Add Electrical Component docstrings

[AayushSabharwal]

Working on adding some docstrings, trying to follow the pattern in MTK.jl docs.

Some questions:

• Do the ModelingToolkit.connect methods need docstrings? Consequently, does ElectricalPin need one?
• Do the ideal components need to specify how their variables are calculated?
• I've studied OpAmps a bit, but I'm not particularly sure how this one works
• How should the pins of sensors (and components in general) be documented? All two-pin components have pins p and n, so should this be mentioned in every sensor/component or left as an assumed standard? What about components with more pins?

EDIT: I'm not sure why I didn't just take the docs from the README

[AayushSabharwal]

It seems like documenting macro usages is not straightforward. The docstrings for @connector function Pin and @connector function DigitalPin seem to be the issue. I did come across this but it doesn't seem to work here, just leads to another error in MTK.jl

[AayushSabharwal]

Apologies for the inactivity. I have a lot going on right now. Would love to get back to this in a week!

[AayushSabharwal]

Documenting Pin and DigitalPin requires @doc to be used, since they are the result of macro evaluations

[AayushSabharwal]

If #18 is on its way to completion, it might be better to wait and migrate the docstrings over to those new functions? It would reduce a lot of duplicacy

[AayushSabharwal]

I'm not familiar with PulseDiff, but the rest of the docstrings are done. I left out the current sources because of the above comment. They would be nearly identical to the docstrings of the corresponding voltage sources.

[ChrisRackauckas]

Looks good thanks!

[ChrisRackauckas]

Test failure

[AayushSabharwal]

I'll look into this. Weird that docs failed tests.

[AayushSabharwal]

It seems that @register _and(x...) is the issue, in src/Electrical/Digital/tables.jl:59. Symbolics deprecated @register for @register_symbolic, and it doesn't like that. This happens on the main branch too.

I tried updating to using @register_symbolic, but that doesn't fix it either and would break compatibility with pre-v4 Symbolics.

[ChrisRackauckas]

we should just drop the earlier Symbolics and update.

[AayushSabharwal]

I'm afraid this still won't precompile, because Symbolics doesn't like the x... in @register_symbolic _and(x...)

[AayushSabharwal]

It seems the deprecation of @register for @register_symbolic broke this. If I use @register everywhere on Symbolics v4.1.0 this works fine

v4.1.1 breaks this, so from looking at the release log, my guess would be that JuliaSymbolics/Symbolics.jl#457 is the relevant PR causing issues

[AayushSabharwal]

MTK v8 removed promote_connect_rule in SciML/ModelingToolkit.jl#1348, which breaks this. What would be the new way to support this?

[ChrisRackauckas]

If the variables are appropriately defined as flow variables this should all happen automatically through the connect function.

[AayushSabharwal]

I guess that should happen in #25 then. Symbolics v4.1.1 still breaks this, and main as well.

[AayushSabharwal]

Tests should pass now with Symbolics v4.2.2 release

[ChrisRackauckas]

Please update this to match master. We just had a pretty major change (#7) and it would be good to copy these over if possible.

[ChrisRackauckas]

Looking at http://mtkstdlib.sciml.ai/dev/API/electrical/, I like how your format defines the ports and everything, so updating it would be great.

[AayushSabharwal]

Sure! I'm a bit swamped with work right now, but I'll get to this as soon as I can.

[ChrisRackauckas]

Oh no, don't merge, rebase this one 😓. You might want to just start from scratch and copy over the major changes.

[AayushSabharwal]

I guess that's a better idea. I spent the past half hour trying to reconcile all of this only for it to show up again :P

Rebasing was annoying, since it seemed to process commit by commit and kept getting merge conflicts in the same files at the same places repeatedly, so I tried --no-rebase. Looks like git still doesn't like how this branch looks. I'll make another one and PR again.