py5 format error at properties: py5.height and py5.width

[claudefalbriard]

I'm new to the py5 project effort, so first a welcome to all members.
My environment is: Linux Ubuntu 18, Anaconda, Python 3.8.5 and Jupyter
When calling py5.height, or py5.width in a sketch, I get a method in place of the expected numeric value. The data returned is:
<bound method Sketch._get_height of <py5.sketch.Sketch object at 0x7f41722c98e0>>
When calling it as a method, like py5.height() its going well, or by using the attributes "height", or "width" with no prefix, also works OK, as expected. Wonder if this is a planned behavior for the two properties.

[hx2A]

Welcome, @claudefalbriard !

Can I see your code? Also, how are you running this? Are you running it using the py5 kernel or using run_sketch?

Are you coding in "imported mode" or "module mode"? It is not a good idea to mix the the two together in the same sketch. I am not sure what the result might be but I suspect it might cause those symptoms.

[hx2A]

BTW, if you like you can put the call to size() in the beginning of setup(), before the other function calls. You don't need settings() anymore. This is similar to programming in Java in the Processing IDE.

[hx2A]

Also, I read this line in your code:

particles = []

and am curious what you are working on. This might not be relevant, but there is a new points() method that lets you draw multiple points instead of calling point() repeatedly in a loop.

[claudefalbriard]

Thanks so much for the detailed clarification.

I really was unaware about the details and differentiated use of the Jupyter P5 kernel and import modes. The use of the P5 kernel simplifies and shortens the code. There might exist more complex code samples where we need to add extensions by Python 3 classes, like the "Particles" and justifies the use of module mode and standard Python 3 kernels, addressing classes and functions. The conflict with the size parameter and use of settings() function had the same origin.

I will keep this discussion in my Jupyter sample code for future reference.

[hx2A]

Sounds great. When you are using the py5 kernel you can import any other Python library you want. It really is just the IPython kernel with a few small customizations.

I'll see what I can do about adding a warning when users execute import py5 from within the py5 kernel. That might be possible, and will help others who run into this same issue.

[hx2A]

I'll see what I can do about adding a warning when users execute import py5 from within the py5 kernel. That might be possible, and will help others who run into this same issue.

Done. Now the py5 kernel (and py5bot, and the run_sketch terminal command) will reject code that tries to execute import py5. Users will see a message like this:

"import py5" found on line 1. Do not import the py5 library, as this has already been done for you. Your code should be written without any "py5." prefixes.