DEV Community: Kedasha

The GitHub Community Roundup: 03-08-24

Kedasha — Fri, 08 Mar 2024 15:59:56 +0000

Hey folks! Welcome to another edition of the GitHub Community Roundup! There's been some amazing content this week that from the community:

I enjoyed this post by @aksrao1998 where he shows us how to triggers workflows in two repositories and share data between the two.
@shariqahmed525 shared pros and cons of using GitHub Copilot which was quite insightful!
Yan Holtz wrote a nice introduction to GitHub! Check it out, especially if you're new to the platform.
// Detect dark theme var iframe = document.getElementById('tweet-1766096644880183681-171'); if (document.body.className.includes('dark-theme')) { iframe.src = "https://platform.twitter.com/embed/Tweet.html?id=1766096644880183681&theme=dark" }
Shinya's post on CodeQL teaches us why the tool is a beneficial game changer for code analysis.
Learn about the GitHub Student Developer pack! An amazing programming resource for students filled with tonns of benefits! Thanks for the great write up Billups.

If you’ve written an article about GitHub or find some great content about GitHub or any of its features or products, please share it with the #GitHub hashtag on Dev.to, Twitter or LinkedIn!

What's new at GitHub this week?

Well, exciting news! GitHub Copilot Chat is now GA in JetBrains IDE! 🚀

Learn more about how you can get started by reading our blogpost or watch the youtube video.

Learn how to create your first terraform module with GitHub by checking out this video.
Did you know that Autocoder was one of the first attempts at automatic coding? Back in 1950?! Yea, me either! Read more about the Hard and soft skills for developers coding in the age of AI where we touch on the brief history of Ai-powered tools.
Learn How GitHub uses merge queue to ship hundreds of changes every day in this blog post
Discover how to keep repository maintainer information accurate through CODEOWNERS file in this blog post.

A huge thanks goes out to all the contributors to the technical community this week!

Happy Friday and we'll see you all next time for the latest GitHub community updates. ✨

How I Bulk Closed 1000+ GitHub Issues with GitHub Actions 🚀

Kedasha — Wed, 17 Jan 2024 18:29:57 +0000

Now why would I want to do this huh? 🤔

Well, we have a private repo where maintainers are able to make requests to join the private maintainer community.

Someone thought it'd be cute to submit 1000+ issues to the repo with the handle @undefined and the title undefined. This was a problem because it was making it difficult to find the actual issues that were being submitted by maintainers that needed to be approved.

We need to do some additional work to the repo before I'm able to add validation to the issue form but in the meantime, we needed to close these issues. So, I thought a temporary solution would be to create a GitHub action that closed all of the issues - with the title containing Pending invitation request for: @undefined'. And it worked!!

We went from having over 1,600+ issues to 64 valid issues in a matter of minutes.

Here's how I did it.

Creating the GitHub Action

The first thing I did was go to my bestie chatty (GitHub Copilot Chat) and asked it if I could close 1000+ issues with a GitHub action. It gave me a few options, but I decided to go with this one:

Then it generated the closeIssue.js file which looked like this:

Sweet, that was the bulk of the work. Now I just needed to add the action to my repo.

I created a file called close_issues.yml in the .github/workflows directory and added the following code:



name: Close Undefined Issues
on:
    workflow_dispatch:
    push:
      branches: [ "main" ]
    pull_request:
      # The branches below must be a subset of the branches above
      branches: [ "main" ]
    schedule:
      - cron: '43 0 * * 1'

jobs:
  close_issues:
    runs-on: ubuntu-latest
    steps:
      - name: Checkout code
        uses: actions/checkout@v2

      - name: Setup Node.js
        uses: actions/setup-node@v2
        with:
          node-version: 14

      - name: Install dependencies
        run: npm install

      - name: Run script
        run: node closeIssues.js
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

I made a few tweaks to the code, but all in all, this code was written largely in collaboration with GitHub Copilot.

The next file I created was the closeIssues.js file. I added the following code:




import { Octokit } from "@octokit/rest";

const octokit = new Octokit({
  auth: process.env.GITHUB_TOKEN,
});

async function closeIssues() {
  const { data: issues } = await octokit.issues.listForRepo({
    owner: 'OWNER',
    repo: 'internal-repo',
  });

  const issuesToClose = issues.filter(issue => issue.title === 'Pending invitation request for: @undefined');

  for (const issue of issuesToClose) {
    await octokit.issues.update({
      owner: 'OWNER',
      repo: 'internal-repo',
      issue_number: issue.number,
      state: 'closed',
    });
  }
}

closeIssues().catch(console.error);

I added a package.json file to my repo, install the needed dependency, and then pushed the code to my repo.

Testing the Action

After about 3 (ok 4!) tries, I finally got the action to run successfully.

Initially, the action closed 30 issues at a time, because that is the limit for the GitHub API. So, I had to read the docs 🤯 to understand how to grab ALL issues that matched the criteria and closed them.

Updates

After reading the docs, I realized that I needed to use the paginate() method to get all of the issues for the specified repo. I also learned that with paginate, the issues are returned as an array, so I no longer needed const { data: issues }. I also needed to use octokit.rest.issues when making requests.

In the end, the code that ran successfully and closed all 1000+ issues AT ONCE looked like this:



import { Octokit } from "@octokit/rest";

const octokit = new Octokit({
    auth: process.env.GITHUB_TOKEN,
});

async function closeIssues() {
    try {
        console.log('Fetching issues...');
        const issues = await octokit.paginate(octokit.rest.issues.listForRepo, {
            owner: 'OWNER',
            repo: 'internal-repo',
            state: 'open',
        });

        console.log(`Fetched ${issues.length} issues.`);
        const issuesToClose = issues.filter(issue => issue.title === 'Pending invitation request for: @undefined');
        console.log(`Found ${issuesToClose.length} issues to close.`);

        for (const issue of issuesToClose) {
            console.log(`Closing issue #${issue.number}...`);
            await octokit.rest.issues.update({
                owner: 'OWNER',
                repo: 'internal-repo',
                issue_number: issue.number,
                state: 'closed',
            });
            console.log(`Closed issue #${issue.number}.`);
        }
    } catch (error) {
        console.error(error);
    }
}

closeIssues();

I added some logging to the code so I could see what was happening during the workflow run. I also added a try/catch block to catch any errors that may occur.

Conclusion

I was very happy that I could do this because it would've taken my teammate HOURS to close all these issues manually. I was able to do it in a matter of minutes! I'm super stoked that GitHub Copilot was able to teach me something new - how to interact with the GitHub API using the Octokit library. 🚀

Have you ever used GitHub Copilot to help you write GitHub Actions? Let me know in the comments below! I hope you learned something new today.

Until next time, happy coding! 😁

Getting Started with GitHub Copilot in the CLI🚀

Kedasha — Fri, 17 Nov 2023 17:09:17 +0000

If you've ever used the terminal, you know the struggle of trying to remember a command, the fear of breaking your system with a random command you found on the interwebs or the frustration of trying to find the right command to do what you want.

That struggle is lightened today, with GitHub Copilot in the CLI! 💃🏼

GitHub Copilot in the CLI provides a chat-like interface in the terminal that allows you to ask questions about the command line.

It is designed to help you with general shell commands, git commands and gh cli commands. This post aims to get you up and running with installing, and using copilot in the cli.

Installation

To install copilot in the cli, you must first install GitHub CLI and complete authentication in an OAuth browser window. If you're not familiar with the GitHub CLI, learn how to use it by reading this blog post!
Since I'm on macOS, I used homebrew as my package manager:



brew install gh
gh auth login

You also need to have an active GitHub Copilot subscription to use copilot in the cli.
Once the gh cli is installed, and you have a copilot subscription, you can install the extension by running:



gh extension install github/gh-copilot

And you are ready to go! 🎉

Available Commands

I recommend running the help command to understand how to interact with copilot in the cli. When you run gh copilot --help you will see the following output:



gh copilot

Your AI command line copilot.

Usage:
  copilot [command]

Available Commands:

  config      Configure options
  explain     Explain a command
  suggest     Suggest a command

Flags:
  -h, --help      help for copilot
  -v, --version   version for copilot

Use "copilot [command] --help" for more information about a command.

There are two primary commands you can use to interact with copilot in the cli: gh copilot explain and gh copilot suggest.

Create an Alias 💡

💡 Hot tip! creating an alias will make using copilot in the cli a part of your regular workflow a lot easier.

After installation, the next thing I advise you to do is to create an alias. This will make interfacing with copilot in the cli a lot smoother because it saves you some keystrokes and you can get help faster.

Since I'm using zsh, I ran the following commands to create aliases:



alias copilot='gh copilot' ; echo 'alias copilot="gh copilot"' >> ~/.zshrc && source ~/.zshrc
alias gcs='gh copilot suggest' ; echo 'alias gcs="gh copilot suggest"' >> ~/.zshrc && source ~/.zshrc
alias gce='gh copilot explain' ; echo 'alias gce="gh copilot explain"' >> ~/.zshrc && source ~/.zshrc

Now, let's get into what these commands mean and a few examples on how to use them.

💡 Remember, I have an alias so I'll be using gce or gcs instead of gh copilot explain or gh copilot suggest.

Explain Command 🤖

There are two ways you can use the copilot cli explain command:

You can run the explain command with no arguments and it will prompt you to enter a command to explain.



gh copilot explain

You can run the explain command and the command you want explained at the same time:



gh copilot explain 'your query'

You can also run the explain command with the -h flag to get more information:



gh copilot explain -h

Suggest Command 🤖

There are 3 ways you can interact with copilot in the cli to get a suggestion:

You can run the suggest command and follow the on screen prompts:



gh copilot suggest

You can run the suggest command and your query at the same time, then select the type of command:



gh copilot suggest 'command you want to run'

You can run the suggest command, your query and the type of command at the same time:



gh copilot suggest 'command you want to run' -t 'type of command'

The -t flag means type of command and can be one of the following:

git
gh
shell

Revising Suggested Commands

Sometimes, the suggestion you get from copilot in the cli is not exactly what you want. You can revise the command by selecting the Revise command option when you receive a suggestion and give copilot about more detail about what you want:

💡 Hot tip! Copilot in the CLI is not there to simply give you the suggested you want. Remember, it is there as your co-pilot, so if the output is not quite what you expected, edit the suggestion or ask for a revision until you get what you need.

Explaining Suggested Commands

When you get a suggestion from copilot in the cli, you can ask copilot to explain the command by selecting the Explain command option:

Copying Suggested Commands

When you get a suggestion from copilot in the cli, you can copy the provided response and run it in your terminal.

Rating Commands 🌟🌟🌟🌟🌟

When you get a suggestion from copilot in the cli, you can rate the command by selecting the Rate command option, and selecting whether or not it was helpful:

Exiting the prompt flow window 🚪

You can exit the prompt flow window by selecting the Exit option:

Examples of using Copilot in the CLI

Now let's take a look at some examples. 💃🏼

1. Explain a dangerous command ⚠️

So, I saw this command online and was curious about what it did so I can ask:



gh copilot explain chmod -r 777 /

💡 Hot tip! When asking copilot to explain commands that have special characters, enclose the command in a string so copilot can understand it better.

As you see, this command is actually very dangerous and can cause you to lose all your data. What I love about the explanation I received was that it not only told me what the command does, but it also told me why it was dangerous.

🚨 NEVER RUN THIS COMMAND ON YOUR MACHINE 🚨

2. Delete a git branch locally and remotely

In the visual example below, I first list the branches in my local repo with the command git branch then I list the branches in my remote repo with the command git branch -r. I then ask copilot in the cli for a suggestion:



gh copilot suggest 'delete a git branch locally and remotely' -t git

I was provided with the following suggestion:



git push origin --delete <branch_name>

git branch -D <branch_name>

I ran the suggested commands, then listed the branches in my local repo and remote repo again to confirm that the branch was deleted. Thankfully, it was. 🙌🏼

View all open issues in a repo

You can also ask copilot in the cli for gh commands. For example, if you wanted to view all open issues in a repo, you can ask copilot in the cli for a suggestion:



gh copilot suggest 'view all open issues' -t gh

I was provided with the following suggestion:



gh issue list --state='open'

I ran the command and was provided with a list of all open issues in the repo I was currently in. See the flow in the visual example below:

3. Kill processes with open files that have been deleted

Copilot in the CLI can also be useful for administrative tasks in the terminal. For example, if you wanted to kill processes with open files that have been deleted, you can ask copilot in the cli for a suggestion:



gh copilot suggest 'kill processes with open files that have been deleted' -t shell

I was provided with the following suggestion:



lsof | grep '(deleted)' | awk '{print $2}' | sort -u | xargs -r kill -9

I asked copilot to explain the suggestion because I wasn't very familiar with the xargs command. See more in the video example below:

Once I ran the suggested command, I was able to run the command lsof | grep '(deleted)' | awk '{print $2}' | sort -u to see the processes that were killed.

4. Exit Vim 😅

This would not be a terminal demo if I didn't ask this one question 😂:



gh copilot suggest 'how to exit vim' -t shell

I was provided with the following suggestion:

:q!

Though this is correct, it's not quite what I was looking for, because while this will take you out of vim, you will loose all your changes in the process. So I asked copilot to revise the command:



? How should this be revised?
> how to exit vim and save changes

Suggestion:

  :wq

Perfect! 💃🏼
See this entire flow in the visual example below:

Conclusion

GitHub Copilot in the CLI is a game-changer that is super useful for reminding you of commands, teaching you new commands or explaining random commands you come across find online.

I hope you enjoy using it! Try it yourself and let me know how you used it. If you have feedback or issues, you can submit an issue in the gh-copilot repo or leave a comment below and I'll be sure to get you an answer.

Have you tried the GitHub Copilot in the CLI? What do you think? Let me know in the comments below! 👇🏼🚀

// Detect dark theme var iframe = document.getElementById('tweet-1725549992792797304-525'); if (document.body.className.includes('dark-theme')) { iframe.src = "https://platform.twitter.com/embed/Tweet.html?id=1725549992792797304&theme=dark" }

How I used GitHub Copilot Chat to debug my deployment workflow 🐥🤖

Kedasha — Mon, 13 Nov 2023 18:02:31 +0000

A few weeks ago I was working on putting together a slide deck for my talk at GitHub Universe 2023. I spoke about how you can use Github Copilot in the CLI to improve your time in the terminal.

I typically use Canva to create my presentations but I reaaaly didn't want to do that this time around because I get really anal about the small details and I didn't want to spend a lot of time on design. I wanted to focus on the content and the delivery.

So of course, I tweeted and someone suggested slidev.

I took a quick gander at it and was quite impresed with the dev experience and built in themes and decided to give it a go. I was able to create the initial draft of my slides pretty quickly and I'm quite happy with the results!

I'm not going to go into the details of how to use it but I'll link to the docs here.

Deploying my slides

So, let's get into the GitHub Actions part. So slidev has a template workflow for deploying your slides to GitHub Pages. I copied the workflow and added it to my repo, pushed to GitHub ad expected magic to happen on the first go. It didn't of course lol.

I ran into a few issues with actions and thought oh, GitHub Copilot is the perfect tool to help me out so I decided to use it to help me debug my workflow.

What is GitHub Actions

If you've never used GitHub actions before, it's essentially a way for you to automate anything in your repos. You can use it to build, test, and deploy your code right from GitHub and you can also use it to automate your project's triage process, label issues, and more.

If you want to learn more about actions, checkout my talk below from Philly ETE Conference this past April.

You can also check out this repo for a full walk through of how to use actions.

Anyway, so my workflow didn't work for a few reasons.

1. I didn't enable Github Pages

I didn't enable GitHub pages in my repo. I know, I know, lol. I was so excited to get my slides deployed that I forgot to enable it. So I enabled GitHub pages to use actions and tried again. Still didn't work.

2. The installation of `package.json` failed

I received an error stating that my directory could not be found.

So my slides are located in a folder and not the root of my repo. I had to update the workflow to point to the correct folder, so I updated the command like so:



name: Install dependencies
run:  cd ../copilot-cli && npm install

But that still didn't work. So I went to copilot and pasted the error message I saw in the actions logs:

What I love about this is that it not only provided a suggestion on how to solve the problem, copilot also gave me a bit of education: I didn't know that workflows were ran from the root of the repo. I thought they were ran from .github/workflows directory, so I learned something new.

I updated my workflow with copilots suggestion and it worked well. But then the slidev build command failed. 😫

3. Slidev build command failed

Since I had the earlier experience of forgetting to update the directory, I knew that I had to update the build command to point to the correct directory. I updated the command like so:



name: Build
run: slidev build --base /gh-copilot-cli/copilot-cli/

This goes into the repo and then goes into the folder where my slides are located and then runs the build command. It failed, so I updated the command to this:



name: Build
run: slidev build --base /copilot-cli/

That makes sense right? Since the workflow is ran from the root of the repo, I should be able to just go into the folder where my slides are located and run the build command. But that failed too.

So, I went to copilot and pasted the error message I saw in the actions logs:

Oh gotcha, I need to cd copilot-cli then point to /copilot-cli directory. I implemented copilot's suggestion and it ran successfully! 💃🏼

But then, I got another error. 😫

4. The post build command failed - no `dist` folder found

You would think by now I would've learned my lesson and updated the command to point to the correct directory. But I didn't. 😅 When the build failed again, I updated the build command to this:



uses: actions/upload-pages-artifact@v1
    with:
        path: copilot-cli/dist

I was like ok that should do it, I'm pointing to the correct directory. But it failed again. So I went to copilot and pasted the error message I saw in the actions logs:

That's what I did, hmm 🤔. I went back to my debugging buddy and it suggested the following:

After listing the directory contents as suggested, it turns out that the dist folder was not being created by the slidev build command. Here's what the logs showed:



Run ls -la copilot-cli
total 536
drwxr-xr-x   6 runner docker   4096 Oct 28 16:46 .
drwxr-xr-x   6 runner docker   4096 Oct 28 16:46 ..
-rw-r--r--   1 runner docker     78 Oct 28 16:46 .gitignore
-rw-r--r--   1 runner docker     57 Oct 28 16:46 .npmrc
-rw-r--r--   1 runner docker      9 Oct 28 16:46 .prettierignore
-rw-r--r--   1 runner docker    267 Oct 28 16:46 README.md
drwxr-xr-x   2 runner docker   4096 Oct 28 16:46 components
drwxr-xr-x   2 runner docker   4096 Oct 28 16:46 layouts
-rw-r--r--   1 runner docker    248 Oct 28 16:46 netlify.toml
drwxr-xr-x 467 runner docker  20480 Oct 28 16:46 node_modules
-rw-r--r--   1 runner docker 299331 Oct 28 16:46 package-lock.json
-rw-r--r--   1 runner docker    431 Oct 28 16:46 package.json
drwxr-xr-x   2 runner docker   4096 Oct 28 16:46 pages
-rw-r--r--   1 runner docker 166229 Oct 28 16:46 pnpm-lock.yaml
-rw-r--r--   1 runner docker   7616 Oct 28 16:46 slides.md
-rw-r--r--   1 runner docker    144 Oct 28 16:46 vercel.json

No dist directory in sight. So of course, I went back to my bestie copilot and pasted the returned directory contents. Here's what it suggested:

I implemented copilot's suggestion and it failed again. I decided to use my eagle eyes and comb through the Build logs. Turns out, the slidev build command wasn't even running! What?! 😫

.... not solved lol

That was a journey lol. 😂

5. Lessons learned

I learned a few things from this experience:

I need to read the docs
GitHub Copilot is a great debugging buddy
I need to read the docs and pay closer attention my files 😅

My slides are still not deployed LOL but here they are in PDF format!

If you've ever worked with slidev before lemme know how I can deploy my slides to gh-pages LOL. GitHub Copilot was still a great rubber ducky! 🐥

Happy coding! 💕
Kedasha

Top 10 Tips for using GitHub from the Command Line

Kedasha — Fri, 13 Oct 2023 21:47:59 +0000

The GitHub CLI (gh cli) is a command line tool that allows you to interact with GitHub without ever leaving your terminal. It's an open source tool that allows you to view and manage issues, pull requests, releases, and more. In this post, we'll cover 10 tips for using the GitHub CLI, but first, how do you install it?

Installing the GitHub CLI 🛠️

I'm using MacOS, so I'll be using Homebrew to install the GitHub CLI. If you're using Windows or Linux, you can find instructions for installing the GitHub CLI here.



brew install gh

Once you've installed the GitHub CLI, you'll need to authenticate with GitHub. You can do this by running the following command:



gh auth login

This will open a browser window and prompt you to login to GitHub. Once you've logged in, you'll be prompted to authenticate with GitHub. Once you've authenticated, you'll be able to use the GitHub CLI. 🎉

CLI Syntax and Usage 📝

The CLI follows a very specific syntax. It generally works as, "gh" "feature" "then what you want to do with the feature" then “any flags associated with the feature” - everything you want to do with the cli takes on this format.

For example, if you wanted to clone a repository using the gh cli, it would be gh repo clone <repo to be cloned>

Let's take a look at some practical tips when using the GitHub CLI.

1. Create a new repository and clone it locally 🎉

You can create a new repository using the gh repo create command. This command will prompt you for the name of the repository, the description, and the visibility. You can also specify the name of the repository as an argument to the command.

You can also create a new repository and clone it locally at the same time by using the --clone flag. Here's what this will look like:



gh repo create your-new-repo --public --clone

The repo was created here and all you need to do now is to push your changes and you're good.

2. Create issues 📝

We can create issues for this repository right from the command line with the gh cli. This command will prompt you for the title and body of the issue. You can also specify the title and body as arguments to the command.



gh issue create -t "issue title" -b "issue body content" -a "assignee"

You can also edit, transfer, delete, close, and reopen issues with the gh cli. You can find more information about this here.

3. Create pull requests 🔄

We can create pull requests by using the gh pr create command and follow the on-screen prompts. You can also specify the title and body as arguments to the command.



gh pr create -t "pull request title" -b "pull request body content" -a "assignee"

4. Review pull requests 👀

Run gh pr create and add the necessary flags: -t for the title, -b for the body -a for the assignee and -d to mark the pr as draft. You can also use gh pr create and follow the on screen prompts to create a new PR with the cli.
Once the PR is created, take note of the PR num and run gh pr view <num> to see the details of the pr in your terminal.

5. View action workflows 👀

We can view the action workflows right from the command line with the gh cli. This command will list all the recent action workflows for the repository.



gh run list

You can also run workflows right from your terminal with the command gh workflow run <workflow name>. This command will run the workflow and prompt you for any required inputs.



gh workflow run <workflow filename>

6. Open your current Repo in the browser 💻

You can open your current repository in the browser right from your terminal. This command will take you to the homepage of your repository.



gh browse

If you you attach the -b option, you can open the current branch in the browser, like so:



gh browse -b "branch name"

7. Create and update Project Boards 📋

Creating a project from your terminal with the command line requires you to have project scope permissions. To update your auth token with project scope permissions, run the following command and follow the on screen prompts:



gh auth refresh -s project,read:project

Once that's confirmed, you can create a project board with the command gh project create and follow the on screen prompts.



gh project create --owner "owner" --title "title"

8. Create edit, and view Releases 🚀

You can create a release with the command gh release create and follow the on screen prompts.



gh release create --title "title" --notes "notes"

You can also edit, dowload assets and delete relases with the gh cli. You can find more information about this here.

9. Start and Stop Codespaces 💻

A Codespace is an online development environment. You can start a codespace with the command gh codespace create and follow the on screen prompts. You can also do it by specifying certain flags, for example, the branch you want to use:



gh codespace create -b "branch"

You can then open up the codespace in VSCode or Jupyter Notebooks with the command gh codespace code [flags] or gh codespace jupyter [flags], respectively, where the flags can include the version of vscode, repo, repo owner and name of the codespace you want to open.

10. Browse and create gh cli extensions 📦

You can run gh extension browse in your terminal to view a list of all available cli extensions to make your terminal experience even better. You can also install ann extension directly from the browsing interface.

To install an extension you like, run gh extension install <repository> [flags] and follow the on screen prompts.
You can also create your own cli extensions with the command gh ext create and follow the on screen prompts.

Getting Help with the gh cli ❓

If you're ever unsure of what to do with the GitHub CLI or how to use it, you can always run gh --help and it will give you a list of all the commands, help topics, extensions, flags and examples you can run with the GitHub CLI.

GitHub Copilot for CLI 🤖

Ever forget a command and have to search for it over and over again? Yea, me too. This is why we built GitHub Copilot for CLI. It provides a chat-like interface in your terminal to help you find commands for GitHub CLI, shell commands and git.

I'll be showing you how to use GitHub Copilot for CLI and how amazing it is at taking your command line skills to the next level. Join me at GitHub Universe next month to see it in action.

And if you can't make it in person, virtual passes are free and open to everyone. Register here.

If you have any questions about the gh cli, feel free to post them below and I'll do my best to answer them. 😄

From TSV to CSV: How GitHub Copilot Chat Made Data Conversion a Breeze!

Kedasha — Tue, 29 Aug 2023 17:44:01 +0000

I recently discovered that GitHub projects has TSV exports! 💃🏼

I was super excited about this but it also made me realize that some folks may want/prefer csv files and so, I asked by bestie - GitHub Copilot Chat - to help me convert a TSV file to a CSV file in Python.
Now, I'm not a Python Developer, so I wasn't sure how I would do this in Python, and while I could use an npm module like PapaParse to accomplish this, I wanted to try another language.

Honestly, it was pretty seamless!

Here's how I did it:

Step 1: Download your TSV file and add it to a folder, then open said folder in VSCode (or your preferred editor).

Step 2: Create a new Python file, I called mine tsv_to_csv.py.

Step 3: Navigate to GitHub Copilot Chat and ask "how do I convert a tsv file to csv with python?"
I was provided with the following code:



import csv

with open('ruby-study-guide.tsv', 'r') as tsvfile:
    reader = csv.reader(tsvfile, delimiter='\t')
    with open('ruby-study-guide.csv', 'w', newline='') as csvfile:
        writer = csv.writer(csvfile)
        for row in reader:
            writer.writerow(row)
    print("Conversion complete!")

If you're not familiar with GitHub Copilot Chat, read this blog post to learn how to get started with it.

Step 4: Paste the code into your file by clicking the "Insert at Cursor" icon.

Step 5: Save your file, run the code with python <<name of your file here>>, for example python tsv_to_csv.py, and enjoy your csv file!

Here's a video/gif on the steps I did above:

Now that you know that GitHub Projects has TSV exports, learn more tidbits that you possibly didn't know by reading this blog post I wrote on 10 things you didn't know you could do with GitHub Projects.

Let me know if you have any questions about GitHub Copilot or GitHub Projects below!

Getting Started with GitHub Copilot Chat in VSCode

Kedasha — Mon, 21 Aug 2023 18:40:33 +0000

GitHub Copilot Chat is an extension that works in your Code Editor or IDE (VS Code or Visual Studio currently) that allows you to have conversations with GitHub Copilot right from your editor.You're able to get code suggestions, build, debug and tests applications with an AI model that understands natural language right from your editor.

Getting started

To get started using GitHub Copilot Chat in VSCode, ensure that you have access to the extension by checking your email for access privileges. You get access privileges through your organization or by being taken off the waitlist for the private beta for individuals. You also need to ensure that you have an active GitHub Copilot subscription.

To verify your access, click on your GitHub profile and select "Try Copilot." If you have access to GitHub Copilot, you will see a message at the top of your settings page indicating such and if you don't have access, you will be routed to another page to start your 30-day free trial.

You have access to GitHub Copilot

You do not have access to GitHub Copilot

Installing GitHub Copilot Chat & asking your first question

Once you verify those details, follow the steps below to start using GitHub Copilot Chat:

Search for "GitHub Copilot Chat" in the vscode extension marketplace, and click the blue install button. After the installation, you'll be prompted to login to your GitHub Account to validate your access to copilot chat.

Be sure to sign in to the GitHub account that has access to your GitHub Copilot subscription and also that has access to Copilot Chat.

Once you've installed the extension, you're ready to use it! You'll see a message/chat icon appear on the side of your code editor, click that icon and start asking programming related questions.

You can start by asking a simple question like "How do I set up a new ruby project?" and it will provide you with a suggestion.

And that's how you start using copilot chat in VSCode.

Slash Commands + GitHub Copilot Chat

There are some really amazing slash commands that you have access to as well that makes it easier to ask for help, create vscode extensions or create unit tests. To start using Slash Commands with GitHub Copilot, type / in the input box and you'll see the multiple options come up.

Choose one of the options and then ask a question to get a suggestion.

Asking GitHub Copilot Chat questions about your code

To ask Copilot Chat specific questions about the code you're working on, open up the file in your editor, and navigate to the chat extension.

Let's say for example this is your first time interacting with this particular repository, and you don't quite understand the code, you can ask copilot chat "what does this file mean?" and it will provide a suggested explanation of what the code is doing.

Using Code Suggestions from GitHub Copilot Chat

Whenever you ask copilot chat for code suggestions, you can accept them by copying the suggestion from the chat interface or clicking the "Insert at Cursor" button. You can also insert the suggestion into a new file, or run the code in the terminal if applicable.

I was able to use GitHub Copilot Chat to tell me how to insert two images side by side using markdown ☺️ I copied to provided code suggestion and pasted it here and voila! I had two images side-by-side. 💃🏽

Opening a Copilot Chat session in your editor

To make your life a little easier, we've made it possible to open up a copilot chat session right in your editor, so you don't have to go back and forth between the chat interface and your files.

To do this, click on the chat icon on the side, and click the three dots at the top and select "Open session in editor" this will move the chat into your editor, similar to how a file would be open.

If you have any issues, questions, comments or concerns, feel free to leave them below and I'll try as best as possible to get you an answer! I hope you enjoy using GitHub copilot Chat as much as I do!

Happy Coding!

How to Enable 2FA for your GitHub Account

Kedasha — Mon, 27 Feb 2023 22:31:21 +0000

Two-factor authentication (2FA) is an extra layer of security used when logging into websites or apps. It requires you to log in with your username, password, and provide an additional form of authentication that is uniquely accessed by you.

By the end of 2023, 2FA will be mandatory for all GitHub accounts - both on web and mobile (iOS and Android). Let’s get a head start on the process by enabling at least 1 form of 2FA on our GitHub accounts.

In your account, you have the option to choose one of the following methods for 2FA:

Authenticator App (TOTP app)
Text/SMS Authentication
GitHub Mobile App
Security Codes

I strongly recommend using an authenticator app, such as Microsoft Authenticator or Authy, to configure 2FA for your account as they are more reliable that SMS/text message based authentication.

Let's walkthrough configuring 2FA on web and mobile using an authenticator app.

Configuring 2FA on the Web

To enable 2FA on GitHub's webpp, follow these steps:

Download an authenticator app. I'm using Microsoft Authenticator.
In your GitHub account, go to Settings
Under Access, select Password and authentication
You will then see an option to enable 2FA, click the green button, "Enable two-factor authentication"
This will take you to a page to setup your authenticator app. Use the authenticator app you downloaded to scan the QR code on the screen.

Once you scan the QR code with the app, enter the generated code in the highlighted text box and click Continue.
The next screen will ask you to download your recovery codes. Click the green download button and save these codes in a password manager - I love Bitwarden.

After you've downloaded and saved your codes, click the green button "I have saved my recovery codes" to confirm your download and complete enabling 2FA on your account.
And you're done! You've successfully enabled 2FA on your account.

You can add a backup 2FA method for your account just in case using a security code (such as a touch ID on mac or a yubikey) or the GitHub mobile app.

Configuring 2FA on Mobile

Getting 2FA enabled on your mobile device is very similar to what we just did on the web app. I'll be using an android device.

Go to the play store on Android or iOS App Store, search for the GitHub Mobile app install it (Im using an Android device).
Once installed, you'll be prompted to sign into your account with your username and password.
Since we just enabled 2FA on the webapp in our account, we'll be prompted to engter the authentication code from the authenticator app we previously downloaded. Enter the code and click Verify
Onced verified, you'll be prompted to authorize your GitHubHub mobile app with your GitHub account. Select Authorize github to continue
Once authorized, you'll be rerouted to the Home screen on mobile.

Wrapup

And that's it! You've successfully enabled 2FA on GitHub web and mobile. Now GitHub your account is more secure. 🔐

If you ever lose access to your 2FA credentials, you can use your recovery codes to gain access to your GitHub account. Learn how to use your recovery codes by reading the GitHub docs on Recovering your account if you lose your 2FA credentials. You can also learn more about 2FA by reading the docs on GitHub.

Codespaces Tips: How to Create Pull Requests Without Leaving Your Code Editor

Kedasha — Wed, 25 Jan 2023 17:17:16 +0000

Codespaces has been a game changer for the way I work with repos on GitHub. For example, I no longer clone repositories to my local machine since there isn’t a need to do so anymore with Codespaces in the browser.

Last week, while pushing code to update an internal workflow, Codespaces asked me an interesting question. Once I hit enter after my regular git push, a notification popped up asking if I wanted to open a PR in Codespaces with GitHub’s Pull Request and Issues extension.

Usually, I would completely ignore these pop up messages but I was intrigued enough to select “yes” and what happened next completely blew my mind - ok, ok, slight exaggeration but I was impressed!

This beautifully intuitive UI popped up on my side panel showing me my branch, the main branch that I need to merge my changes into, a title box with my commit message and a description section for my PR needs. There was even an option to create the PR as a draft, see below:

At this point, I was not only intrigued, I was also impressed! You’re telling me that I don’t need to exit my code editor (codespace) to create a PR?!

I continued creating the PR by hitting Create, which triggered a pop-up asking if I wanted to publish my changes in a new upstream branch and then create the PR.

I obliged, clicked Publish Branch, and my PR was created with another pop-up asking if I wanted to periodically run git fetch. Color me impressed by this extension. Color me impressed.

But, just in case you don’t get that pop-up asking if you want to open a PR, to create a PR in Codespaces, click on the GitHub icon in the editor. This will pull up a side panel option that’s like a Pull Request Dashboard. Once there, click Create Pull Request as shown below, and create your PR!

If you’ve used the extension before in VSCode, you’d already know how to do this and it works the same in Codespaces. What I did appreciate is that this extension built into Codespaces was intelligent enough to ask me if I wanted to use it.

I’m super glad I clicked yes on that pop-up notification instead of ignoring it. It just goes to show that sometimes, pop-ups can be useful.

Have you tried this extension before? Or Codespaces? Let me know your thoughts in the comments and give GitHub Pull Request and Issues extensions a try!

Happy Coding!

The role software developers play in the cybersecurity space

Kedasha — Thu, 10 Mar 2022 19:54:49 +0000

Cybersecurity is the intentional practice of securing networks, data, and devices from unauthorized users. With the growing increase in cyberattacks and security vulnerabilities (most recently, the Log4j vulnerability), it’s becoming increasingly important for developers to understand how to secure applications and think like cybercriminals to prevent these attacks from severely harming customers.

As software developers, we have a unique opportunity to be one of the first lines of defense against cyberattacks. The software development lifecycle focuses on implementing core functionality in software and applications; code quality and security is often an afterthought. However, our understanding of core cybersecurity principles can make or break the applications that we build.

Since 2003, the Open Web Application Security Project (OWASP) has highlighted the 10 most critical security risks to web applications via OWASP Top 10 list. This is “globally recognized by developers as the first step towards more secure coding.” However, some of the same vulnerabilities appear year over year with little to no improvements on the quality of software code being deployed to production environments.

Organizations deem it unacceptable for software teams to knowingly ship products with functional defects. It is time for organizations to also find it unacceptable to ship products with security defects. Security starts and ends with us - the developers. Unless a software team intentionally focuses on code quality and security, vulnerabilities will find their way into shipped products, and cybercriminals will exploit those vulnerabilities.

As developers, we can help prevent cyberattacks by proactively implementing security controls in our code. We can accomplish this in the following ways:

Take the time to resolve high-severity alerts by keeping packages and dependencies updated. We can use tools like Renovate and WhiteSource to automatically scan for updates.
Identify and understand the typical vulnerabilities for our tech stacks. We can use a tool such as the CVE to search for vulnerabilities in the software that we use.
Test for what our code is and is not meant to do.
Upskill by taking security courses on platforms such as Udemy. For example, this course teaches cybersecurity for developers.
Understand how to perform security testing such as vulnerability scanning and penetration tests. There are many tools available to automate this.
Review the OWASP Top 10 list to understand the most common security attacks and how to prevent them.
At the very least, be familiar with the following three attacks:
- Broken access control
- Cryptographic failures
- Injection attacks

We control the security of the software and applications we build with code. Understanding cybersecurity is important because it protects the users and intellectual property of the companies that we work for.

Come learn security with me in the Lacework Community, where I’ll cover fundamental security topics for software developers.

From Social Worker to Software Engineer 💃

Kedasha — Mon, 07 Oct 2019 03:45:28 +0000

Six months ago I began my journey into the world of software engineering by joining the Full-Stack Web Development & Computer Science program at Lambda School. The last time I wrote a blog post, I wrote from the perspective of a part-time student. So much has changed since then so I figured it was time to write an updated post. 📒

Transitioning to the Full-time cohort 💻

At the end of May 2019, I decided to transition to the full-time cohort to accelerate my progress and learning in the program. I was working up to 35hrs per week in addition to coding for at least 8hrs per day. Needless to say it was quite stressful and overwhelming. 😩 😩 😩

In August, and after speaking with my partner, I left my full-time job, started working as a part-time Team Section Lead at Lambda School and kept pushing through the program. August & September were the two most trying times. Apart from the stress & shock of learning Reactjs & Redux back in June, participating in the 2 months in-house apprenticeship (labs) was maddening - primarily because I was coding all day with my team then had to switch to managing students & team leads in the part time cohort. My day started at 8am and ended at midnight most days. It was (and still is) a lot of work. 💤 😫

By the time labs ended, I was completely burnt out. 😩 So I asked for a short break from school. I wanted to be in the right mindset before going into the hardest part of the curriculum - computer science fundamentals. Additionally, I wanted to give myself time to enjoy the Grace Hopper Conference which was immediately after completing labs. 💃

Grace Hopper Celebration 🎉

Sometime this year I received a registration scholarship from AnitaB to attend Grace Hopper. This was my very first time attending a tech conference and I was determined to make connections and engage with as many employers as possible. 😀

While I was extremely surprised at first to see how underrepresented WOC were at the conference, I made it my goal to present myself as best as possible and tried to standout among the large computer science crowd. I used the wonderful interpersonal skills I gained from working in the field of Social Work, Non-Profit Organizations & Government, to connect with recruiters on an individual level and create meaningful conversations. 💬

I spoke about my journey, my former career and my current tech stack and was surprised at how receptive most employers were to my story. This increased my confidence and allowed me to shine even brighter when I was speaking with the Engineering Director at Northwestern Mutual's (NM) booth. 🏦

I'm happy to say that I was hired by NM as an Associate Engineer and will begin that journey in January 2020 after completing Lambda School in December. 😲 I still cannot believe this happened because my goal was to work in the fintech space post graduation. 😲 😀

The Journey Ahead ✈️

As I continue my learning at Lambda and prepare to move across the country with my partner, I'm simply amazed at how much I have learned and how much I am yet to discover. Discovering code in December 2018 has been a transformative experience. ✨ ✨

As a Jr Software Engineer, I am so excited to see what lies ahead. 😀

My First Three Weeks at Lambda School as a PT Student

Kedasha — Sat, 06 Apr 2019 20:36:26 +0000

This week I completed my first three weeks in Lambda School's part-time Full-Stack Web Engineering and Computer Science bootcamp. As the program is 18 months long, I have quite the way to go.

I wanted to take the time to write about what the part-time program is like for anyone considering enrolling in a PT cohort. I hope my insight will give you an idea of what to expect upon enrolling.

Why I decided to attend Lambda School

Five months ago I had no idea what web development was. At the time I was searching for a more robust career trajectory and came across the world of software engineering and web development. After teaching myself HTML5/CSS3 on codecademy I decided in early January that I wanted to make this my new career path.

My intention was to use as many free resources and Udemy to further my education, build as many projects as possible and apply for a job within 9 months. It was all planned + going quite well until in late January I came across JavaScript and hit a mental roadblock. I realized that I needed a more structured learning path to assist me in my learning, and decided to apply to a bootcamp to accelerate && support my progress. Enter: Lambda School.

The First Three Weeks

My cohort started on March 18th and I am honestly amazed at what I've been able to accomplish in the past three weeks as a part time student. For the first portion of the curriculum we focus on web development fundamentals. In 3 short weeks I have solidified my knowledge in semantic html5, CSS3, responsive web design and accessibility.

During the first week I learned about Git and the command line and I must say that I am starting to get quite comfortable with using these essential web developer tools. I have created pull requests, cloned repositories, create branches in existing repos and have started to create really amazing relationships with students in my cohort. I also created my first portfolio website! How exciting!

The Next Three Weeks

In the next three weeks I will have a solid understanding of pre-processing and start learning more about fundamental Javascript concepts. I will continue to do more advanced projects and continue to build relationships with the amazing students I am learning with.

If you would like to see what I have done so far, see my Github repos here.

Until next time, thanks for reading! :)

DEV Community: Kedasha

The GitHub Community Roundup: 03-08-24

What's new at GitHub this week?

How I Bulk Closed 1000+ GitHub Issues with GitHub Actions 🚀

Creating the GitHub Action

Testing the Action

Updates

Conclusion

Getting Started with GitHub Copilot in the CLI🚀

Installation

Available Commands

Create an Alias 💡

Explain Command 🤖

Suggest Command 🤖

Revising Suggested Commands

Explaining Suggested Commands

Copying Suggested Commands

Rating Commands 🌟🌟🌟🌟🌟

Exiting the prompt flow window 🚪

Examples of using Copilot in the CLI

1. Explain a dangerous command ⚠️

2. Delete a git branch locally and remotely

View all open issues in a repo

3. Kill processes with open files that have been deleted

4. Exit Vim 😅

Conclusion

How I used GitHub Copilot Chat to debug my deployment workflow 🐥🤖

Deploying my slides

What is GitHub Actions

1. I didn't enable Github Pages

2. The installation of package.json failed

3. Slidev build command failed

4. The post build command failed - no dist folder found

5. Lessons learned

Top 10 Tips for using GitHub from the Command Line

Installing the GitHub CLI 🛠️

CLI Syntax and Usage 📝

1. Create a new repository and clone it locally 🎉

2. Create issues 📝

3. Create pull requests 🔄

4. Review pull requests 👀

5. View action workflows 👀

6. Open your current Repo in the browser 💻

7. Create and update Project Boards 📋

8. Create edit, and view Releases 🚀

9. Start and Stop Codespaces 💻

10. Browse and create gh cli extensions 📦

Getting Help with the gh cli ❓

GitHub Copilot for CLI 🤖

From TSV to CSV: How GitHub Copilot Chat Made Data Conversion a Breeze!

Getting Started with GitHub Copilot Chat in VSCode

Getting started

Installing GitHub Copilot Chat & asking your first question

Slash Commands + GitHub Copilot Chat

Asking GitHub Copilot Chat questions about your code

Using Code Suggestions from GitHub Copilot Chat

Opening a Copilot Chat session in your editor

How to Enable 2FA for your GitHub Account

Configuring 2FA on the Web

Configuring 2FA on Mobile

Wrapup

Codespaces Tips: How to Create Pull Requests Without Leaving Your Code Editor

The role software developers play in the cybersecurity space

From Social Worker to Software Engineer 💃

Transitioning to the Full-time cohort 💻

Grace Hopper Celebration 🎉

The Journey Ahead ✈️

My First Three Weeks at Lambda School as a PT Student

Why I decided to attend Lambda School

The First Three Weeks

The Next Three Weeks

2. The installation of `package.json` failed

4. The post build command failed - no `dist` folder found