progress.txt

CAS issues i'm saying fuck it for
	sometimes want negative value instead
		with pendulum, have to set the position to -ybob so it's down
	for the linkage specifying L gives a random gues for theta
		hasinfinitesolutions SHOULD work BUT...
			(1-sin(x)^2)^0.5-cos(x) is actually only sometimes zero cause the square is moved over to the other side
			(1-sin(x)^2)-cos(x)^2 would of course have worked
	mohrs circle
		circle equation and sin and cos --> contradiction cause sin and cos only gets one sign, which could be wrong
		circle equation and tan --> flips sign, as it moves past pi/2
		need just sin and cos for it to work
	WHY CANT SOLVE sheet (truss example), only issue when you shift the top two nodes vertically, there is too much unknown, not sure why


should remove "Sheet saved" after typing


visual ramp could instead be derived from triangular prism primitive


water wave showing forces on columns and conservation of mass





bugs
	NONE FUCK YES
		there are actually still bugs but i dont fucking care


shortcuts



wheel in arbitrary orientation
	can use instead of bevel for the differential (would actually be clearer)

curvedarrow needs scale


if there's an error before the solve block, ideally the input values for saving should be restored after the error is fixed

undo doesnt clear visuals?????????
	i thought i fixed this
	i think undo and redo shouldn't run the sheet each time, that introduces annoying lag
		doesnt work cause you lose sub table
	might just undo and redo to last run (e.g. Ctrl+Enter)

for some reason setting zoom messes up MQ (changes where the cursur is initially on focus)

ellipse primitive
	for phase state diagram and water particle motion

forces on wave

composite section

instead of grouping into major topics, i think it's better to group into smaller things, and go deep into each
	
waterwave no work at 0 amplitude

visuals
	statics
		rections point and distributed load
		hanging cable
		moment from point loads, influence
		space truss
		2d truss, could have a node primitve with 5 forces, set ones you don't need to 0
	dynamics
		friction wheels <-- cant actually be done, can only get the final state
		pendulum (small angle, show forces)
			unfortunately will have to hardcode sinusoidal motion
		spinning pendulum
		spring mass SHM, show phase state diagram
		projectile motion
		ramp (hollow tube)
		pendulum with shifted center
		comparison of forces on spinning vs regular pendulum
		gyroscope?
		feynman spinning plate (again would need arbitrary axis wheel)

	optics
		multiple refraction layers
		circular mirror
		lens (might need lens primitive)
		prism refraction
	solid mechanics
		tension in series
		torsion in series
		section properties
		mohr
		3d stress strain
		stress in beam
	fluid mechanics
		pipe (single and then in series, then maybe network)
		wave (show particle motion)
		open channel (hyrdaulic depth)
		open channel (expansion, contraction, raising lowering bottom)
		specific energy of channel
		maybe hydrostatics, not sure
	soil mechanics
		soil properties derivation (no visual)
		soil components
		point loads stress
		triaxial (bringing back mohr stuff)
			mu and cohesion
	mechanisms
		4 bar linkage
		quick/springy return
		planetary gear
		automatic transmission
		rotation to oscillating motion
		




To do at end

	make sure all error messages make sense
		error message convert variable names to plain text (using greek mapping i used for glowscript text) (assuming i keep the variables, not sure)

	some sort of indication that the sheet is loading when you first go on the site to a particular sheet

	only show solutions steps when it successfully solves 

	catch all error so it doesn't break if some weird error comes up




visuals to create

	water wave showing particle motion

	basics
		helices


	phase state diagram with spring on mass

	

	basic statics, equilibrium loads
		but could also have distributed loads
		of course much more limited than skyciv but nice how you'd see loads update real time
		could also have spring supports, basic sort of static indeterminancy
			interesting cause i'm not exactly sure what the behavior is

	light ray refraction
		could also have prisms and stuff, interface at an angle
	
	surface bending with poisson
		could fix 0 bending one way, and determines moment required

	multiple loads on a cable
		assume cable is weightless, so just straight lines

	ramps
		rolling, moment of inertia
		mech advantage

	the pulling wheel by string "paradox"

	equilibrium of chemical reactions?
		concentrations at equilibrium



	ideal gas law
		n of little spheres and color for T, and arrows for P
		kind of pointless though unless you do something else with it
		could go all out and actually create the model with collisions
			could just do 2d
			could even show expected vs actual averaged pressure

		not super useful to hav a cas at that point, since P would almost always be the output
			could also show plot (or at least a point on the plot), then moving it shows the different parameters

	buoyancy?
		not really much you can do with it without integrating

	tiny space truss
		super simple things
		still would have to deal with 3d eqns, annoying cause i don't have vectors 

	planes connected in shear
		not that interesting but pretty simple
			would just need a quad primitive, which is anyway a gs primitive so super easy


	wedge
		super simple, could just be 2d, three forces

	multiple stacked blocks in friction
		issue is i don't have inequalities, max, or min, so i can't say something like when one of the surfaces slips

	the arched stack blocks thing
		issue is i don't have sum
			might still be ok, since you would just add to the one above


	trapezoidal distribution
	
	random stuff
		balloon, how the pressure after a certain point goes down cause the rubber thins
			or the two balloons thing, issue is it would need to find the two different solutions
		you could have load suspended 3 points in 3d space, show how the tension changes with the angles
		car differential
		different stuff for mechanical advantage

	hydrualics
		pipes (network)
		pipes in series with grad lines
		weirs (discharge of different shapes?)
	mechanics
		basically all the stuff from old seeciv 
		mohr's circle <-- could be very useful
		torsion (and with multiple rods)
		beam moment shear (distributions would be built into the visual)
		bending (honestly not that interesting a visual)
	3bar linkage, already have 
		could do other mechanisms even 3D
	statics
		might need vector primitive to be very useful
		could have forces on something balanced out
	geotech/foundations
 		soil layers stresses
			would want to have functions, but could use trapezoids
		soil stress over area from loads
			would need color plot (more generally subbing an expresion instead of a value)
		foundations
			issue is a lot of them are piecewise
	
	structural
		section properties

		beam loading??
			again would need function
	
	physics

		springs and forces and stuff (probably not dynamic)

		projectile motion (would just need sphere primitive)

	water treatment

		size of things like the spinny things
	
	numerical analysis??? need to think about that


Future??

	CAS
		
		0=0 should be removed, at least from eqns to display
		simplify top exponent out and top product coefficients????
			getting stuck on this

		tests with remove vars too?
		y=x^2 complexity is taken as 3 since *-1 adds to the complexity --> too much
		haven't handled this: combine_solve_term(eqn_to_tree("a*(a+a*b)+d"),"a")


			
	Efficiency

		could be even faster by only simplifying where it's subbed in for sub_in function
		could save the tree from block to block so it doesn't have to parse repeatedly



	input expressions instead of just values to plotting

	make visuals classes that could inherit from one another
	Draggable blocks
	loading a sheet has to have a callback to update the variable tracker
	Symbolic if statements
	Max, min, and inequalities