StrawberryPuding

I improved the OpenAPI docs (added actual help messages, example values etc+improved type annotation) and deployed it! https://queue-api.strawberry.hackclub.app/docs

I improved cookie management and fought with the LWPCookieJar for a bit. it can't load more than 1 cookie for some reason? is it stupid?
I also added a function for updating project info, including the screenshot (always remember to set the file type!)

(sorry for long devlog, i've mostly been working on this for siege)
I set up cookie management for auth-protected sites. Turns out Siege is pretty secure, and applies trickery such as rotating session cookies, so I had to work with that and save all the current cookie.
I also implemented a function that parses the projects/Armory page for info about each of your projects.

I added sanitisation to all my database queries (silly me forgot before and just did f strings :p ). I also added endpoints for getting the length of a queue, and changing the read and write keys of a queue using its delete key.

I wired up my BMS IC (bq25622e) and its battery connector. i'm a bit worried that the 2x2 connector might take up quite a lot of space, but at worst i can use smaller headers to make wiring easier later.

(test devlog for SoM ship showcase) I considered the potential of users and reviewers alike not reading the conveniently named README,
so I added some checks that tell the user to fix their session.json if it is missing or broken

I added argparse support, the tool now takes project ID+devlog image+devlog text (or a readable file). well, let's test it :P

Added a controls hint, score counter and collision physics for the target. I'm probably going to ship this for SoM now and improve it (read: make sure I can play this without my eyes exploding) for Siege later. 7 hours of js, and instead of getting better at the language, i got better at hating it :pf:

Added a visual aiming indicator and simple physics to the projectile

(this is a test devlog) if you're seeing this, i fiddled around with multipart forms enough to get this devlog submitted

(this is a test devlog) if you're seeing this, i fiddled around with multipart forms enough to get this devlog submitted

Made the functions that write my own multipart form data (apparently, requests can do that itself? woah,,,), but the server was always returning 400 (Bad Request). I found a way to probably fix this. posting this devlog so that the next one can be testing

All this beautifulsoup work just to get the csrf token pf
I also got cookies kind of set up with a separate session.json file so that i don't leak my session cookies :p

(worked for more than this time, but kicad-wakatime didn't catch that)
I wired up USB PD in my schematic, with some ESD protection components so that the devboard doesn't get fried

added PUT and DELETE endpoints and tested the whole thing. read, write and delete passwords all work!

Got the FastAPI server working! sqlite was annoying me about database locks, so I had to rework my code a bit. with the basic POST and GET methods done, I'll add queue creation and deletion to the API next

I completed all the functions I needed for queue manipulation. these respectively create/drop a queue database and add it to/remove it from the queue_manager structure

I added the queue database structure, wrote functions which interact with it and added support for multiple queue databases+another database to keep track of them. if endpointer wants me to use a database, i will sure as hell do that!

I made a script that makes samples from prettier functions like sin and cos. Used it to make better demos and a thumbnail from this project in a feeble effort to get better voting performance :P

Reworked the csv loading process to support x,y pairs from the csv. This allows points to be drawn in the output, rather than just bars. I had to fix a lot of things like Y-axis labeling errors caused by DRY not working well :P

I routed the entire PCB, added silkscreen and made this 3D render! (took a lot more than just 2 hours, but kicad-wakatime crashed)

After some research on resistor values, I started placing down my component footprints, hopefully it will fit into the size limit

(sorry for long devlog, I was sprinting this for TC and forgot that SoM exists)
I added the entire function plotting feature! Since TC kind of discouraged packages by the self-containment requirement, I wrote my own math statement parser, which then finds the value of each sample. The resulting data is written to csv instead of just being passed to graph rendering, so it can be used as data input for other programs.