@ArHIM

READY TO SHIP.

I cleaned up minor UI confusion and made it clear for anyone who opens the extension to understand what’s happening and start using it instantly.

Also removed friction (unnecesary settings user needed to decide) so someone can instantly go into focus session.

Lastly, made the schedule screen easy to pick. At first glance, all know whats happening.

I foolishly didn’t make a devlog in a while becuase I thought that I couldn’t while my project was being reviewed. Sorry.

But now I’ll try to summarize the changes made, and there were a LOT LOT:

From what I recall:

• Support multiple servers

the bot will automatically join and run in any discord server immediately. (before I needed to manually add the server ID)
Server admins can use “/setpollchannel” to specify where they want the polls to go.
Servers can individually set if they want polls to @everyone or a specific role or none (before it was set to @everyone which was not a good thing to do).

• More consistent daily polls

Moved to newer free AI models for generating the questions for the polls.
Automatically retry the AI service if it is too slow or has an error. (before it quit fast resulting in lots of errors.)
Auto-cut answers to fit within the <55 character limit for buttons on discord. (avoiding failure)
If there are issues posting a scheduled morning poll to one server, it won’t block it from posting other scheduled polls for other servers.
also the same poll goes to all servers instead of a seperate poll being generated for each one.

• Command changes

Merged some overlapping commands for a simpler command list.
MAJOR BUG: “/knowledge” command could potentially overwrite/destroy the existing bot settings. Resolved that issue.
Updated /help

• Other

Automated tests were tweaked to increase performance.
Upgraded project deps

ALMOST READY TO SHIP LETS GOO!

I made a pre-built zip file to get it ready to ship. I also compiled the extension files for popup UI, and background worker.

Then ran the package script to put everything into a neat zip file called focus-blocker-v0.2.0.zip.

The zip contains only the actual extension files and icons. So no one needs to download the whole thing or install node to try it out.

So I pushed the code. Created a 0.2.0 version tag and published release to github with the zip file. And added simple instructions to README.

To try it, just download the zip file. Unzip it.

Then go to chrome://extensions.

Activate Developer mode. And click on Load unpacked.

And that’s it! Try it out!

Holy efficiency!

Primary changes were to revamp how blocking of windows apps works. It used to poll open windows periodically - so there would be a delay of a few seconds before closing a blocked app. Now the background service just gets a process creation event. If a blocked app tries to open, the app detection works right away to close it.

Made a screen for the blocked app in the desktop ui. Added a Back to Focus button. Also added some debouncing to not get popup’d too much if an app tries to reopen itself.

Fixed a number of corner cases for apps that were getting through. For example, some apps/games can be launched through another launcher, so added some checks to find the actual program binary. Did some trimming of the windows path prefixes. Fixed some edge cases with really long file paths. Removed the background toast notifications (which were silently failing, b/c background services on windows operate in Session 0)

In developer area, I grouped all the startup/test commands into start.bat and test.bat as two batch files. Got the desktop test suite a bit faster by switching from Vitest to happy-dom. Added some mocked data so we can test the frontend without the bg service running. Lastly but not least, cleaned up dead network filter code.

More README.md edits. And, I tried making testing the product easier/more accessible. Either just watch the demo video, or join the servere in which you can use the bot freely! Both in the project desc.

PLEASE ALLOW MY SHIP, SHIPWRITERSS 🥹🥹

okay, finally ready to ship.

• made readme + worked on it cleaned it made it nice etc. big deal

• polished small changes or things that bugged me

• made sure it met all project reqs.

and now, I submit

It actually workssss lets goooo!!!

(But the thing is turns out hackatime wasn’t counting time for the most of the time I worked. BROOOOO. Anyways, I figured and fixed it but most of the work was done by then… )

After a good amount of working, exploring, and js having fun, it works.

The site uses the api to get today’s picture + title + explanation.

I also added a date picker to look at past ones (but the user interactions for that is not the best :( WILL WORK ON IT THO!). made the max date to today so people cant pick future dates.

The hardest part was different data formats. NASA posts a yt video instead of an image. It broke so I needed to fix it. I added a check to check for that. Then it matches and works w the appropriate type of image/vid/yt.

Then I spent some time on the styling. Following the guide, I made it dark purple theme, added fonts, made these zig zag kinda neon pink borders on the sides. On desktop/website it looks great. Phone… i have no idea.

It all works. Finally! Its ready to ship after I add a quick readme. But I’ll surely make some changes after that.

Im heavily following the guide, so this might seem like boring. But im leaning and making progres…

Quick one:

• cleaned out default template files and icons that came with Vite.

• I also set up GitHub Actions. it auto-deploys to GitHub Pages whenever I push to main which is pretty cool tbh.

Now we actually start writing code. cant wait! or can i?

This is the start of my project for the ‘Give Your Website a Pulse’ mission. The mission itself seems very interesting to me tbh.

I set it up using vite and vanilla JavaScript.

Also, created Git repo and made first commit.

It works locally!! Yayyy

I also got the NASA api and now Im looking forward to building this lets gooo

Turns out my readme SOUNDED AI generated.

I changed up the whole format and structure. Idk if that made it easier or harder to read but now I ship again!

Finally, we are DONE

Actually added color to make it nice to read. And changed layout
Edited text a bit again because it can always be better.
Added more contact info and just info in general.
Cleaned up code!
Also added an IMAGE of me (ai generated bc idk)

Even tho it is VERY simple right now, I think its ok to submit. I will, of course, keep working on it in the future and keep refining it over time

This was relatively quick, thanks to the guides.
Plus stardance motivated me a lot to complete and work on this. LETS GOOOO

Added more personal information about my life.

like, projects, interests, what i do in free time

• changed text a bit to be more FUN to read…

Started the site!

Just started the site, laying out foundations and creating first section with basic info (according to guide)

I have high hopes…

Longggg session today.
shipped a lot of cool features tho.

1. made multi-server management much easy just in case this blows up (fingers crossed)

2. toggled the AI prompt for generating summary a bit cuz we all hate niche ai outputs

3. cleaned up docs

4. improved messages and templates it uses to send discord messages. made it simpler. otherwise, it was all accross the place.

Code quality matters much more than people realize. This session was all about clearing up the code. Merging duplicate parts, unifying systems, and making it just easier to read and understand overall. Also, TESTS help a lot. Run a suite of tests after every change. Keep adding tests for new changes too!