box2d physics library #63

HMart81 · 2025-03-28T15:50:44Z

Box2D is a popular 2D Physics engine for 2d games, docs and more on https://box2d.org/

add ID values for GuiIconName to make it easy to embed icons id's in strings

changed loadDirectoryFiles to raylib 55 to follow new function signture and unloadDirectoryFiles missed the necessary argument

Box2D is a 2D Physics engine, see more on https://box2d.org/

Tried to follow c3 vendor guide lines "conform to C3 stdlib layout (tabs, Allman braces)" didn't do this first because was unaware of them, don't known if the tabs stuff is fully correct, thou I changed the leading ones to real tab, hope is enough.

lerno · 2025-03-30T00:28:53Z

This looks like this is also making changes to the raylib bindings? Also, it's 0.7.0 and 0.6.8, not 7.0.0 and 6.8.0 :D

HMart81 · 2025-03-30T13:25:57Z

Oh sorry about that! I'm still a noob at this github technical stuff (yeh shows I'm not a professional programmer, i'm just a self taught one), if possible ignore the raygui and raylib stuff.
And I most have been on drugs about the 6.8.0 and 7.0.0 versioning numbers, I don't know how the hell I thought that.

HMart81 · 2025-03-30T13:34:11Z

Thou looking at the two raylib changes I did (I did them some time ago, so I forgot about it...) they do make sense, as those were wrong function definitions/bindings in the original, unless those functions were already corrected in the mean time, I recommend that you also merge my raylib changes.

lerno · 2025-03-31T16:43:46Z

I've removed the 0.6.8 versions and I'm doing 0.7.0 now exclusively. Could you update the bindings and changes for it?

HMart81 · 2025-03-31T17:25:05Z

Yes I can. :)

lerno · 2025-04-01T23:11:50Z

Could you do these changes?

HMart81 · 2025-04-02T12:15:37Z

I already did the updates to 0.7.0, if that is what you are asking?

lerno · 2025-04-02T23:00:59Z

No, there are some new changes requested, see in the boxes above!

HMart81 · 2025-04-04T11:58:07Z

Sorry for my ignorance but what boxes above? Can you point me to what you are talking about. Or can you write what you want me to change in a comment here?
I'm used to tortoise svn for most of my programing, so I still don't know all ins and outs of this website.

libraries/box2d.c3l/box2d.c3i

@lerno

changed the code to follow more closely @lerno suggestions (still not tested at this point...)

now fully compiles

lerno · 2025-04-06T23:38:08Z

I'm sorry I haven't checked this yet. I'll try to have time soon. 🙏

lerno · 2025-05-08T09:32:23Z

libraries/box2d.c3l/box2d.c3i

+}
+
+// Body id references a body instance. This should be treated as an opaque handle.
+struct B2BodyId


I would make these implement comparisons using operator overloading with macro methods.

I may have bitten more than I can chew...
ok... any examples I can look at, so I can see what you mean?
Btw If I didn't said this before, I will say it now, I never used box2d before this, I know almost nothing about it, I just heard Unity3D used it and thought a 2D physics engine, would be cool to use with C3 for simple 2D games, so probably for me to really optimize its library for C3, I would need to first learn well its ins and outs, but I will see what I can do.

lerno · 2025-05-08T09:37:18Z

libraries/box2d.c3l/box2d.c3i

+//=====================================================================================================
+module vendor::box2d::b2::b2math;
+
+alias B2Vec2 = float[2]; // NOTE(HM) to support C3 vector math using (float[<2>])


You can make B2Vec2 a typedef and then implement the different vector ops on it, but it's probably easier to leave it as is, since a float[2] will implicitly convert to a float[<2>] and vice versa. See if you want to use operator overloading on structs such as B2Rot.

If it can stay like that for now better, it can always be changed later by me or others... the reality, I don't have much time to dedicate to this now. Not if you don't want to wait one more month for this... :(
Don't know how many people are waiting for this bindings, but they are always free to comment here. (...and also help btw....)

can this stay like this?

lerno · 2025-05-08T09:38:00Z

libraries/box2d.c3l/manifest.json

+    "macos-x64" : {
+      "link-args" : [],
+      "dependencies" : [],
+      "linked-libraries" : []


Please add "linked-libraries" to macos and windows too.

this is mostly done?

lerno · 2025-05-08T09:39:22Z

libraries/raygui.c3l/raygui.c3i

-  ICON_254,
-  ICON_255,
+// val is the ID number for the "#iconID#sometext" ability
+enum GuiIconName : int (int val)  {


If it is the same as the ordinal the int val part can be removed.

I have the impression we already discussed this? Even if is useless, I think those should stay?
Thou I will probably make them into a comment, or better, thinking about it, make them into a inline string, so you can use the enum member directly and just concatenate them into other strings?

like so:

enum GuiIconName : (inline String str) { CURSOR_HAND = "#19#", } if( rgui::button(rec, CURSOR_HAND.zstr_tcopy()) ) { // }

Anyway why I think they should stay, the main reason is, because it removes the need to count each enum by hand, to know its value visually.
Imagine you saw a online tutorial on box2d and it used "#128#" on it (many do use the id directly like that), looking at that first time, what icon is that id?
If those index numbers are not there, for you to immediately connect the dots, you will have to count each enum member to find who is the 128 one...

But like I said, probably making them into a simple // comment or inline string is enough... but I'm not sure, I may be forgetting something.

I did this modifications

edited the manifest file and provided a compiled box2d library for windows-x64 I use, but imo the best is for others to compile box2d themselves for their own CPU architecture... Unfortunately cannot provide macos and linux ones at this time. Changed the raygui icon enum to be a inline ZString so you can pass them directly to raygui functions as zero terminated strings for the "#id#" feature.

lerno · 2025-05-12T22:15:16Z

Hmm.. it has some conflicts now.

lerno · 2025-05-13T12:35:58Z

Let's merge this and let improvements come later.

HMart81 · 2025-05-13T12:39:40Z

Let's merge this and let improvements come later.

Agree and thanks for the patience in dealing with a noob like me. And sorry for not doing more but other work got in the way.

HMart81 added 4 commits January 6, 2025 23:07

Update raygui.c3i

4963dc4

add ID values for GuiIconName to make it easy to embed icons id's in strings

Update raylib.c3i

17bc1d5

changed loadDirectoryFiles to raylib 55 to follow new function signture and unloadDirectoryFiles missed the necessary argument

add a box2d vendor library

5497b68

Box2D is a 2D Physics engine, see more on https://box2d.org/

Update box2d.c3i

14e5575

Tried to follow c3 vendor guide lines "conform to C3 stdlib layout (tabs, Allman braces)" didn't do this first because was unaware of them, don't known if the tabs stuff is fully correct, thou I changed the leading ones to real tab, hope is enough.

change readme to refer to correct C3 versions

d127792

update box2d to c3 0.7.0

67dda1e

lerno requested changes Apr 4, 2025

View reviewed changes

libraries/box2d.c3l/box2d.c3i Outdated Show resolved Hide resolved

libraries/box2d.c3l/box2d.c3i Outdated Show resolved Hide resolved

libraries/box2d.c3l/box2d.c3i Outdated Show resolved Hide resolved

libraries/box2d.c3l/box2d.c3i Outdated Show resolved Hide resolved

HMart81 added 2 commits April 5, 2025 15:23

Update box2d.c3i

dcd9e07

changed the code to follow more closely @lerno suggestions (still not tested at this point...)

Update box2d.c3i

4c91278

now fully compiles

HMart81 requested a review from lerno April 6, 2025 15:26

lerno requested changes May 8, 2025

View reviewed changes

Merge branch 'main' into hm_vendor

66192fe

lerno merged commit 7d98837 into c3lang:main May 13, 2025
2 checks passed

HMart81 deleted the hm_vendor branch May 13, 2025 12:45

box2d physics library #63

box2d physics library #63

Uh oh!

Conversation

HMart81 commented Mar 28, 2025

Uh oh!

lerno commented Mar 30, 2025

Uh oh!

HMart81 commented Mar 30, 2025

Uh oh!

HMart81 commented Mar 30, 2025

Uh oh!

lerno commented Mar 31, 2025

Uh oh!

HMart81 commented Mar 31, 2025

Uh oh!

lerno commented Apr 1, 2025

Uh oh!

HMart81 commented Apr 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

lerno commented Apr 2, 2025

Uh oh!

HMart81 commented Apr 4, 2025

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

lerno commented Apr 6, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

HMart81 May 8, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

lerno commented May 12, 2025

Uh oh!

lerno commented May 13, 2025

Uh oh!

Uh oh!

HMart81 commented May 13, 2025

Uh oh!

Uh oh!

HMart81 commented Apr 2, 2025 •

edited

Loading

HMart81 May 8, 2025 •

edited

Loading