Drafts

• 11ty https://www.rockyourcode.com/how-to-deploy-eleventy-to-github-pages-with-github-actions/
• hugo puts it in the base url https://gohugo.io/getting-started/configuration/#baseurl
• mkdocs uses a special cli build command https://squidfunk.github.io/mkdocs-material/publishing-your-site/#github-pages

Upon first running an aws cli command using localstack you might end up with the following error.

Unable to locate credentials. You can configure credentials by running "aws configure".

Easy way #

The easy easiest way is to leverage a package called awscli-local.

pipx install awscli-local

Leveraging the awscli #

If you want to use the cli pro

pipx install awscli

aws config --profile localstack
# put what you want for the keys, but enter a valid region like us-east-1

alias aws='aws --endpoint-url http://localhost:4566 --profile localstack'

npx create-react-app todoreact

import React,{useState,useEffect} from 'react';
import './App.css';

function App() {
  const [data,setData]=useState([]);
  const [newName,setNewName]=useState([]);
  const getData=()=>{
    fetch('/api'
    ,{
      headers : {
        'Content-Type': 'application/json',
        'Accept': 'application/json'
       }
    }
    )
      .then(function(response){
        return response.json();
      })
      .then(function(myJson) {
        setData(myJson)
      });
  }
  useEffect(()=>{
    getData()
  },[])

  const addItem= async () => {
    const rawResponse = await fetch('/api/add/', {
      method: 'POST',
      headers: {
        'Accept': 'application/json',
        'Content-Type': 'application/json'
      },

    body: JSON.stringify({"name": newName})
    });
    const content = await rawResponse;

    console.log(content);
    getData()
  }




  return (
    <div className="App">
     {
       data && data.length>0 && data.map((item)=><p>{item.id}{item.priority}{item.name}<button>raise priority</button></p>)
     }
    <input type='text' value={newName} onChange={(e) => (setNewName(e.target.value))} />
    <button onClick={addItem} >add item</button>
    </div>
  );
}

export default App;

Hatch allows you to specify direct references for dependencies in your pyproject.toml file. This is useful when you want to depend on a package that is not available on PyPI or when you want to use a specific version from a Git repository. Often used for unreleased packages, or unreleased versions of packages.

docs

[project]
dependencies = ['markata', 'markata-todoui@git+https://github.com/waylonwalker/markata-todoui']

[tool.hatch.metadata]
allow-direct-references=true

https://gist.github.com/YoEight/d19112db56cd8f93835bf2d009d617f7

xrandr is a great cli to manage your windows in a linux distro using x11, which is most of them. The issue is that I can never remember all the flags to the command, and if you are using it with something like a laptop using a dock the names of all the displays tend to change every time you redock. This makes it really hard to make scripts that work right every time.

Homepage #

Check out the deresmos/xrandr-manager for more details on it.

installation #

xrander-manager is a python cli application that is simply a nice interface into xrandr. So you must have xrandr already installed, which is generally just there on any x11 window manager, I’ve never had to install it.

As with any python cli that is indended to be used as a global/system level cli application I always install them with pipx. This automates the process of creating a virtual environment for xrandr-manager for me, and does not clutter up my system packages with its dependencies that may eventually clash with another that I want to use.

# prereqs (xrandr, pipx)
pipx install xrandr-manager

set main monitor #

First if your main display is not set to the correct monitor set your main display first.

xrandr-manager -m HDMI-0
xrandr-manager -m DP-0

prompt mode #

If you dont know the name of your monitors and and don’t want to dig through xrandr, you can just run --prompt and tab complete to fill set your main display.

xrandr-manager --prompt

direction #

This is what I most often use xrandr-manager for. Once you have the main display set you can tell it where to put the other monitor. I’ve only tried this with two monitors, I have no idea what happens with more monitors.

xrandr-manager -d right
xrandr-manager -d left
xrandr-manager -d above
xrandr-manager -d below

mirror #

One thing that I always need to jump through hoops to do is mirror. Occasionally I want to mirror so that more people can see the screen while we are split screen gaming. This has seemed like a pain in any other xrandr utility, but trivial in xrandr-manager.

xrandr-manager --mirror

It logs out the xrandr command #

One nice thing about xrandr-manager is that it echos out the xrandr command that it’s running. This is nice because you can toss this behind a hotkey or an init script.

Guis #

Ya there are guis that do this. I’ve had good luck with arandr. It’s more intuitive to drag windows around like what you would do in windows. Every once in awhile it messes up and my polybar overlaps my windows, or my windows end up only on half the screen.

There are also graphics card specific utilities, Ive used nvidia x server settings and it mostly works similar to arandr.

jq has some syntax that will sneak up on you with complexity. It looks so good, and so understandable, but everytime I go to use it myself, I don’t get it. ijq is an interactive alternative to jq that gives you and nice repl that you can iterate on queries quickly.

paru -Syu ijq

Here are some other articles, I decided to link at the time of writing this article.

JUT | Read Notebooks in the Terminal

Comprehensive guide to creating kedro nodes

Kedro - My Data Is Not A Table

I love getting faster in my workflow, something I have recently added in is creating GitHub repos with the cli. I often create little examples of projects, but they just end up on my machine and not anywhere that someone else can see, mostly because it takes more effort to go create a repo. TIL you can create a repo right from the command line and push to it immediately.

gh repo create waylonwalker-cli

want to see what this repo I created is about? #

Check out what I created here.

pipx run waylonwalker

totally guessed at this post’s date

I’m still trying to understand this one, but this is how you force a python object to stop atexit.

import atexit

class Server:
    def __init__(
        self,
        auto_restart: bool = True,
        directory: Union[str, "Path"] = None,
        port: int = 8000,
    ):
        if directory is None:
            from markata import Markata

            m = Markata()
            directory = m.config["output_dir"]

        self.directory = directory
        self.port = find_port(port=port)
        self.start_server()
        atexit.register(self.kill)

    def start_server(self):
        import subprocess

        self.cmd = [
            "python",
            "-m",
            "http.server",
            str(self.port),
            "--directory",
            self.directory,
        ]

        self.proc = subprocess.Popen(
            self.cmd,
            stderr=subprocess.PIPE,
            stdout=subprocess.PIPE,
        )
        self.start_time = time.time()


    def kill(self):
        self.auto_restart = False
        self.proc.kill()

    def __rich__(self) -> Panel:
        if not self.proc.poll():
            return Panel(
                f"[green]serving on port: [gold1]{self.port} [green]using pid: [gold1]{self.proc.pid} [green]uptime: [gold1]{self.uptime} [green]link: [gold1] http://localhost:{self.port}[/]",
                border_style="blue",
                title="server",
            )

        else:
            if self.auto_restart:
                self.start_server()

            return Panel(f"[red]server died", title="server", border_style="red")

Whenever you are installing python packages, you should always use a virtual environment. pip makes this easy to follow by adding some configuration to pip.

require-virtualenv #

Pip is the pacakage tool for python. It installs third-party packages and is configurable. One of the configuration settings that I highly reccommend everyone to add is require-virtualenv. This will stop pip from installing any packages if you have not activated a virtualenv.

why #

python packages often require many different dependencies, sometimes packages are up to date and sometimes they require different versions of dependencies. If you install everything in one environment its easy to end up with version conflict issues that are really hard to resolve, especially since your system environment cannot easily be restarted.

PIPX my one exception #

My one exception that I put in my system level packages is pipx. pipx is very handy as it manages virtual environments for you and is intended for command line utilities that would end up in your system env or require you to manually manage virtual environments without it.

pip config #

Your pip config might be found in either ~/.pip/pip.conf or ~/.config/pip/pip.conf. You can either use the pip config set command or edit one of these files manually.

pip config set global.require-virtualenv True

Now you sould see this in your ~/.config/pip/pip.conf

[global]
require-virtualenv = True

pip config debug #

If you want to know where pip is looking for configuration on your system, and what files are setting a certain config you can use pip config debug to find it.

❯ pip config debug

env_var:
env:
global:
  /etc/xdg/xdg-awesome/pip/pip.conf, exists: False
  /etc/xdg/pip/pip.conf, exists: False
  /etc/pip.conf, exists: False
site:
  /home/waylon/git/waylonwalker.com/.venv/pip.conf, exists: False
user:
  /home/waylon/.pip/pip.conf, exists: False
  /home/waylon/.config/pip/pip.conf, exists: True
    global.require-virtualenv: True

saved my bacon #

This setting recently saved me when I modified my .envrc file my virtual environment deactivated, so when I went to pip install something it gave me an error that it was not active. Situations like this are an easy way to pollute your system with packages that it does not need installed.

TLDR #

Run this at your command line to avoid polluting your system environment by mistake before running any pip command.

pip config set global.require-virtualenv True

I’ve been trying to adopt pyenv for a few months, but have been completely blocked by this issue on one of the main machines I use. Whenever I start up ipython I get the following error.

ImportError: No module named '_sqlite3

I talked about why and how to use pyenv along with my first impressions in this post

pyenv/issues/678 #

According to #678 I need to install libsqlite3-dev on ubuntu to resolve this issue.

install libsqlite3-dev #

libsqlite3-dev can be installed using apt

sudo apt install libsqlite3-dev

But wait…. #

When I make a fresh env and install ipython I still get the same error and I am still not able to use ipython with pyenv.

ImportError: No module named '_sqlite3

re-install python #

After having this issue for awhile an coming back to #678 several times I realized that libsqlite3-dev needs to be installed while during install.

pyenv install 3.8.13

I think I had tried this several times, but was missing the -y option each time. You gotta read errors like this, I am really good at glossing over them.

Let’s never have this issue again. #

When you spend months living with little errors like this and finally fix it, its good to make sure that it never happens again. Whenever I start a new ubuntu machine I run an ansible playbook that does all the setup for me. I added libsqlite3-dev to my core install in 64c85ca now it will be on all of my machines and not break again.

Sometimes you have a pretty old branch you are trying to merge into and you are absolutely sure what you have is what you want, and therefore you don’t want to deal with any sort of merge conflicts, you would rather just tell git to use my version and move on.

update main #

The first step is to make sure your local copy of the branch you are moving into is up to date.

git checkout main
git pull

update your feature branch #

It’s also worth updating your feature branch before doing the merge. Maybe you have teammates that have updated the repo, or you popped in a quick change from the web ui. It’s simple and worth checking.

git checkout my-feature
git pull

start the merge #

Merge the changes from main into my-feature branch.

git merge main

Now is where the merge conflict may have started. If you are completely sure that your copy is correct you can --ours, if you are completely sure that main is correct, you can --theirs.

git checkout --ours .
git merge --continue

This will pop open your configured git.core.editor or $EDTIOR. If you have not configured your editor, it will default to vim. Close vim with <escape>:x, accepting the merge message.

Now push your changes that do not clash with main and finish your pr.

git push