The hovercraft can now complete tasks by hovering over a task marker until the progress bar fills up.

Next up: Keeping track of what's on the hovercraft.

Note: Due to complicated reasons, my screenshots will consist of me taking pictures of my screen with my phone.

Started coding tonight, challenge registration today.

As I need a project simple enough to be finished by the end of the week, I'll try to make a platformer as fun as possible.

Due to the fact that this was our least favorite theme for the contest and we had a hard time coming up with an interesting idea, we were off to a slow start. After a while we finally managed to outline a concept that we all thought was doable and would be fun to play in the end.
As we lost a lot of time struggling there we only started development on Monday.

Yet we are making good progress at this point and currently we are able to load a tile map, have implemented a (hopefully useful) architecture and the player is able to move. 

So while we started off slow we now start picking up momentum...

We're still getting our core gameplay together. Meanwhile the art workflow seems be working pretty well and we're seeing all kinds of fun character art showing up. This is the first time we've tried to incorporate automated testing into our development for PyWeek and I'm really glad we did because our game model is showing up all kinds of edge cases and it would be very hard to be sure it is working otherwise.

Anyway, more details coming soon. In the meantime if you know what we drew here then please let us know!

"Cops" was my #5 theme choice. I'm not inclined to make a game about police right now.

"Copse" is a homophone that means a small cluster of trees, so that's what I'll make a game about! I'm expecting the possibility of comments saying I used the theme wrong, but I'll live. As long as it's clear I didn't ignore the theme. Here's some stock image inspiration:

I know I'm supposed to make a game, but I've spent my entire afternoon trying to make my brain think about aspect ratios.

Every engine I've used so far had some sort of automatic scaling for when the user resized their window. But, python? Yeah, forget about any kind of help.

This was a pain. There was absolutely 0 documentation about how to implement this on Arcade, which is the module I'm using. I want my game to keep it's aspect ratio no matter how the user resizes the window, by adding black bars to the sides. It took me 45 minutes for my brain to use 5% of it's processing power and realize this:

(orig height/orig width) = (new height/new width)
Therefore
(orig height/orig width) * new width = new height
(orig width/orig height) * new height = new width

So simple, but my small brain kept searching for complicated equations, algorithms and formulas that were just not necessary.

Afterwards, I spend a good time looking at the docs. That set_viewport function wasn't even that complicated. But my brain refuses to believe things can be simple. I'm kind of contradicting myself now, aren't I?

That was not the end of it. This is the code I made:

if width <= height:
    # Width is smaller or equal
    r_width = width
    s_width = 0
    r_height= math.ceil((500/800)*width)
    s_height = math.floor((height - r_height)/2)
    self.set_viewport(-s_width, r_width+s_width, -s_height, r_height+s_height)
    print(f"\nResized to {width}, {height}.\nWidth is smaller.\nResult is {width}, {(500/800)*width}\n")
else:
    # Height is smaller
    print(f"\nResized to {width}, {height}.\nHeight is smaller.\nResult is {(800/500)*height}, {height}\n")
    r_width = math.ceil((800/500)*height)
    s_width = math.floor((width - r_width)/2)
    r_height= height
    s_height = 0
    self.set_viewport(-s_width, r_width+s_width, -s_height, r_height+s_height)

It basically tries to make the content inside as big as possible within the space available in the window. But since I'm not accounting for aspect ratios in the conditions, it doesn't detect when it has to switch modes very well. Also, making the width smaller breaks it and doesn't correctly center or scale the content. Here's a picture to embarrass myself.

TL;DR: Two days and I can't even handle aspect ratios. I better calm down and lower expectatives. Good luck to everyone!

The Engine for the NPCs is there now:
    - it can move,
    - hit the wall ( and react),
    - has some looks to choose from,
    - can change the looks depending on the state ( standing, walking, ..),
    
    There are some managing classes, to take care of some NPC stuff... ( like images to be loaded, size ratios, animation frames, position, wall detection,....):
        - the "Cloth" class // for all the clothes of an NPC
        - the "walker" class // for walking throu the game

There is a beginning of a manager (that is there for):
    - NPCs don't just walk into each other ( in some way ..),
    - NPCs will be stoppable from outside (= the user),

There is a Vector class... // we made it rather simple... not much to say...