Repository navigation
[p5.strands] Reenabling and implementing FES in strands #7899
Description
Activity
@lukeplowden I’d like to work on this issue. Could you please assign it to me?
Hi @Satish-Medar, thank you for being interested. I have been away this month hence the slow reply.
I am just finishing up a refactor in #8009 which is pertinent to this issue. You can see here I've started a small file to handle FES.
Here, in the new main strands file, you can see FES is disabled/ re-enabled when strands is initialised.
One good approach for now is to identify all of the places where FES will be used in strands. You could start by looking at all my references to FES in that branch and taking inventory, also cross referencing with this document and making sure the style aligns.
Reacted by kitHi @lukeplowden , @ksen0 ,
I would like to work on this issue, First I had planned to go
- through p5 strands module and create inventory where FES is disabled or needs to implement. I will also follow the style with FES contribution guide.
I will paste the inventory list here in this thread, after a review, I can implement
FES handlers to correctly bypass or validate the dynamic GLSL nodesIf this approach sounds good, I can start this
could you assign this to me ??I just go through the strandNode which exposes typeinfo. so I am thinking to update the FES validators
So when they receive argument for strandNode being true. they will be dynamically validate against base type of typeinfoHi @Ayush4958 I've noticed you have been making many comments on multiple open issues. Please don't do too many requests for assignment and feedback, this kind of behavior results in many notifications and slower times to respond/review. We appreciate the interest but right now the bottleneck is responding to comments and reviewing. If you want, helping to review open PRs by testing them or providing informed feedback would be very helpful to improve overall efficiency. Thanks for understanding!
Sorry for commenting on multiple issues
I will try to review the opened pr
And will provide informed feedbackThanks ...
Metadata
Metadata
Assignees
Type
Projects
- Status
Increasing access
FES is a core part of making p5.js approachable, but it is currently disabled in p5.strands
Most appropriate sub-area of p5.js?
Feature enhancement details
p5.strands disables FES entirely whilst it compiles the user's code to GLSL. This is because there are some functions with shared scope and name clashes, in p5.strands and the core p5.js library... see #7849. This causes two types of false negatives, to my knowledge:
_validateParameters, since it expects the core function signature. For example, if you usesin(10)in strands, it will be converted tosin(dynamicNode(10)), and the actual received arg will not be of type Number. FES will therefore warn thatsinexpects a number. This is also the case foratan()which GLSL allows 1 or 2 arguments for, whereas p5.js only allows one.