsMode"="automatic"

As you can see from the table and from the graph, downloading the test file usually takes 200-250ms on Github Pages and Cloudflare, while on Netlify the times as 2 times longer with a more unreliable time for the latter.