[p5.js 2.0 RFC Proposal]: expand keyIsDown() to work with characters as arguments

[kjhollen]

Increasing access

Having some symmetry where you can also do:

keyIsDown('a');

would be useful for teaching beginners.

Which types of changes would be made?

• Breaking change (Add-on libraries or sketches will work differently even if their code stays the same.)
• Systemic change (Many features or contributor workflows will be affected.)
• Overdue change (Modifications will be made that have been desirable for a long time.)
• Unsure (The community can help to determine the type of change.)

Most appropriate sub-area of p5.js?

• Accessibility
• Color
• Core/Environment/Rendering
• Data
• DOM
• Events
• Image
• IO
• Math
• Typography
• Utilities
• WebGL
• Build process
• Unit testing
• Internationalization
• Friendly errors
• Other (specify if possible)

What's the problem?

keyIsDown() currently only works with keycodes. Sometimes this is confusing for my students, because they can use this for things like:

keyIsDown(LEFT_ARROW);
keyIsDown(ENTER);

but if they want to check if an alphanumeric key is down, the code is different:

keyIsPressed && key == 'a'

What's the solution?

keyIsDown() would need an addition that checks the type of the argument and allows passing text/characters. This means p5.js would have some symmetry where you can also do:

keyIsDown('a');

Pros (updated based on community comments)

Example list:

• Consistency: This proposal increases API consistency by giving users a way to enter the name of an alphanumeric key for keyIsDown();
• can use keyIsDown() without having to look up keycodes

Cons (updated based on community comments)

• would this be confusing for cases like keyIsDown(4) where the user might mean the number 4? (note that 3, 8, and 9 are valid keycodes, so we can't just allow all numbers < 10)

Proposal status

Under review

[wwwld1]

Hi, I'd like to work on this, thank you!

[RuimingShen]

Hi, I'd like to work on this, thank you! I think I solve it.

[wwwld1]

Sorry I don't know if this is a right pin @limzykenneth, I saw that you are under I/O and I am currently working on the implementation for this. It seems to be a conflict for the const "LEFT_ARROW" with the purposed solutions of using String as a input. I can propose 2 solutions.

1. We change the validateParameters for keyIsDown to allow both type number and string for the code parameter (this allows to keep the constants DOWN_ARROW, RIGHT_ARROW, etc.
2. Or we can keep things more consistent, and changing the parameter to allow allow strings, and make users type "DOWN_ARROW", "LEFT_ARROW", etc instead (and then maybe delete the constants for the arrow and other escape keys?).

[wwwld1]

Hey,

I made a working draft PR above for keyIsDown() to accept alphanumeric String params such as 'w' or 'W'. Regarding the confusing case of the int 4 versus string '4', this implementation includes both where the string parameter e.g. '4' reflects the ASCII value of '4' while the int parameter e.g. 4 reflects the ASCII code 4.

The documentation is also updated to reflect the changes with examples now included for the new changes.

Let me know if any unit tests are necessary for these changes (based on the current tests for keyIsDown() I do not think we need to add any additional tests)

Feedback is much appreciated :)

[mvicky2592]

@wwwld1 I think keyIsDown should also accept keys with names longer than one character such as "arrowUp", "command", and "space"