Platform checks

tuneup

.zed/tasks.json

@@ -55,6 +55,16 @@
     "command": "odin run draw/examples -debug -out=out/debug/draw-examples -- hellope-shapes",
     "cwd": "$ZED_WORKTREE_ROOT",
   },
+  {
+    "label": "Run draw hellope-text example",
+    "command": "odin run draw/examples -debug -out=out/debug/draw-examples -- hellope-text",
+    "cwd": "$ZED_WORKTREE_ROOT",
+  },
+  {
+    "label": "Run draw hellope-custom example",
+    "command": "odin run draw/examples -debug -out=out/debug/draw-examples -- hellope-custom",
+    "cwd": "$ZED_WORKTREE_ROOT",
+  },
   // ---------------------------------------------------------------------------------------------------------------------
   // ----- Other ------------------------
   // ---------------------------------------------------------------------------------------------------------------------

draw/README.md

@@ -555,3 +555,26 @@ odin run meta -- gen-shaders
 ```
 
 Requires `glslangValidator` and `spirv-cross` on PATH.
+
+### Shader format selection
+
+The library embeds shader bytecode per compile target — MSL + `main0` entry point on Darwin (via
+`spirv-cross --msl`, which renames `main` because it is reserved in Metal), SPIR-V + `main` entry
+point elsewhere. Three compile-time constants in `draw.odin` expose the build's shader configuration:
+
+| Constant                      | Type                      | Darwin    | Other      |
+| ----------------------------- | ------------------------- | --------- | ---------- |
+| `PLATFORM_SHADER_FORMAT_FLAG` | `sdl.GPUShaderFormatFlag` | `.MSL`    | `.SPIRV`   |
+| `PLATFORM_SHADER_FORMAT`      | `sdl.GPUShaderFormat`     | `{.MSL}`  | `{.SPIRV}` |
+| `SHADER_ENTRY`                | `cstring`                 | `"main0"` | `"main"`   |
+
+Pass `PLATFORM_SHADER_FORMAT` to `sdl.CreateGPUDevice` so SDL selects a backend compatible with the
+embedded bytecode:
+
+```
+gpu := sdl.CreateGPUDevice(draw.PLATFORM_SHADER_FORMAT, true, nil)
+```
+
+At init time the library calls `sdl.GetGPUShaderFormats(device)` to verify the active backend
+accepts `PLATFORM_SHADER_FORMAT_FLAG`. If it does not, `draw.init` returns `false` with a
+descriptive log message showing both the embedded and active format sets.

draw/draw.odin

@@ -1,21 +1,27 @@
 package draw
 
-import clay "../vendor/clay"
 import "base:runtime"
 import "core:c"
 import "core:log"
-
+import "core:math"
 import "core:strings"
 import sdl "vendor:sdl3"
 import sdl_ttf "vendor:sdl3/ttf"
 
+import clay "../vendor/clay"
+
 when ODIN_OS == .Darwin {
-	SHADER_TYPE :: sdl.GPUShaderFormat{.MSL}
-	ENTRY_POINT :: "main0"
+	PLATFORM_SHADER_FORMAT_FLAG :: sdl.GPUShaderFormatFlag.MSL
+	SHADER_ENTRY :: cstring("main0")
+	BASE_VERT_2D_RAW :: #load("shaders/generated/base_2d.vert.metal")
+	BASE_FRAG_2D_RAW :: #load("shaders/generated/base_2d.frag.metal")
 } else {
-	SHADER_TYPE :: sdl.GPUShaderFormat{.SPIRV}
-	ENTRY_POINT :: "main"
+	PLATFORM_SHADER_FORMAT_FLAG :: sdl.GPUShaderFormatFlag.SPIRV
+	SHADER_ENTRY :: cstring("main")
+	BASE_VERT_2D_RAW :: #load("shaders/generated/base_2d.vert.spv")
+	BASE_FRAG_2D_RAW :: #load("shaders/generated/base_2d.frag.spv")
 }
+PLATFORM_SHADER_FORMAT :: sdl.GPUShaderFormat{PLATFORM_SHADER_FORMAT_FLAG}
 
 BUFFER_INIT_SIZE :: 256
 INITIAL_LAYER_SIZE :: 5
@@ -35,14 +41,14 @@ BLUE :: Color{0, 0, 255, 255}
 BLANK :: Color{0, 0, 0, 0}
 
 // Convert clay.Color ([4]c.float in 0–255 range) to Color.
-color_from_clay :: proc(cc: clay.Color) -> Color {
-	return Color{u8(cc[0]), u8(cc[1]), u8(cc[2]), u8(cc[3])}
+color_from_clay :: proc(clay_color: clay.Color) -> Color {
+	return Color{u8(clay_color[0]), u8(clay_color[1]), u8(clay_color[2]), u8(clay_color[3])}
 }
 
 // Convert Color to [4]f32 in 0.0–1.0 range. Useful for SDL interop (e.g. clear color).
-color_to_f32 :: proc(c: Color) -> [4]f32 {
+color_to_f32 :: proc(color: Color) -> [4]f32 {
 	INV :: 1.0 / 255.0
-	return {f32(c[0]) * INV, f32(c[1]) * INV, f32(c[2]) * INV, f32(c[3]) * INV}
+	return {f32(color[0]) * INV, f32(color[1]) * INV, f32(color[2]) * INV, f32(color[3]) * INV}
 }
 
 // ---------------------------------------------------------------------------------------------------------------------
@@ -52,8 +58,8 @@ color_to_f32 :: proc(c: Color) -> [4]f32 {
 Rectangle :: struct {
 	x:      f32,
 	y:      f32,
-	w: f32,
-	h: f32,
+	width:  f32,
+	height: f32,
 }
 
 Sub_Batch_Kind :: enum u8 {
@@ -100,7 +106,8 @@ Global :: struct {
 	tmp_text_batches:  [dynamic]TextBatch,
 	tmp_primitives:    [dynamic]Primitive,
 	tmp_sub_batches:   [dynamic]Sub_Batch,
-	clay_mem:          [^]u8,
+	tmp_uncached_text: [dynamic]^sdl_ttf.Text, // Uncached TTF_Text objects to destroy after end()
+	clay_memory:       [^]u8,
 	msaa_texture:      ^sdl.GPUTexture,
 	curr_layer_index:  uint,
 	max_layers:        int,
@@ -112,8 +119,8 @@ Global :: struct {
 	max_primitives:    int,
 	max_sub_batches:   int,
 	dpi_scaling:       f32,
-	msaa_w:            u32,
-	msaa_h:            u32,
+	msaa_width:        u32,
+	msaa_height:       u32,
 	sample_count:      sdl.GPUSampleCount,
 	clay_z_index:      i16,
 	cleared:           bool,
@@ -169,25 +176,26 @@ init :: proc(
 		tmp_text_batches  = make([dynamic]TextBatch, 0, BUFFER_INIT_SIZE, allocator = allocator),
 		tmp_primitives    = make([dynamic]Primitive, 0, BUFFER_INIT_SIZE, allocator = allocator),
 		tmp_sub_batches   = make([dynamic]Sub_Batch, 0, BUFFER_INIT_SIZE, allocator = allocator),
+		tmp_uncached_text = make([dynamic]^sdl_ttf.Text, 0, 16, allocator = allocator),
 		odin_context      = odin_context,
 		dpi_scaling       = sdl.GetWindowDisplayScale(window),
-		clay_mem         = make([^]u8, min_memory_size, allocator = allocator),
+		clay_memory       = make([^]u8, min_memory_size, allocator = allocator),
 		sample_count      = resolved_sample_count,
 		pipeline_2d_base  = pipeline,
 		text_cache        = text_cache,
 	}
 	log.debug("Window DPI scaling:", GLOB.dpi_scaling)
-	arena := clay.CreateArenaWithCapacityAndMemory(min_memory_size, GLOB.clay_mem)
+	arena := clay.CreateArenaWithCapacityAndMemory(min_memory_size, GLOB.clay_memory)
 	window_width, window_height: c.int
 	sdl.GetWindowSize(window, &window_width, &window_height)
 
 	clay.Initialize(arena, {f32(window_width), f32(window_height)}, {handler = clay_error_handler})
-	clay.SetMeasureTextFunction(measure_text, nil)
+	clay.SetMeasureTextFunction(measure_text_clay, nil)
 	return true
 }
 
-// TODO every x frames nuke max values in case of edge cases where max gets set very high
-// Called at the end of every frame
+// TODO Either every x frames nuke max values in case of edge cases where max gets set very high
+// or leave to application code to decide the right time for resize
 resize_global :: proc() {
 	if len(GLOB.layers) > GLOB.max_layers do GLOB.max_layers = len(GLOB.layers)
 	shrink(&GLOB.layers, GLOB.max_layers)
@@ -216,7 +224,9 @@ destroy :: proc(device: ^sdl.GPUDevice, allocator := context.allocator) {
 	delete(GLOB.tmp_text_batches)
 	delete(GLOB.tmp_primitives)
 	delete(GLOB.tmp_sub_batches)
-	free(GLOB.clay_mem, allocator)
+	for ttf_text in GLOB.tmp_uncached_text do sdl_ttf.DestroyText(ttf_text)
+	delete(GLOB.tmp_uncached_text)
+	free(GLOB.clay_memory, allocator)
 	if GLOB.msaa_texture != nil {
 		sdl.ReleaseGPUTexture(device, GLOB.msaa_texture)
 	}
@@ -229,6 +239,9 @@ clear_global :: proc() {
 	GLOB.curr_layer_index = 0
 	GLOB.clay_z_index = 0
 	GLOB.cleared = false
+	// Destroy uncached TTF_Text objects from the previous frame (after end() has submitted draw data)
+	for ttf_text in GLOB.tmp_uncached_text do sdl_ttf.DestroyText(ttf_text)
+	clear(&GLOB.tmp_uncached_text)
 	clear(&GLOB.layers)
 	clear(&GLOB.scissors)
 	clear(&GLOB.tmp_shape_verts)
@@ -244,7 +257,7 @@ clear_global :: proc() {
 // ---------------------------------------------------------------------------------------------------------------------
 
 @(private = "file")
-measure_text :: proc "c" (
+measure_text_clay :: proc "c" (
 	text: clay.StringSlice,
 	config: ^clay.TextElementConfig,
 	user_data: rawptr,
@@ -252,12 +265,12 @@ measure_text :: proc "c" (
 	context = GLOB.odin_context
 	text := string(text.chars[:text.length])
 	c_text := strings.clone_to_cstring(text, context.temp_allocator)
-	w, h: c.int
-	if !sdl_ttf.GetStringSize(get_font(config.fontId, config.fontSize), c_text, 0, &w, &h) {
+	width, height: c.int
+	if !sdl_ttf.GetStringSize(get_font(config.fontId, config.fontSize), c_text, 0, &width, &height) {
 		log.panicf("Failed to measure text: %s", sdl.GetError())
 	}
 
-	return clay.Dimensions{width = f32(w) / GLOB.dpi_scaling, height = f32(h) / GLOB.dpi_scaling}
+	return clay.Dimensions{width = f32(width) / GLOB.dpi_scaling, height = f32(height) / GLOB.dpi_scaling}
 }
 
 // ---------------------------------------------------------------------------------------------------------------------
@@ -274,8 +287,8 @@ begin :: proc(bounds: Rectangle) -> ^Layer {
 		bounds = sdl.Rect {
 			x = i32(bounds.x * GLOB.dpi_scaling),
 			y = i32(bounds.y * GLOB.dpi_scaling),
-			w = i32(bounds.w * GLOB.dpi_scaling),
-			h = i32(bounds.h * GLOB.dpi_scaling),
+			w = i32(bounds.width * GLOB.dpi_scaling),
+			h = i32(bounds.height * GLOB.dpi_scaling),
 		},
 	}
 	append(&GLOB.scissors, scissor)
@@ -305,8 +318,8 @@ new_layer :: proc(prev_layer: ^Layer, bounds: Rectangle) -> ^Layer {
 		bounds = sdl.Rect {
 			x = i32(bounds.x * GLOB.dpi_scaling),
 			y = i32(bounds.y * GLOB.dpi_scaling),
-			w = i32(bounds.w * GLOB.dpi_scaling),
-			h = i32(bounds.h * GLOB.dpi_scaling),
+			w = i32(bounds.width * GLOB.dpi_scaling),
+			h = i32(bounds.height * GLOB.dpi_scaling),
 		},
 	}
 	append(&GLOB.scissors, scissor)
@@ -336,14 +349,19 @@ prepare_sdf_primitive :: proc(layer: ^Layer, prim: Primitive) {
 
 // Submit a text element to the given layer for rendering.
 // Copies SDL_ttf vertices directly (with baked position) and copies indices for indexed drawing.
-prepare_text :: proc(layer: ^Layer, txt: Text) {
-	data := sdl_ttf.GetGPUTextDrawData(txt.ref)
+prepare_text :: proc(layer: ^Layer, text: Text) {
+	data := sdl_ttf.GetGPUTextDrawData(text.sdl_text)
 	if data == nil {
 		return // nil is normal for empty text
 	}
 
 	scissor := &GLOB.scissors[layer.scissor_start + layer.scissor_len - 1]
 
+	// Snap base position to integer physical pixels to avoid atlas sub-pixel
+	// sampling blur (and the off-by-one bottom-row clip that comes with it).
+	base_x := math.round(text.position[0] * GLOB.dpi_scaling)
+	base_y := math.round(text.position[1] * GLOB.dpi_scaling)
+
 	for data != nil {
 		vertex_start := u32(len(GLOB.tmp_text_verts))
 		index_start := u32(len(GLOB.tmp_text_indices))
@@ -354,11 +372,7 @@ prepare_text :: proc(layer: ^Layer, txt: Text) {
 			uv := data.uv[i]
 			append(
 				&GLOB.tmp_text_verts,
-				Vertex {
-					position = {pos.x + txt.position[0] * GLOB.dpi_scaling, -pos.y + txt.position[1] * GLOB.dpi_scaling},
-					uv = {uv.x, uv.y},
-					color = txt.color,
-				},
+				Vertex{position = {pos.x + base_x, -pos.y + base_y}, uv = {uv.x, uv.y}, color = text.color},
 			)
 		}
 
@@ -384,6 +398,54 @@ prepare_text :: proc(layer: ^Layer, txt: Text) {
 	}
 }
 
+// Submit a text element with a 2D affine transform applied to vertices.
+// Used by the high-level `text` proc when rotation or a non-zero origin is specified.
+// NOTE: xform must be in physical (DPI-scaled) pixel space — the caller pre-scales
+// pos and origin by GLOB.dpi_scaling before building the transform.
+prepare_text_transformed :: proc(layer: ^Layer, text: Text, transform: Transform_2D) {
+	data := sdl_ttf.GetGPUTextDrawData(text.sdl_text)
+	if data == nil {
+		return
+	}
+
+	scissor := &GLOB.scissors[layer.scissor_start + layer.scissor_len - 1]
+
+	for data != nil {
+		vertex_start := u32(len(GLOB.tmp_text_verts))
+		index_start := u32(len(GLOB.tmp_text_indices))
+
+		for i in 0 ..< data.num_vertices {
+			pos := data.xy[i]
+			uv := data.uv[i]
+			// SDL_ttf gives glyph positions in physical pixels relative to text origin.
+			// The transform is already in physical-pixel space (caller pre-scaled),
+			// so we apply directly — no per-vertex DPI divide/multiply.
+			append(
+				&GLOB.tmp_text_verts,
+				Vertex{position = apply_transform(transform, {pos.x, -pos.y}), uv = {uv.x, uv.y}, color = text.color},
+			)
+		}
+
+		append(&GLOB.tmp_text_indices, ..data.indices[:data.num_indices])
+
+		batch_idx := u32(len(GLOB.tmp_text_batches))
+		append(
+			&GLOB.tmp_text_batches,
+			TextBatch {
+				atlas_texture = data.atlas_texture,
+				vertex_start = vertex_start,
+				vertex_count = u32(data.num_vertices),
+				index_start = index_start,
+				index_count = u32(data.num_indices),
+			},
+		)
+
+		append_or_extend_sub_batch(scissor, layer, .Text, batch_idx, 1)
+
+		data = data.next
+	}
+}
+
 // Append a new sub-batch or extend the last one if same kind and contiguous.
 @(private)
 append_or_extend_sub_batch :: proc(
@@ -415,6 +477,19 @@ clay_error_handler :: proc "c" (errorData: clay.ErrorData) {
 	log.error("Clay error:", errorData.errorType, errorData.errorText)
 }
 
+// Called for each Clay `RenderCommandType.Custom` render command that
+// `prepare_clay_batch` encounters.
+//
+// - `layer` is the layer the command belongs to (post-z-index promotion).
+// - `bounds` is already translated into the active layer's coordinate system
+//   and pre-DPI, matching what the built-in shape procs expect.
+// - `render_data` is Clay's `CustomRenderData` for the element, exposing
+//   `backgroundColor`, `cornerRadius`, and the `customData` pointer the caller
+//   attached to `clay.CustomElementConfig.customData`.
+//
+// The callback must not call `new_layer` or `prepare_clay_batch`.
+Custom_Draw :: #type proc(layer: ^Layer, bounds: Rectangle, render_data: clay.CustomRenderData)
+
 ClayBatch :: struct {
 	bounds: Rectangle,
 	cmds:   clay.ClayArray(clay.RenderCommand),
@@ -426,6 +501,7 @@ prepare_clay_batch :: proc(
 	batch: ^ClayBatch,
 	mouse_wheel_delta: [2]f32,
 	frame_time: f32 = 0,
+	custom_draw: Custom_Draw = nil,
 ) {
 	mouse_pos: [2]f32
 	mouse_flags := sdl.GetMouseState(&mouse_pos.x, &mouse_pos.y)
@@ -447,8 +523,8 @@ prepare_clay_batch :: proc(
 		bounds := Rectangle {
 			x      = render_command.boundingBox.x + layer.bounds.x,
 			y      = render_command.boundingBox.y + layer.bounds.y,
-			w = render_command.boundingBox.width,
-			h = render_command.boundingBox.height,
+			width  = render_command.boundingBox.width,
+			height = render_command.boundingBox.height,
 		}
 
 		if render_command.zIndex > GLOB.clay_z_index {
@@ -466,32 +542,17 @@ prepare_clay_batch :: proc(
 			render_data := render_command.renderData.text
 			txt := string(render_data.stringContents.chars[:render_data.stringContents.length])
 			c_text := strings.clone_to_cstring(txt, context.temp_allocator)
-			sdl_text := GLOB.text_cache.cache[render_command.id]
-
-			if sdl_text == nil {
-				// Cache a SDL text object
-				sdl_text = sdl_ttf.CreateText(
-					GLOB.text_cache.engine,
-					get_font(render_data.fontId, render_data.fontSize),
+			// Clay render-command IDs are derived via Clay's internal HashNumber (Jenkins-family)
+			// and namespaced with .Clay so they can never collide with user-provided custom text IDs.
+			sdl_text := cache_get_or_update(
+				Cache_Key{render_command.id, .Clay},
 				c_text,
-					0,
+				get_font(render_data.fontId, render_data.fontSize),
 			)
-				if sdl_text == nil {
-					log.panicf("Failed to create SDL text for clay render command: %s", sdl.GetError())
-				}
-				GLOB.text_cache.cache[render_command.id] = sdl_text
-			} else {
-				if !sdl_ttf.SetTextString(sdl_text, c_text, 0) {
-					log.panicf("Failed to update SDL text string: %s", sdl.GetError())
-				}
-			}
-
 			prepare_text(layer, Text{sdl_text, {bounds.x, bounds.y}, color_from_clay(render_data.textColor)})
 		case clay.RenderCommandType.Image:
 		case clay.RenderCommandType.ScissorStart:
-			if bounds.w == 0 || bounds.h == 0 {
-				continue
-			}
+			if bounds.width == 0 || bounds.height == 0 do continue
 
 			curr_scissor := &GLOB.scissors[layer.scissor_start + layer.scissor_len - 1]
 
@@ -502,8 +563,8 @@ prepare_clay_batch :: proc(
 					bounds          = sdl.Rect {
 						c.int(bounds.x * GLOB.dpi_scaling),
 						c.int(bounds.y * GLOB.dpi_scaling),
-						c.int(bounds.w * GLOB.dpi_scaling),
-						c.int(bounds.h * GLOB.dpi_scaling),
+						c.int(bounds.width * GLOB.dpi_scaling),
+						c.int(bounds.height * GLOB.dpi_scaling),
 					},
 				}
 				append(&GLOB.scissors, new)
@@ -512,8 +573,8 @@ prepare_clay_batch :: proc(
 				curr_scissor.bounds = sdl.Rect {
 					c.int(bounds.x * GLOB.dpi_scaling),
 					c.int(bounds.y * GLOB.dpi_scaling),
-					c.int(bounds.w * GLOB.dpi_scaling),
-					c.int(bounds.h * GLOB.dpi_scaling),
+					c.int(bounds.width * GLOB.dpi_scaling),
+					c.int(bounds.height * GLOB.dpi_scaling),
 				}
 			}
 		case clay.RenderCommandType.ScissorEnd:
@@ -532,25 +593,23 @@ prepare_clay_batch :: proc(
 			render_data := render_command.renderData.border
 			cr := render_data.cornerRadius
 			color := color_from_clay(render_data.color)
-			thick := f32(render_data.width.top)
+			thickness := f32(render_data.width.top)
 			radii := [4]f32{cr.topLeft, cr.topRight, cr.bottomRight, cr.bottomLeft}
 
 			if radii == {0, 0, 0, 0} {
-				rectangle_lines(layer, bounds, color, thick)
+				rectangle_lines(layer, bounds, color, thickness)
 			} else {
-				rectangle_corners_lines(layer, bounds, radii, color, thick)
+				rectangle_corners_lines(layer, bounds, radii, color, thickness)
+			}
+		case clay.RenderCommandType.Custom: if custom_draw != nil {
+					custom_draw(layer, bounds, render_command.renderData.custom)
 				}
-		case clay.RenderCommandType.Custom:
 		}
 	}
 }
 
 // Render primitives. clear_color is the background fill before any layers are drawn.
-end :: proc(
-	device: ^sdl.GPUDevice,
-	window: ^sdl.Window,
-	clear_color: Color = BLACK,
-) {
+end :: proc(device: ^sdl.GPUDevice, window: ^sdl.Window, clear_color: Color = BLACK) {
 	cmd_buffer := sdl.AcquireGPUCommandBuffer(device)
 	if cmd_buffer == nil {
 		log.panicf("Failed to acquire GPU command buffer: %s", sdl.GetError())
@@ -561,13 +620,9 @@ end :: proc(
 	upload(device, copy_pass)
 	sdl.EndGPUCopyPass(copy_pass)
 
-	// Resize dynamic arrays
-	// TODO: This should only be called occasionally, not every frame.
-	resize_global()
-
 	swapchain_texture: ^sdl.GPUTexture
-	w, h: u32
-	if !sdl.WaitAndAcquireGPUSwapchainTexture(cmd_buffer, window, &swapchain_texture, &w, &h) {
+	width, height: u32
+	if !sdl.WaitAndAcquireGPUSwapchainTexture(cmd_buffer, window, &swapchain_texture, &width, &height) {
 		log.panicf("Failed to acquire swapchain texture: %s", sdl.GetError())
 	}
 
@@ -583,16 +638,16 @@ end :: proc(
 	render_texture := swapchain_texture
 
 	if use_msaa {
-		ensure_msaa_texture(device, sdl.GetGPUSwapchainTextureFormat(device, window), w, h)
+		ensure_msaa_texture(device, sdl.GetGPUSwapchainTextureFormat(device, window), width, height)
 		render_texture = GLOB.msaa_texture
 	}
 
-	cc := color_to_f32(clear_color)
+	clear_color_f32 := color_to_f32(clear_color)
 
 	// Draw layers. One render pass per layer; sub-batches draw in submission order within each scissor.
 	for &layer, index in GLOB.layers {
 		log.debug("Drawing layer", index)
-		draw_layer(device, window, cmd_buffer, render_texture, w, h, cc, &layer)
+		draw_layer(device, window, cmd_buffer, render_texture, width, height, clear_color_f32, &layer)
 	}
 
 	// Resolve MSAA render texture to the swapchain.
@@ -624,15 +679,15 @@ end :: proc(
 max_sample_count :: proc(device: ^sdl.GPUDevice, window: ^sdl.Window) -> sdl.GPUSampleCount {
 	format := sdl.GetGPUSwapchainTextureFormat(device, window)
 	counts := [?]sdl.GPUSampleCount{._8, ._4, ._2}
-	for sc in counts {
-		if sdl.GPUTextureSupportsSampleCount(device, format, sc) do return sc
+	for count in counts {
+		if sdl.GPUTextureSupportsSampleCount(device, format, count) do return count
 	}
 	return ._1
 }
 
 @(private = "file")
-ensure_msaa_texture :: proc(device: ^sdl.GPUDevice, format: sdl.GPUTextureFormat, w, h: u32) {
-	if GLOB.msaa_texture != nil && GLOB.msaa_w == w && GLOB.msaa_h == h {
+ensure_msaa_texture :: proc(device: ^sdl.GPUDevice, format: sdl.GPUTextureFormat, width, height: u32) {
+	if GLOB.msaa_texture != nil && GLOB.msaa_width == width && GLOB.msaa_height == height {
 		return
 	}
 	if GLOB.msaa_texture != nil {
@@ -644,18 +699,18 @@ ensure_msaa_texture :: proc(device: ^sdl.GPUDevice, format: sdl.GPUTextureFormat
 			type = .D2,
 			format = format,
 			usage = {.COLOR_TARGET},
-			width = w,
-			height = h,
+			width = width,
+			height = height,
 			layer_count_or_depth = 1,
 			num_levels = 1,
 			sample_count = GLOB.sample_count,
 		},
 	)
 	if GLOB.msaa_texture == nil {
-		log.panicf("Failed to create MSAA texture (%dx%d): %s", w, h, sdl.GetError())
+		log.panicf("Failed to create MSAA texture (%dx%d): %s", width, height, sdl.GetError())
 	}
-	GLOB.msaa_w = w
-	GLOB.msaa_h = h
+	GLOB.msaa_width = width
+	GLOB.msaa_height = height
 }
 
 // ---------------------------------------------------------------------------------------------------------------------
@@ -683,9 +738,21 @@ Vertex_Uniforms :: struct {
 }
 
 // Push projection, dpi scale, and rendering mode as a single uniform block (slot 0).
-push_globals :: proc(cmd_buffer: ^sdl.GPUCommandBuffer, w: f32, h: f32, mode: Draw_Mode = .Tessellated) {
+push_globals :: proc(
+	cmd_buffer: ^sdl.GPUCommandBuffer,
+	width: f32,
+	height: f32,
+	mode: Draw_Mode = .Tessellated,
+) {
 	globals := Vertex_Uniforms {
-		projection = ortho_rh(left = 0.0, top = 0.0, right = f32(w), bottom = f32(h), near = -1.0, far = 1.0),
+		projection = ortho_rh(
+			left = 0.0,
+			top = 0.0,
+			right = f32(width),
+			bottom = f32(height),
+			near = -1.0,
+			far = 1.0,
+		),
 		scale      = GLOB.dpi_scaling,
 		mode       = mode,
 	}
@@ -757,3 +824,115 @@ destroy_buffer :: proc(device: ^sdl.GPUDevice, buffer: ^Buffer) {
 	sdl.ReleaseGPUBuffer(device, buffer.gpu)
 	sdl.ReleaseGPUTransferBuffer(device, buffer.transfer)
 }
+
+// ---------------------------------------------------------------------------------------------------------------------
+// ----- Transform ------------------------
+// ---------------------------------------------------------------------------------------------------------------------
+
+// 2x3 affine transform for 2D pivot-rotation.
+// Used internally by rotation-aware drawing procs.
+Transform_2D :: struct {
+	m00, m01: f32, // row 0: rotation/scale
+	m10, m11: f32, // row 1: rotation/scale
+	tx, ty:   f32, // translation
+}
+
+// Build a pivot-rotation transform.
+//
+// Semantics (raylib-style):
+//   The point whose local coordinates equal `origin` lands at `pos` in world space.
+//   The rest of the shape rotates around that pivot.
+//
+// Formula:  p_world = pos + R(θ) · (p_local - origin)
+//
+// Parameters:
+//   pos          – world-space position where the pivot lands.
+//   origin       – pivot point in local space (measured from the shape's natural reference point).
+//   rotation_deg – rotation in degrees, counter-clockwise.
+//
+build_pivot_rotation :: proc(position: [2]f32, origin: [2]f32, rotation_deg: f32) -> Transform_2D {
+	radians := math.to_radians(rotation_deg)
+	cos_angle := math.cos(radians)
+	sin_angle := math.sin(radians)
+	return Transform_2D {
+		m00 = cos_angle,
+		m01 = -sin_angle,
+		m10 = sin_angle,
+		m11 = cos_angle,
+		tx = position.x - (cos_angle * origin.x - sin_angle * origin.y),
+		ty = position.y - (sin_angle * origin.x + cos_angle * origin.y),
+	}
+}
+
+// Apply the transform to a local-space point, producing a world-space point.
+apply_transform :: #force_inline proc(transform: Transform_2D, point: [2]f32) -> [2]f32 {
+	return {
+		transform.m00 * point.x + transform.m01 * point.y + transform.tx,
+		transform.m10 * point.x + transform.m11 * point.y + transform.ty,
+	}
+}
+
+// Fast-path check callers use BEFORE building a transform.
+// Returns true if either the origin is non-zero or rotation is non-zero,
+// meaning a transform actually needs to be computed.
+needs_transform :: #force_inline proc(origin: [2]f32, rotation: f32) -> bool {
+	return origin != {0, 0} || rotation != 0
+}
+
+// ---------------------------------------------------------------------------------------------------------------------
+// ----- Procedure Groups ------------------------
+// ---------------------------------------------------------------------------------------------------------------------
+
+center_of :: proc {
+	center_of_rectangle,
+	center_of_triangle,
+	center_of_text,
+}
+
+top_left_of :: proc {
+	top_left_of_rectangle,
+	top_left_of_triangle,
+	top_left_of_text,
+}
+
+top_of :: proc {
+	top_of_rectangle,
+	top_of_triangle,
+	top_of_text,
+}
+
+top_right_of :: proc {
+	top_right_of_rectangle,
+	top_right_of_triangle,
+	top_right_of_text,
+}
+
+left_of :: proc {
+	left_of_rectangle,
+	left_of_triangle,
+	left_of_text,
+}
+
+right_of :: proc {
+	right_of_rectangle,
+	right_of_triangle,
+	right_of_text,
+}
+
+bottom_left_of :: proc {
+	bottom_left_of_rectangle,
+	bottom_left_of_triangle,
+	bottom_left_of_text,
+}
+
+bottom_of :: proc {
+	bottom_of_rectangle,
+	bottom_of_triangle,
+	bottom_of_text,
+}
+
+bottom_right_of :: proc {
+	bottom_right_of_rectangle,
+	bottom_right_of_triangle,
+	bottom_right_of_text,
+}

draw/examples/hellope.odin

@@ -2,10 +2,9 @@ package examples
 
 import "../../draw"
 import "../../vendor/clay"
-import "core:c"
+import "core:math"
 import "core:os"
 import sdl "vendor:sdl3"
-import sdl_ttf "vendor:sdl3/ttf"
 
 JETBRAINS_MONO_REGULAR_RAW :: #load("fonts/JetBrainsMono-Regular.ttf")
 JETBRAINS_MONO_REGULAR: draw.Font_Id = max(draw.Font_Id) // Max so we crash if registration is forgotten
@@ -13,25 +12,28 @@ JETBRAINS_MONO_REGULAR: draw.Font_Id = max(draw.Font_Id) // Max so we crash if r
 hellope_shapes :: proc() {
 	if !sdl.Init({.VIDEO}) do os.exit(1)
 	window := sdl.CreateWindow("Hellope!", 500, 500, {.HIGH_PIXEL_DENSITY})
-	gpu := sdl.CreateGPUDevice({.MSL}, true, nil)
+	gpu := sdl.CreateGPUDevice(draw.PLATFORM_SHADER_FORMAT, true, nil)
 	if !sdl.ClaimWindowForGPUDevice(gpu, window) do os.exit(1)
 	if !draw.init(gpu, window) do os.exit(1)
 
+	spin_angle: f32 = 0
+
 	for {
 		defer free_all(context.temp_allocator)
 		ev: sdl.Event
 		for sdl.PollEvent(&ev) {
 			if ev.type == .QUIT do return
 		}
-		base_layer := draw.begin({w = 500, h = 500})
+		spin_angle += 1
+		base_layer := draw.begin({width = 500, height = 500})
 
 		// Background
 		draw.rectangle(base_layer, {0, 0, 500, 500}, {40, 40, 40, 255})
 
-		// Shapes demo
+		// ----- Shapes without rotation (existing demo) -----
 		draw.rectangle(base_layer, {20, 20, 200, 120}, {80, 120, 200, 255})
-		draw.rectangle_lines(base_layer, {20, 20, 200, 120}, draw.WHITE, thick = 2)
-		draw.rectangle_rounded(base_layer, {240, 20, 240, 120}, 0.3, {200, 80, 80, 255})
+		draw.rectangle_lines(base_layer, {20, 20, 200, 120}, draw.WHITE, thickness = 2)
+		draw.rectangle(base_layer, {240, 20, 240, 120}, {200, 80, 80, 255}, roundness = 0.3)
 		draw.rectangle_gradient(
 			base_layer,
 			{20, 160, 460, 60},
@@ -41,33 +43,86 @@ hellope_shapes :: proc() {
 			{255, 255, 0, 255},
 		)
 
-		draw.circle(base_layer, {120, 320}, 60, {100, 200, 100, 255})
-		draw.circle_lines(base_layer, {120, 320}, 60, draw.WHITE, thick = 2)
-		draw.circle_gradient(base_layer, {300, 320}, 60, {255, 200, 50, 255}, {200, 50, 50, 255})
-		draw.ring(base_layer, {430, 320}, 30, 55, 0, 270, {100, 100, 220, 255})
+		// ----- Rotation demos -----
 
-		draw.triangle(base_layer, {60, 420}, {180, 480}, {20, 480}, {220, 180, 60, 255})
-		draw.line(base_layer, {220, 420}, {460, 480}, {255, 255, 100, 255}, thick = 3)
-		draw.poly(base_layer, {350, 450}, 6, 40, {180, 100, 220, 255}, rotation = 30)
-		draw.poly_lines(base_layer, {350, 450}, 6, 40, draw.WHITE, rotation = 30, thick = 2)
+		// Rectangle rotating around its center
+		rect := draw.Rectangle{100, 320, 80, 50}
+		draw.rectangle(
+			base_layer,
+			rect,
+			{100, 200, 100, 255},
+			origin = draw.center_of(rect),
+			rotation = spin_angle,
+		)
+		draw.rectangle_lines(
+			base_layer,
+			rect,
+			draw.WHITE,
+			thickness = 2,
+			origin = draw.center_of(rect),
+			rotation = spin_angle,
+		)
+
+		// Rounded rectangle rotating around its center
+		rrect := draw.Rectangle{230, 300, 100, 80}
+		draw.rectangle(
+			base_layer,
+			rrect,
+			{200, 100, 200, 255},
+			roundness = 0.4,
+			origin = draw.center_of(rrect),
+			rotation = spin_angle,
+		)
+
+		// Ellipse rotating around its center (tilted ellipse)
+		draw.ellipse(base_layer, {410, 340}, 50, 30, {255, 200, 50, 255}, rotation = spin_angle)
+
+		// Circle orbiting a point (moon orbiting planet)
+		planet_pos := [2]f32{100, 450}
+		moon_pos := planet_pos + {0, -40}
+		draw.circle(base_layer, planet_pos, 8, {200, 200, 200, 255}) // planet (stationary)
+		draw.circle(base_layer, moon_pos, 5, {100, 150, 255, 255}, origin = {0, 40}, rotation = spin_angle) // moon orbiting
+
+		// Ring arc rotating in place
+		draw.ring(base_layer, {250, 450}, 15, 30, 0, 270, {100, 100, 220, 255}, rotation = spin_angle)
+
+		// Triangle rotating around its center
+		tv1 := [2]f32{350, 420}
+		tv2 := [2]f32{420, 480}
+		tv3 := [2]f32{340, 480}
+		draw.triangle(
+			base_layer,
+			tv1,
+			tv2,
+			tv3,
+			{220, 180, 60, 255},
+			origin = draw.center_of(tv1, tv2, tv3),
+			rotation = spin_angle,
+		)
+
+		// Polygon rotating around its center (already had rotation; now with origin for orbit)
+		draw.polygon(base_layer, {460, 450}, 6, 30, {180, 100, 220, 255}, rotation = spin_angle)
+		draw.polygon_lines(base_layer, {460, 450}, 6, 30, draw.WHITE, rotation = spin_angle, thickness = 2)
 
 		draw.end(gpu, window)
 	}
 }
 
 hellope_text :: proc() {
+	HELLOPE_ID :: 1
+	ROTATING_SENTENCE_ID :: 2
+	MEASURED_ID :: 3
+	CORNER_SPIN_ID :: 4
+
 	if !sdl.Init({.VIDEO}) do os.exit(1)
-	window := sdl.CreateWindow("Hellope!", 500, 500, {.HIGH_PIXEL_DENSITY})
-	gpu := sdl.CreateGPUDevice({.MSL}, true, nil)
+	window := sdl.CreateWindow("Hellope!", 600, 600, {.HIGH_PIXEL_DENSITY})
+	gpu := sdl.CreateGPUDevice(draw.PLATFORM_SHADER_FORMAT, true, nil)
 	if !sdl.ClaimWindowForGPUDevice(gpu, window) do os.exit(1)
 	if !draw.init(gpu, window) do os.exit(1)
 	JETBRAINS_MONO_REGULAR = draw.register_font(JETBRAINS_MONO_REGULAR_RAW)
 
 	FONT_SIZE :: u16(24)
-	TEXT_ID :: u32(1)
-
-	font := draw.get_font(JETBRAINS_MONO_REGULAR, FONT_SIZE)
-	dpi := sdl.GetWindowDisplayScale(window)
+	spin_angle: f32 = 0
 
 	for {
 		defer free_all(context.temp_allocator)
@@ -75,28 +130,74 @@ hellope_text :: proc() {
 		for sdl.PollEvent(&ev) {
 			if ev.type == .QUIT do return
 		}
-		base_layer := draw.begin({w = 500, h = 500})
+		spin_angle += 0.5
+		base_layer := draw.begin({width = 600, height = 600})
 
 		// Grey background
-		draw.rectangle(base_layer, {0, 0, 500, 500}, {127, 127, 127, 255})
+		draw.rectangle(base_layer, {0, 0, 600, 600}, {127, 127, 127, 255})
 
-		// Measure and center text
-		tw, th: c.int
-		sdl_ttf.GetStringSize(font, "Hellope!", 0, &tw, &th)
-		text_w := f32(tw) / dpi
-		text_h := f32(th) / dpi
-		pos_x := (500.0 - text_w) / 2.0
-		pos_y := (500.0 - text_h) / 2.0
+		// ----- Text API demos -----
 
-		txt := draw.text(
-			TEXT_ID,
+		// Cached text with id — TTF_Text reused across frames (good for text-heavy apps)
+		draw.text(
+			base_layer,
 			"Hellope!",
-			{pos_x, pos_y},
+			{300, 80},
+			JETBRAINS_MONO_REGULAR,
+			FONT_SIZE,
 			color = draw.WHITE,
-			font_id = JETBRAINS_MONO_REGULAR,
-			font_size = FONT_SIZE,
+			origin = draw.center_of("Hellope!", JETBRAINS_MONO_REGULAR, FONT_SIZE),
+			id = HELLOPE_ID,
+		)
+
+		// Rotating sentence — verifies multi-word text rotation around center
+		draw.text(
+			base_layer,
+			"Hellope World!",
+			{300, 250},
+			JETBRAINS_MONO_REGULAR,
+			FONT_SIZE,
+			color = {255, 200, 50, 255},
+			origin = draw.center_of("Hellope World!", JETBRAINS_MONO_REGULAR, FONT_SIZE),
+			rotation = spin_angle,
+			id = ROTATING_SENTENCE_ID,
+		)
+
+		// Uncached text (no id) — created and destroyed each frame, simplest usage
+		draw.text(
+			base_layer,
+			"Top-left anchored",
+			{20, 450},
+			JETBRAINS_MONO_REGULAR,
+			FONT_SIZE,
+			color = draw.WHITE,
+		)
+
+		// Measure text for manual layout
+		size := draw.measure_text("Measured!", JETBRAINS_MONO_REGULAR, FONT_SIZE)
+		draw.rectangle(base_layer, {300 - size.x / 2, 380, size.x, size.y}, {60, 60, 60, 200})
+		draw.text(
+			base_layer,
+			"Measured!",
+			{300, 380},
+			JETBRAINS_MONO_REGULAR,
+			FONT_SIZE,
+			color = draw.WHITE,
+			origin = draw.top_of("Measured!", JETBRAINS_MONO_REGULAR, FONT_SIZE),
+			id = MEASURED_ID,
+		)
+
+		// Rotating text anchored at top-left (no origin offset) — spins around top-left corner
+		draw.text(
+			base_layer,
+			"Corner spin",
+			{150, 530},
+			JETBRAINS_MONO_REGULAR,
+			FONT_SIZE,
+			color = {100, 200, 255, 255},
+			rotation = spin_angle,
+			id = CORNER_SPIN_ID,
 		)
-		draw.prepare_text(base_layer, txt)
 
 		draw.end(gpu, window)
 	}
@@ -105,14 +206,14 @@ hellope_text :: proc() {
 hellope_clay :: proc() {
 	if !sdl.Init({.VIDEO}) do os.exit(1)
 	window := sdl.CreateWindow("Hellope!", 500, 500, {.HIGH_PIXEL_DENSITY})
-	gpu := sdl.CreateGPUDevice({.MSL}, true, nil)
+	gpu := sdl.CreateGPUDevice(draw.PLATFORM_SHADER_FORMAT, true, nil)
 	if !sdl.ClaimWindowForGPUDevice(gpu, window) do os.exit(1)
 	if !draw.init(gpu, window) do os.exit(1)
 	JETBRAINS_MONO_REGULAR = draw.register_font(JETBRAINS_MONO_REGULAR_RAW)
 
 	text_config := clay.TextElementConfig {
 		fontId    = JETBRAINS_MONO_REGULAR,
-		fontSize  = 24,
+		fontSize  = 36,
 		textColor = {255, 255, 255, 255},
 	}
 
@@ -122,8 +223,8 @@ hellope_clay :: proc() {
 		for sdl.PollEvent(&ev) {
 			if ev.type == .QUIT do return
 		}
-		base_layer := draw.begin({w = 500, h = 500})
-		clay.SetLayoutDimensions({width = base_layer.bounds.w, height = base_layer.bounds.h})
+		base_layer := draw.begin({width = 500, height = 500})
+		clay.SetLayoutDimensions({width = base_layer.bounds.width, height = base_layer.bounds.height})
 		clay.BeginLayout()
 		if clay.UI()(
 		{
@@ -145,3 +246,106 @@ hellope_clay :: proc() {
 		draw.end(gpu, window)
 	}
 }
+
+hellope_custom :: proc() {
+	if !sdl.Init({.VIDEO}) do os.exit(1)
+	window := sdl.CreateWindow("Hellope Custom!", 600, 400, {.HIGH_PIXEL_DENSITY})
+	gpu := sdl.CreateGPUDevice(draw.PLATFORM_SHADER_FORMAT, true, nil)
+	if !sdl.ClaimWindowForGPUDevice(gpu, window) do os.exit(1)
+	if !draw.init(gpu, window) do os.exit(1)
+	JETBRAINS_MONO_REGULAR = draw.register_font(JETBRAINS_MONO_REGULAR_RAW)
+
+	text_config := clay.TextElementConfig {
+		fontId    = JETBRAINS_MONO_REGULAR,
+		fontSize  = 24,
+		textColor = {255, 255, 255, 255},
+	}
+
+	gauge := Gauge {
+		value = 0.73,
+		color = {50, 200, 100, 255},
+	}
+	gauge2 := Gauge {
+		value = 0.45,
+		color = {200, 100, 50, 255},
+	}
+	spin_angle: f32 = 0
+
+	for {
+		defer free_all(context.temp_allocator)
+		ev: sdl.Event
+		for sdl.PollEvent(&ev) {
+			if ev.type == .QUIT do return
+		}
+
+		spin_angle += 1
+		gauge.value = (math.sin(spin_angle * 0.02) + 1) * 0.5
+		gauge2.value = (math.cos(spin_angle * 0.03) + 1) * 0.5
+
+		base_layer := draw.begin({width = 600, height = 400})
+		clay.SetLayoutDimensions({width = base_layer.bounds.width, height = base_layer.bounds.height})
+		clay.BeginLayout()
+
+		if clay.UI()(
+		{
+			id = clay.ID("outer"),
+			layout = {
+				sizing = {clay.SizingGrow({}), clay.SizingGrow({})},
+				childAlignment = {x = .Center, y = .Center},
+				layoutDirection = .TopToBottom,
+				childGap = 20,
+			},
+			backgroundColor = {50, 50, 50, 255},
+		},
+		) {
+			if clay.UI()({id = clay.ID("title"), layout = {sizing = {clay.SizingFit({}), clay.SizingFit({})}}}) {
+				clay.Text("Custom Draw Demo", &text_config)
+			}
+
+			if clay.UI()(
+			{
+				id = clay.ID("gauge"),
+				layout = {sizing = {clay.SizingFixed(300), clay.SizingFixed(30)}},
+				custom = {customData = &gauge},
+				backgroundColor = {80, 80, 80, 255},
+			},
+			) {}
+
+			if clay.UI()(
+			{
+				id = clay.ID("gauge2"),
+				layout = {sizing = {clay.SizingFixed(300), clay.SizingFixed(30)}},
+				custom = {customData = &gauge2},
+				backgroundColor = {80, 80, 80, 255},
+			},
+			) {}
+		}
+
+		clay_batch := draw.ClayBatch {
+			bounds = base_layer.bounds,
+			cmds   = clay.EndLayout(),
+		}
+		draw.prepare_clay_batch(base_layer, &clay_batch, {0, 0}, custom_draw = draw_custom)
+		draw.end(gpu, window)
+	}
+
+	Gauge :: struct {
+		value: f32,
+		color: draw.Color,
+	}
+
+	draw_custom :: proc(layer: ^draw.Layer, bounds: draw.Rectangle, render_data: clay.CustomRenderData) {
+		gauge := cast(^Gauge)render_data.customData
+
+		// Background from clay's backgroundColor
+		draw.rectangle(layer, bounds, draw.color_from_clay(render_data.backgroundColor), roundness = 0.25)
+
+		// Fill bar
+		fill := bounds
+		fill.width *= gauge.value
+		draw.rectangle(layer, fill, gauge.color, roundness = 0.25)
+
+		// Border
+		draw.rectangle_lines(layer, bounds, draw.WHITE, thickness = 2, roundness = 0.25)
+	}
+}

draw/examples/main.odin

@@ -57,17 +57,18 @@ main :: proc() {
 	args := os.args
 	if len(args) < 2 {
 		fmt.eprintln("Usage: examples <example_name>")
-		fmt.eprintln("Available examples: hellope-shapes, hellope-text, hellope-clay")
+		fmt.eprintln("Available examples: hellope-shapes, hellope-text, hellope-clay, hellope-custom")
 		os.exit(1)
 	}
 
 	switch args[1] {
 	case "hellope-clay": hellope_clay()
+	case "hellope-custom": hellope_custom()
 	case "hellope-shapes": hellope_shapes()
 	case "hellope-text": hellope_text()
 	case:
 		fmt.eprintf("Unknown example: %v\n", args[1])
-		fmt.eprintln("Available examples: hellope-shapes, hellope-text, hellope-clay")
+		fmt.eprintln("Available examples: hellope-shapes, hellope-text, hellope-clay, hellope-custom")
 		os.exit(1)
 	}
 }

draw/pipeline_2d_base.odin

@@ -103,7 +103,8 @@ Primitive :: struct {
 	bounds:     [4]f32, //  0: min_x, min_y, max_x, max_y (world-space, pre-DPI)
 	color:      Color, // 16: u8x4, unpacked in shader via unpackUnorm4x8
 	kind_flags: u32, // 20: (kind as u32) | (flags as u32 << 8)
-	_pad:        [2]f32,       // 24: alignment to vec4 boundary
+	rotation:   f32, // 24: shader self-rotation in radians (used by RRect, Ellipse)
+	_pad:       f32, // 28: alignment to vec4 boundary
 	params:     Shape_Params, // 32: two vec4s of shape params
 }
 
@@ -143,32 +144,34 @@ create_pipeline_2d_base :: proc(
 		if pipeline.sdl_pipeline != nil do sdl.ReleaseGPUGraphicsPipeline(device, pipeline.sdl_pipeline)
 	}
 
-	when ODIN_OS == .Darwin {
-		base_2d_vert_raw := #load("shaders/generated/base_2d.vert.metal")
-		base_2d_frag_raw := #load("shaders/generated/base_2d.frag.metal")
-	} else {
-		base_2d_vert_raw := #load("shaders/generated/base_2d.vert.spv")
-		base_2d_frag_raw := #load("shaders/generated/base_2d.frag.spv")
+	active_shader_formats := sdl.GetGPUShaderFormats(device)
+	if PLATFORM_SHADER_FORMAT_FLAG not_in active_shader_formats {
+		log.errorf(
+			"draw: no embedded shader matches active GPU formats; this build supports %v but device reports %v",
+			PLATFORM_SHADER_FORMAT,
+			active_shader_formats,
+		)
+		return pipeline, false
 	}
 
-	log.debug("Loaded", len(base_2d_vert_raw), "vert bytes")
-	log.debug("Loaded", len(base_2d_frag_raw), "frag bytes")
+	log.debug("Loaded", len(BASE_VERT_2D_RAW), "vert bytes")
+	log.debug("Loaded", len(BASE_FRAG_2D_RAW), "frag bytes")
 
 	vert_info := sdl.GPUShaderCreateInfo {
-		code_size           = len(base_2d_vert_raw),
-		code                = raw_data(base_2d_vert_raw),
-		entrypoint          = ENTRY_POINT,
-		format              = SHADER_TYPE,
+		code_size           = len(BASE_VERT_2D_RAW),
+		code                = raw_data(BASE_VERT_2D_RAW),
+		entrypoint          = SHADER_ENTRY,
+		format              = {PLATFORM_SHADER_FORMAT_FLAG},
 		stage               = .VERTEX,
 		num_uniform_buffers = 1,
 		num_storage_buffers = 1,
 	}
 
 	frag_info := sdl.GPUShaderCreateInfo {
-		code_size    = len(base_2d_frag_raw),
-		code         = raw_data(base_2d_frag_raw),
-		entrypoint   = ENTRY_POINT,
-		format       = SHADER_TYPE,
+		code_size    = len(BASE_FRAG_2D_RAW),
+		code         = raw_data(BASE_FRAG_2D_RAW),
+		entrypoint   = SHADER_ENTRY,
+		format       = {PLATFORM_SHADER_FORMAT_FLAG},
 		stage        = .FRAGMENT,
 		num_samplers = 1,
 	}
@@ -239,31 +242,31 @@ create_pipeline_2d_base :: proc(
 	}
 
 	// Create vertex buffer
-	vb_ok: bool
-	pipeline.vertex_buffer, vb_ok = create_buffer(
+	vert_buf_ok: bool
+	pipeline.vertex_buffer, vert_buf_ok = create_buffer(
 		device,
 		size_of(Vertex) * BUFFER_INIT_SIZE,
 		sdl.GPUBufferUsageFlags{.VERTEX},
 	)
-	if !vb_ok do return pipeline, false
+	if !vert_buf_ok do return pipeline, false
 
 	// Create index buffer (used by text)
-	ib_ok: bool
-	pipeline.index_buffer, ib_ok = create_buffer(
+	idx_buf_ok: bool
+	pipeline.index_buffer, idx_buf_ok = create_buffer(
 		device,
 		size_of(c.int) * BUFFER_INIT_SIZE,
 		sdl.GPUBufferUsageFlags{.INDEX},
 	)
-	if !ib_ok do return pipeline, false
+	if !idx_buf_ok do return pipeline, false
 
 	// Create primitive storage buffer (used by SDF instanced drawing)
-	pb_ok: bool
-	pipeline.primitive_buffer, pb_ok = create_buffer(
+	prim_buf_ok: bool
+	pipeline.primitive_buffer, prim_buf_ok = create_buffer(
 		device,
 		size_of(Primitive) * BUFFER_INIT_SIZE,
 		sdl.GPUBufferUsageFlags{.GRAPHICS_STORAGE_READ},
 	)
-	if !pb_ok do return pipeline, false
+	if !prim_buf_ok do return pipeline, false
 
 	// Create static 6-vertex unit quad buffer (two triangles, TRIANGLELIST)
 	pipeline.unit_quad_buffer = sdl.CreateGPUBuffer(
@@ -296,73 +299,73 @@ create_pipeline_2d_base :: proc(
 
 	// Upload white pixel and unit quad data in a single command buffer
 	white_pixel := [4]u8{255, 255, 255, 255}
-	white_transfer := sdl.CreateGPUTransferBuffer(
+	white_transfer_buf := sdl.CreateGPUTransferBuffer(
 		device,
 		sdl.GPUTransferBufferCreateInfo{usage = .UPLOAD, size = size_of(white_pixel)},
 	)
-	if white_transfer == nil {
+	if white_transfer_buf == nil {
 		log.errorf("Failed to create white pixel transfer buffer: %s", sdl.GetError())
 		return pipeline, false
 	}
-	defer sdl.ReleaseGPUTransferBuffer(device, white_transfer)
+	defer sdl.ReleaseGPUTransferBuffer(device, white_transfer_buf)
 
-	white_ptr := sdl.MapGPUTransferBuffer(device, white_transfer, false)
+	white_ptr := sdl.MapGPUTransferBuffer(device, white_transfer_buf, false)
 	if white_ptr == nil {
 		log.errorf("Failed to map white pixel transfer buffer: %s", sdl.GetError())
 		return pipeline, false
 	}
 	mem.copy(white_ptr, &white_pixel, size_of(white_pixel))
-	sdl.UnmapGPUTransferBuffer(device, white_transfer)
+	sdl.UnmapGPUTransferBuffer(device, white_transfer_buf)
 
-	quad_verts := [6]Vertex{
-		{position = {0, 0}}, {position = {1, 0}}, {position = {0, 1}},
-		{position = {0, 1}}, {position = {1, 0}}, {position = {1, 1}},
+	quad_verts := [6]Vertex {
+		{position = {0, 0}},
+		{position = {1, 0}},
+		{position = {0, 1}},
+		{position = {0, 1}},
+		{position = {1, 0}},
+		{position = {1, 1}},
 	}
-	quad_transfer := sdl.CreateGPUTransferBuffer(
+	quad_transfer_buf := sdl.CreateGPUTransferBuffer(
 		device,
 		sdl.GPUTransferBufferCreateInfo{usage = .UPLOAD, size = size_of(quad_verts)},
 	)
-	if quad_transfer == nil {
+	if quad_transfer_buf == nil {
 		log.errorf("Failed to create unit quad transfer buffer: %s", sdl.GetError())
 		return pipeline, false
 	}
-	defer sdl.ReleaseGPUTransferBuffer(device, quad_transfer)
+	defer sdl.ReleaseGPUTransferBuffer(device, quad_transfer_buf)
 
-	quad_ptr := sdl.MapGPUTransferBuffer(device, quad_transfer, false)
+	quad_ptr := sdl.MapGPUTransferBuffer(device, quad_transfer_buf, false)
 	if quad_ptr == nil {
 		log.errorf("Failed to map unit quad transfer buffer: %s", sdl.GetError())
 		return pipeline, false
 	}
 	mem.copy(quad_ptr, &quad_verts, size_of(quad_verts))
-	sdl.UnmapGPUTransferBuffer(device, quad_transfer)
+	sdl.UnmapGPUTransferBuffer(device, quad_transfer_buf)
 
-	upload_cmd := sdl.AcquireGPUCommandBuffer(device)
-	if upload_cmd == nil {
+	upload_cmd_buffer := sdl.AcquireGPUCommandBuffer(device)
+	if upload_cmd_buffer == nil {
 		log.errorf("Failed to acquire command buffer for init upload: %s", sdl.GetError())
 		return pipeline, false
 	}
-	upload_pass := sdl.BeginGPUCopyPass(upload_cmd)
+	upload_pass := sdl.BeginGPUCopyPass(upload_cmd_buffer)
 
 	sdl.UploadToGPUTexture(
 		upload_pass,
-		sdl.GPUTextureTransferInfo{transfer_buffer = white_transfer},
+		sdl.GPUTextureTransferInfo{transfer_buffer = white_transfer_buf},
 		sdl.GPUTextureRegion{texture = pipeline.white_texture, w = 1, h = 1, d = 1},
 		false,
 	)
 
 	sdl.UploadToGPUBuffer(
 		upload_pass,
-		sdl.GPUTransferBufferLocation{transfer_buffer = quad_transfer},
-		sdl.GPUBufferRegion{
-			buffer = pipeline.unit_quad_buffer,
-			offset = 0,
-			size   = size_of(quad_verts),
-		},
+		sdl.GPUTransferBufferLocation{transfer_buffer = quad_transfer_buf},
+		sdl.GPUBufferRegion{buffer = pipeline.unit_quad_buffer, offset = 0, size = size_of(quad_verts)},
 		false,
 	)
 
 	sdl.EndGPUCopyPass(upload_pass)
-	if !sdl.SubmitGPUCommandBuffer(upload_cmd) {
+	if !sdl.SubmitGPUCommandBuffer(upload_cmd_buffer) {
 		log.errorf("Failed to submit init upload command buffer: %s", sdl.GetError())
 		return pipeline, false
 	}
@@ -409,16 +412,16 @@ upload :: proc(device: ^sdl.GPUDevice, pass: ^sdl.GPUCopyPass) {
 			sdl.GPUBufferUsageFlags{.VERTEX},
 		)
 
-		v_array := sdl.MapGPUTransferBuffer(device, GLOB.pipeline_2d_base.vertex_buffer.transfer, false)
-		if v_array == nil {
+		vert_array := sdl.MapGPUTransferBuffer(device, GLOB.pipeline_2d_base.vertex_buffer.transfer, false)
+		if vert_array == nil {
 			log.panicf("Failed to map vertex transfer buffer: %s", sdl.GetError())
 		}
 		if shape_vert_size > 0 {
-			mem.copy(v_array, raw_data(GLOB.tmp_shape_verts), int(shape_vert_size))
+			mem.copy(vert_array, raw_data(GLOB.tmp_shape_verts), int(shape_vert_size))
 		}
 		if text_vert_size > 0 {
 			mem.copy(
-				rawptr(uintptr(v_array) + uintptr(shape_vert_size)),
+				rawptr(uintptr(vert_array) + uintptr(shape_vert_size)),
 				raw_data(GLOB.tmp_text_verts),
 				int(text_vert_size),
 			)
@@ -428,11 +431,7 @@ upload :: proc(device: ^sdl.GPUDevice, pass: ^sdl.GPUCopyPass) {
 		sdl.UploadToGPUBuffer(
 			pass,
 			sdl.GPUTransferBufferLocation{transfer_buffer = GLOB.pipeline_2d_base.vertex_buffer.transfer},
-			sdl.GPUBufferRegion{
-				buffer = GLOB.pipeline_2d_base.vertex_buffer.gpu,
-				offset = 0,
-				size   = total_vert_size,
-			},
+			sdl.GPUBufferRegion{buffer = GLOB.pipeline_2d_base.vertex_buffer.gpu, offset = 0, size = total_vert_size},
 			false,
 		)
 	}
@@ -449,21 +448,17 @@ upload :: proc(device: ^sdl.GPUDevice, pass: ^sdl.GPUCopyPass) {
 			sdl.GPUBufferUsageFlags{.INDEX},
 		)
 
-		i_array := sdl.MapGPUTransferBuffer(device, GLOB.pipeline_2d_base.index_buffer.transfer, false)
-		if i_array == nil {
+		idx_array := sdl.MapGPUTransferBuffer(device, GLOB.pipeline_2d_base.index_buffer.transfer, false)
+		if idx_array == nil {
 			log.panicf("Failed to map index transfer buffer: %s", sdl.GetError())
 		}
-		mem.copy(i_array, raw_data(GLOB.tmp_text_indices), int(index_size))
+		mem.copy(idx_array, raw_data(GLOB.tmp_text_indices), int(index_size))
 		sdl.UnmapGPUTransferBuffer(device, GLOB.pipeline_2d_base.index_buffer.transfer)
 
 		sdl.UploadToGPUBuffer(
 			pass,
 			sdl.GPUTransferBufferLocation{transfer_buffer = GLOB.pipeline_2d_base.index_buffer.transfer},
-			sdl.GPUBufferRegion{
-				buffer = GLOB.pipeline_2d_base.index_buffer.gpu,
-				offset = 0,
-				size   = index_size,
-			},
+			sdl.GPUBufferRegion{buffer = GLOB.pipeline_2d_base.index_buffer.gpu, offset = 0, size = index_size},
 			false,
 		)
 	}
@@ -480,25 +475,17 @@ upload :: proc(device: ^sdl.GPUDevice, pass: ^sdl.GPUCopyPass) {
 			sdl.GPUBufferUsageFlags{.GRAPHICS_STORAGE_READ},
 		)
 
-		p_array := sdl.MapGPUTransferBuffer(
-			device, GLOB.pipeline_2d_base.primitive_buffer.transfer, false,
-		)
-		if p_array == nil {
+		prim_array := sdl.MapGPUTransferBuffer(device, GLOB.pipeline_2d_base.primitive_buffer.transfer, false)
+		if prim_array == nil {
 			log.panicf("Failed to map primitive transfer buffer: %s", sdl.GetError())
 		}
-		mem.copy(p_array, raw_data(GLOB.tmp_primitives), int(prim_size))
+		mem.copy(prim_array, raw_data(GLOB.tmp_primitives), int(prim_size))
 		sdl.UnmapGPUTransferBuffer(device, GLOB.pipeline_2d_base.primitive_buffer.transfer)
 
 		sdl.UploadToGPUBuffer(
 			pass,
-			sdl.GPUTransferBufferLocation{
-				transfer_buffer = GLOB.pipeline_2d_base.primitive_buffer.transfer,
-			},
-			sdl.GPUBufferRegion{
-				buffer = GLOB.pipeline_2d_base.primitive_buffer.gpu,
-				offset = 0,
-				size   = prim_size,
-			},
+			sdl.GPUTransferBufferLocation{transfer_buffer = GLOB.pipeline_2d_base.primitive_buffer.transfer},
+			sdl.GPUBufferRegion{buffer = GLOB.pipeline_2d_base.primitive_buffer.gpu, offset = 0, size = prim_size},
 			false,
 		)
 	}
@@ -510,8 +497,8 @@ draw_layer :: proc(
 	window: ^sdl.Window,
 	cmd_buffer: ^sdl.GPUCommandBuffer,
 	render_texture: ^sdl.GPUTexture,
-	swapchain_w: u32,
-	swapchain_h: u32,
+	swapchain_width: u32,
+	swapchain_height: u32,
 	clear_color: [4]f32,
 	layer: ^Layer,
 ) {
@@ -521,9 +508,7 @@ draw_layer :: proc(
 				cmd_buffer,
 				&sdl.GPUColorTargetInfo {
 					texture = render_texture,
-					clear_color = sdl.FColor {
-						clear_color[0], clear_color[1], clear_color[2], clear_color[3],
-					},
+					clear_color = sdl.FColor{clear_color[0], clear_color[1], clear_color[2], clear_color[3]},
 					load_op = .CLEAR,
 					store_op = .STORE,
 				},
@@ -540,9 +525,7 @@ draw_layer :: proc(
 		cmd_buffer,
 		&sdl.GPUColorTargetInfo {
 			texture = render_texture,
-			clear_color = sdl.FColor {
-				clear_color[0], clear_color[1], clear_color[2], clear_color[3],
-			},
+			clear_color = sdl.FColor{clear_color[0], clear_color[1], clear_color[2], clear_color[3]},
 			load_op = GLOB.cleared ? .LOAD : .CLEAR,
 			store_op = .STORE,
 		},
@@ -569,21 +552,19 @@ draw_layer :: proc(
 	)
 
 	// Shorthand aliases for frequently-used pipeline resources
-	main_vbuf := GLOB.pipeline_2d_base.vertex_buffer.gpu
+	main_vert_buf := GLOB.pipeline_2d_base.vertex_buffer.gpu
 	unit_quad := GLOB.pipeline_2d_base.unit_quad_buffer
-	white     := GLOB.pipeline_2d_base.white_texture
+	white_texture := GLOB.pipeline_2d_base.white_texture
 	sampler := GLOB.pipeline_2d_base.sampler
-	w         := f32(swapchain_w)
-	h         := f32(swapchain_h)
+	width := f32(swapchain_width)
+	height := f32(swapchain_height)
 
 	// Initial GPU state: tessellated mode, main vertex buffer, no atlas bound yet
-	push_globals(cmd_buffer, w, h, .Tessellated)
-	sdl.BindGPUVertexBuffers(
-		render_pass, 0, &sdl.GPUBufferBinding{buffer = main_vbuf, offset = 0}, 1,
-	)
+	push_globals(cmd_buffer, width, height, .Tessellated)
+	sdl.BindGPUVertexBuffers(render_pass, 0, &sdl.GPUBufferBinding{buffer = main_vert_buf, offset = 0}, 1)
 
 	current_mode: Draw_Mode = .Tessellated
-	current_vbuf := main_vbuf
+	current_vert_buf := main_vert_buf
 	current_atlas: ^sdl.GPUTexture
 
 	// Text vertices live after shape vertices in the GPU vertex buffer
@@ -596,76 +577,69 @@ draw_layer :: proc(
 			switch batch.kind {
 			case .Shapes:
 				if current_mode != .Tessellated {
-					push_globals(cmd_buffer, w, h, .Tessellated)
+					push_globals(cmd_buffer, width, height, .Tessellated)
 					current_mode = .Tessellated
 				}
-				if current_vbuf != main_vbuf {
-					sdl.BindGPUVertexBuffers(
-						render_pass, 0,
-						&sdl.GPUBufferBinding{buffer = main_vbuf, offset = 0}, 1,
-					)
-					current_vbuf = main_vbuf
+				if current_vert_buf != main_vert_buf {
+					sdl.BindGPUVertexBuffers(render_pass, 0, &sdl.GPUBufferBinding{buffer = main_vert_buf, offset = 0}, 1)
+					current_vert_buf = main_vert_buf
 				}
-				if current_atlas != white {
+				if current_atlas != white_texture {
 					sdl.BindGPUFragmentSamplers(
-						render_pass, 0,
-						&sdl.GPUTextureSamplerBinding{texture = white, sampler = sampler}, 1,
+						render_pass,
+						0,
+						&sdl.GPUTextureSamplerBinding{texture = white_texture, sampler = sampler},
+						1,
 					)
-					current_atlas = white
+					current_atlas = white_texture
 				}
 				sdl.DrawGPUPrimitives(render_pass, batch.count, 1, batch.offset, 0)
 
 			case .Text:
 				if current_mode != .Tessellated {
-					push_globals(cmd_buffer, w, h, .Tessellated)
+					push_globals(cmd_buffer, width, height, .Tessellated)
 					current_mode = .Tessellated
 				}
-				if current_vbuf != main_vbuf {
-					sdl.BindGPUVertexBuffers(
-						render_pass, 0,
-						&sdl.GPUBufferBinding{buffer = main_vbuf, offset = 0}, 1,
-					)
-					current_vbuf = main_vbuf
+				if current_vert_buf != main_vert_buf {
+					sdl.BindGPUVertexBuffers(render_pass, 0, &sdl.GPUBufferBinding{buffer = main_vert_buf, offset = 0}, 1)
+					current_vert_buf = main_vert_buf
 				}
-				chunk := &GLOB.tmp_text_batches[batch.offset]
-				if current_atlas != chunk.atlas_texture {
+				text_batch := &GLOB.tmp_text_batches[batch.offset]
+				if current_atlas != text_batch.atlas_texture {
 					sdl.BindGPUFragmentSamplers(
-						render_pass, 0,
-						&sdl.GPUTextureSamplerBinding {
-							texture = chunk.atlas_texture,
-							sampler = sampler,
-						},
+						render_pass,
+						0,
+						&sdl.GPUTextureSamplerBinding{texture = text_batch.atlas_texture, sampler = sampler},
 						1,
 					)
-					current_atlas = chunk.atlas_texture
+					current_atlas = text_batch.atlas_texture
 				}
 				sdl.DrawGPUIndexedPrimitives(
 					render_pass,
-					chunk.index_count,
+					text_batch.index_count,
 					1,
-					chunk.index_start,
-					i32(text_vertex_gpu_base + chunk.vertex_start),
+					text_batch.index_start,
+					i32(text_vertex_gpu_base + text_batch.vertex_start),
 					0,
 				)
 
 			case .SDF:
 				if current_mode != .SDF {
-					push_globals(cmd_buffer, w, h, .SDF)
+					push_globals(cmd_buffer, width, height, .SDF)
 					current_mode = .SDF
 				}
-				if current_vbuf != unit_quad {
-					sdl.BindGPUVertexBuffers(
-						render_pass, 0,
-						&sdl.GPUBufferBinding{buffer = unit_quad, offset = 0}, 1,
-					)
-					current_vbuf = unit_quad
+				if current_vert_buf != unit_quad {
+					sdl.BindGPUVertexBuffers(render_pass, 0, &sdl.GPUBufferBinding{buffer = unit_quad, offset = 0}, 1)
+					current_vert_buf = unit_quad
 				}
-				if current_atlas != white {
+				if current_atlas != white_texture {
 					sdl.BindGPUFragmentSamplers(
-						render_pass, 0,
-						&sdl.GPUTextureSamplerBinding{texture = white, sampler = sampler}, 1,
+						render_pass,
+						0,
+						&sdl.GPUTextureSamplerBinding{texture = white_texture, sampler = sampler},
+						1,
 					)
-					current_atlas = white
+					current_atlas = white_texture
 				}
 				sdl.DrawGPUPrimitives(render_pass, 6, batch.count, 0, batch.offset)
 			}

draw/shaders/generated/base_2d.frag.metal

@@ -24,32 +24,41 @@ struct main0_in
     float4 f_params [[user(locn2)]];
     float4 f_params2 [[user(locn3)]];
     uint f_kind_flags [[user(locn4)]];
+    float f_rotation [[user(locn5), flat]];
 };
 
+static inline __attribute__((always_inline))
+float2 apply_rotation(thread const float2& p, thread const float& angle)
+{
+    float cr = cos(-angle);
+    float sr = sin(-angle);
+    return float2x2(float2(cr, sr), float2(-sr, cr)) * p;
+}
+
 static inline __attribute__((always_inline))
 float sdRoundedBox(thread const float2& p, thread const float2& b, thread float4& r)
 {
-    float2 _56;
+    float2 _61;
     if (p.x > 0.0)
     {
-        _56 = r.xy;
+        _61 = r.xy;
     }
     else
     {
-        _56 = r.zw;
+        _61 = r.zw;
     }
-    r.x = _56.x;
-    r.y = _56.y;
-    float _73;
+    r.x = _61.x;
+    r.y = _61.y;
+    float _78;
     if (p.y > 0.0)
     {
-        _73 = r.x;
+        _78 = r.x;
     }
     else
     {
-        _73 = r.y;
+        _78 = r.y;
     }
-    r.x = _73;
+    r.x = _78;
     float2 q = (abs(p) - b) + float2(r.x);
     return (fast::min(fast::max(q.x, q.y), 0.0) + length(fast::max(q, float2(0.0)))) - r.x;
 }
@@ -142,16 +151,23 @@ fragment main0_out main0(main0_in in [[stage_in]], texture2d<float> tex [[textur
         float4 r = float4(in.f_params.zw, in.f_params2.xy);
         soft = fast::max(in.f_params2.z, 1.0);
         float stroke_px = in.f_params2.w;
-        float2 param = in.f_local_or_uv;
-        float2 param_1 = b;
-        float4 param_2 = r;
-        float _453 = sdRoundedBox(param, param_1, param_2);
-        d = _453;
+        float2 p_local = in.f_local_or_uv;
+        if (in.f_rotation != 0.0)
+        {
+            float2 param = p_local;
+            float param_1 = in.f_rotation;
+            p_local = apply_rotation(param, param_1);
+        }
+        float2 param_2 = p_local;
+        float2 param_3 = b;
+        float4 param_4 = r;
+        float _491 = sdRoundedBox(param_2, param_3, param_4);
+        d = _491;
         if ((flags & 1u) != 0u)
         {
-            float param_3 = d;
-            float param_4 = stroke_px;
-            d = sdf_stroke(param_3, param_4);
+            float param_5 = d;
+            float param_6 = stroke_px;
+            d = sdf_stroke(param_5, param_6);
         }
     }
     else
@@ -161,14 +177,14 @@ fragment main0_out main0(main0_in in [[stage_in]], texture2d<float> tex [[textur
             float radius = in.f_params.x;
             soft = fast::max(in.f_params.y, 1.0);
             float stroke_px_1 = in.f_params.z;
-            float2 param_5 = in.f_local_or_uv;
-            float param_6 = radius;
-            d = sdCircle(param_5, param_6);
+            float2 param_7 = in.f_local_or_uv;
+            float param_8 = radius;
+            d = sdCircle(param_7, param_8);
             if ((flags & 1u) != 0u)
             {
-                float param_7 = d;
-                float param_8 = stroke_px_1;
-                d = sdf_stroke(param_7, param_8);
+                float param_9 = d;
+                float param_10 = stroke_px_1;
+                d = sdf_stroke(param_9, param_10);
             }
         }
         else
@@ -178,15 +194,22 @@ fragment main0_out main0(main0_in in [[stage_in]], texture2d<float> tex [[textur
                 float2 ab = in.f_params.xy;
                 soft = fast::max(in.f_params.z, 1.0);
                 float stroke_px_2 = in.f_params.w;
-                float2 param_9 = in.f_local_or_uv;
-                float2 param_10 = ab;
-                float _511 = sdEllipse(param_9, param_10);
-                d = _511;
+                float2 p_local_1 = in.f_local_or_uv;
+                if (in.f_rotation != 0.0)
+                {
+                    float2 param_11 = p_local_1;
+                    float param_12 = in.f_rotation;
+                    p_local_1 = apply_rotation(param_11, param_12);
+                }
+                float2 param_13 = p_local_1;
+                float2 param_14 = ab;
+                float _560 = sdEllipse(param_13, param_14);
+                d = _560;
                 if ((flags & 1u) != 0u)
                 {
-                    float param_11 = d;
-                    float param_12 = stroke_px_2;
-                    d = sdf_stroke(param_11, param_12);
+                    float param_15 = d;
+                    float param_16 = stroke_px_2;
+                    d = sdf_stroke(param_15, param_16);
                 }
             }
             else
@@ -197,10 +220,10 @@ fragment main0_out main0(main0_in in [[stage_in]], texture2d<float> tex [[textur
                     float2 b_1 = in.f_params.zw;
                     float width = in.f_params2.x;
                     soft = fast::max(in.f_params2.y, 1.0);
-                    float2 param_13 = in.f_local_or_uv;
-                    float2 param_14 = a;
-                    float2 param_15 = b_1;
-                    d = sdSegment(param_13, param_14, param_15) - (width * 0.5);
+                    float2 param_17 = in.f_local_or_uv;
+                    float2 param_18 = a;
+                    float2 param_19 = b_1;
+                    d = sdSegment(param_17, param_18, param_19) - (width * 0.5);
                 }
                 else
                 {
@@ -218,26 +241,18 @@ fragment main0_out main0(main0_in in [[stage_in]], texture2d<float> tex [[textur
                         {
                             angle += 6.283185482025146484375;
                         }
-                        float ang_start = start_rad;
-                        float ang_end = end_rad;
-                        if (ang_start < 0.0)
-                        {
-                            ang_start += 6.283185482025146484375;
-                        }
-                        if (ang_end < 0.0)
-                        {
-                            ang_end += 6.283185482025146484375;
-                        }
-                        float _615;
+                        float ang_start = mod(start_rad, 6.283185482025146484375);
+                        float ang_end = mod(end_rad, 6.283185482025146484375);
+                        float _654;
                         if (ang_end > ang_start)
                         {
-                            _615 = float((angle >= ang_start) && (angle <= ang_end));
+                            _654 = float((angle >= ang_start) && (angle <= ang_end));
                         }
                         else
                         {
-                            _615 = float((angle >= ang_start) || (angle <= ang_end));
+                            _654 = float((angle >= ang_start) || (angle <= ang_end));
                         }
-                        float in_arc = _615;
+                        float in_arc = _654;
                         if (abs(ang_end - ang_start) >= 6.282185077667236328125)
                         {
                             in_arc = 1.0;
@@ -262,9 +277,9 @@ fragment main0_out main0(main0_in in [[stage_in]], texture2d<float> tex [[textur
                             d = (length(p) * cos(bn)) - radius_1;
                             if ((flags & 1u) != 0u)
                             {
-                                float param_16 = d;
-                                float param_17 = stroke_px_3;
-                                d = sdf_stroke(param_16, param_17);
+                                float param_20 = d;
+                                float param_21 = stroke_px_3;
+                                d = sdf_stroke(param_20, param_21);
                             }
                         }
                     }
@@ -272,9 +287,9 @@ fragment main0_out main0(main0_in in [[stage_in]], texture2d<float> tex [[textur
             }
         }
     }
-    float param_18 = d;
-    float param_19 = soft;
-    float alpha = sdf_alpha(param_18, param_19);
+    float param_22 = d;
+    float param_23 = soft;
+    float alpha = sdf_alpha(param_22, param_23);
     out.out_color = float4(in.f_color.xyz, in.f_color.w * alpha);
     return out;
 }

draw/shaders/generated/base_2d.vert.metal

@@ -15,7 +15,8 @@ struct Primitive
     float4 bounds;
     uint color;
     uint kind_flags;
-    float2 _pad;
+    float rotation;
+    float _pad;
     float4 params;
     float4 params2;
 };
@@ -25,7 +26,8 @@ struct Primitive_1
     float4 bounds;
     uint color;
     uint kind_flags;
-    float2 _pad;
+    float rotation;
+    float _pad;
     float4 params;
     float4 params2;
 };
@@ -42,6 +44,7 @@ struct main0_out
     float4 f_params [[user(locn2)]];
     float4 f_params2 [[user(locn3)]];
     uint f_kind_flags [[user(locn4)]];
+    float f_rotation [[user(locn5)]];
     float4 gl_Position [[position]];
 };
 
@@ -52,7 +55,7 @@ struct main0_in
     float4 v_color [[attribute(2)]];
 };
 
-vertex main0_out main0(main0_in in [[stage_in]], constant Uniforms& _12 [[buffer(0)]], const device Primitives& _70 [[buffer(1)]], uint gl_InstanceIndex [[instance_id]])
+vertex main0_out main0(main0_in in [[stage_in]], constant Uniforms& _12 [[buffer(0)]], const device Primitives& _72 [[buffer(1)]], uint gl_InstanceIndex [[instance_id]])
 {
     main0_out out = {};
     if (_12.mode == 0u)
@@ -62,17 +65,19 @@ vertex main0_out main0(main0_in in [[stage_in]], constant Uniforms& _12 [[buffer
         out.f_params = float4(0.0);
         out.f_params2 = float4(0.0);
         out.f_kind_flags = 0u;
+        out.f_rotation = 0.0;
         out.gl_Position = _12.projection * float4(in.v_position * _12.dpi_scale, 0.0, 1.0);
     }
     else
     {
         Primitive p;
-        p.bounds = _70.primitives[int(gl_InstanceIndex)].bounds;
-        p.color = _70.primitives[int(gl_InstanceIndex)].color;
-        p.kind_flags = _70.primitives[int(gl_InstanceIndex)].kind_flags;
-        p._pad = _70.primitives[int(gl_InstanceIndex)]._pad;
-        p.params = _70.primitives[int(gl_InstanceIndex)].params;
-        p.params2 = _70.primitives[int(gl_InstanceIndex)].params2;
+        p.bounds = _72.primitives[int(gl_InstanceIndex)].bounds;
+        p.color = _72.primitives[int(gl_InstanceIndex)].color;
+        p.kind_flags = _72.primitives[int(gl_InstanceIndex)].kind_flags;
+        p.rotation = _72.primitives[int(gl_InstanceIndex)].rotation;
+        p._pad = _72.primitives[int(gl_InstanceIndex)]._pad;
+        p.params = _72.primitives[int(gl_InstanceIndex)].params;
+        p.params2 = _72.primitives[int(gl_InstanceIndex)].params2;
         float2 corner = in.v_position;
         float2 world_pos = mix(p.bounds.xy, p.bounds.zw, corner);
         float2 center = (p.bounds.xy + p.bounds.zw) * 0.5;
@@ -81,6 +86,7 @@ vertex main0_out main0(main0_in in [[stage_in]], constant Uniforms& _12 [[buffer
         out.f_params = p.params;
         out.f_params2 = p.params2;
         out.f_kind_flags = p.kind_flags;
+        out.f_rotation = p.rotation;
         out.gl_Position = _12.projection * float4(world_pos * _12.dpi_scale, 0.0, 1.0);
     }
     return out;

draw/shaders/source/base_2d.frag

@@ -6,6 +6,7 @@ layout(location = 1) in vec2 f_local_or_uv;
 layout(location = 2) in vec4 f_params;
 layout(location = 3) in vec4 f_params2;
 layout(location = 4) flat in uint f_kind_flags;
+layout(location = 5) flat in float f_rotation;
 
 // --- Output ---
 layout(location = 0) out vec4 out_color;
@@ -82,6 +83,15 @@ float sdf_stroke(float d, float stroke_width) {
     return abs(d) - stroke_width * 0.5;
 }
 
+// Rotate a 2D point by the negative of the given angle (inverse rotation).
+// Used to rotate the sampling frame opposite to the shape's rotation so that
+// the SDF evaluates correctly for the rotated shape.
+vec2 apply_rotation(vec2 p, float angle) {
+    float cr = cos(-angle);
+    float sr = sin(-angle);
+    return mat2(cr, sr, -sr, cr) * p;
+}
+
 // ---------------------------------------------------------------------------
 // main
 // ---------------------------------------------------------------------------
@@ -113,11 +123,16 @@ void main() {
         soft = max(f_params2.z, 1.0);
         float stroke_px = f_params2.w;
 
-        d = sdRoundedBox(f_local_or_uv, b, r);
+        vec2 p_local = f_local_or_uv;
+        if (f_rotation != 0.0) {
+            p_local = apply_rotation(p_local, f_rotation);
+        }
+
+        d = sdRoundedBox(p_local, b, r);
         if ((flags & 1u) != 0u) d = sdf_stroke(d, stroke_px);
     }
     else if (kind == 2u) {
-        // Circle
+        // Circle — rotationally symmetric, no rotation needed
         float radius = f_params.x;
         soft = max(f_params.y, 1.0);
         float stroke_px = f_params.z;
@@ -131,11 +146,16 @@ void main() {
         soft = max(f_params.z, 1.0);
         float stroke_px = f_params.w;
 
-        d = sdEllipse(f_local_or_uv, ab);
+        vec2 p_local = f_local_or_uv;
+        if (f_rotation != 0.0) {
+            p_local = apply_rotation(p_local, f_rotation);
+        }
+
+        d = sdEllipse(p_local, ab);
         if ((flags & 1u) != 0u) d = sdf_stroke(d, stroke_px);
     }
     else if (kind == 4u) {
-        // Segment (capsule line)
+        // Segment (capsule line) — no rotation (excluded)
         vec2 a = f_params.xy; // already in local physical pixels
         vec2 b = f_params.zw;
         float width = f_params2.x;
@@ -144,7 +164,7 @@ void main() {
         d = sdSegment(f_local_or_uv, a, b) - width * 0.5;
     }
     else if (kind == 5u) {
-        // Ring / Arc
+        // Ring / Arc — rotation handled by CPU angle offset, no shader rotation
         float inner = f_params.x;
         float outer = f_params.y;
         float start_rad = f_params.z;
@@ -157,10 +177,8 @@ void main() {
         // Angular clip
         float angle = atan(f_local_or_uv.y, f_local_or_uv.x);
         if (angle < 0.0) angle += 2.0 * PI;
-        float ang_start = start_rad;
-        float ang_end = end_rad;
-        if (ang_start < 0.0) ang_start += 2.0 * PI;
-        if (ang_end < 0.0) ang_end += 2.0 * PI;
+        float ang_start = mod(start_rad, 2.0 * PI);
+        float ang_end = mod(end_rad, 2.0 * PI);
 
         float in_arc = (ang_end > ang_start)
             ? ((angle >= ang_start && angle <= ang_end) ? 1.0 : 0.0) : ((angle >= ang_start || angle <= ang_end) ? 1.0 : 0.0);
@@ -169,7 +187,7 @@ void main() {
         d = in_arc > 0.5 ? d_ring : 1e30;
     }
     else if (kind == 6u) {
-        // Regular N-gon
+        // Regular N-gon — has its own rotation in params, no Primitive.rotation used
         float radius = f_params.x;
         float rotation = f_params.y;
         float sides = f_params.z;

draw/shaders/source/base_2d.vert

@@ -11,6 +11,7 @@ layout(location = 1) out vec2 f_local_or_uv;
 layout(location = 2) out vec4 f_params;
 layout(location = 3) out vec4 f_params2;
 layout(location = 4) flat out uint f_kind_flags;
+layout(location = 5) flat out float f_rotation;
 
 // ---------- Uniforms (single block — avoids spirv-cross reordering on Metal) ----------
 layout(set = 1, binding = 0) uniform Uniforms {
@@ -24,7 +25,8 @@ struct Primitive {
     vec4 bounds; // 0-15:  min_x, min_y, max_x, max_y
     uint color; // 16-19: packed u8x4 (unpack with unpackUnorm4x8)
     uint kind_flags; // 20-23: kind | (flags << 8)
-    vec2 _pad; // 24-31: padding
+    float rotation; // 24-27: shader self-rotation in radians
+    float _pad; // 28-31: alignment padding
     vec4 params; // 32-47: shape params part 1
     vec4 params2; // 48-63: shape params part 2
 };
@@ -42,6 +44,7 @@ void main() {
         f_params = vec4(0.0);
         f_params2 = vec4(0.0);
         f_kind_flags = 0u;
+        f_rotation = 0.0;
 
         gl_Position = projection * vec4(v_position * dpi_scale, 0.0, 1.0);
     } else {
@@ -57,6 +60,7 @@ void main() {
         f_params = p.params;
         f_params2 = p.params2;
         f_kind_flags = p.kind_flags;
+        f_rotation = p.rotation;
 
         gl_Position = projection * vec4(world_pos * dpi_scale, 0.0, 1.0);
     }

draw/text.odin

@@ -1,6 +1,8 @@
 package draw
 
+import "core:c"
 import "core:log"
+import "core:strings"
 import sdl "vendor:sdl3"
 import sdl_ttf "vendor:sdl3/ttf"
 
@@ -11,11 +13,21 @@ Font_Key :: struct {
 	size: u16,
 }
 
+Cache_Source :: enum u8 {
+	Custom,
+	Clay,
+}
+
+Cache_Key :: struct {
+	id:     u32,
+	source: Cache_Source,
+}
+
 Text_Cache :: struct {
 	engine:     ^sdl_ttf.TextEngine,
 	font_bytes: [dynamic][]u8,
 	sdl_fonts:  map[Font_Key]^sdl_ttf.Font,
-	cache:      map[u32]^sdl_ttf.Text,
+	cache:      map[Cache_Key]^sdl_ttf.Text,
 }
 
 // Internal for fetching SDL TTF font pointer for rendering
@@ -66,37 +78,194 @@ register_font :: proc(bytes: []u8) -> (id: Font_Id, ok: bool) #optional_ok {
 }
 
 Text :: struct {
-	ref:      ^sdl_ttf.Text,
+	sdl_text: ^sdl_ttf.Text,
 	position: [2]f32,
 	color:    Color,
 }
 
-text :: proc(
-	id: u32,
-	txt: cstring,
-	pos: [2]f32,
-	font_id: Font_Id,
-	font_size: u16 = 44,
-	color: Color = {0, 0, 0, 255},
-) -> Text {
-	sdl_text := GLOB.text_cache.cache[id]
-	if sdl_text == nil {
-		sdl_text = sdl_ttf.CreateText(GLOB.text_cache.engine, get_font(font_id, font_size), txt, 0)
+// ---------------------------------------------------------------------------------------------------------------------
+// ----- Text cache lookup -------------
+// ---------------------------------------------------------------------------------------------------------------------
+
+// Shared cache lookup/create/update logic used by both the `text` proc and the Clay render path.
+// Returns the cached (or newly created) TTF_Text pointer.
+@(private)
+cache_get_or_update :: proc(key: Cache_Key, c_str: cstring, font: ^sdl_ttf.Font) -> ^sdl_ttf.Text {
+	existing, found := GLOB.text_cache.cache[key]
+	if !found {
+		sdl_text := sdl_ttf.CreateText(GLOB.text_cache.engine, font, c_str, 0)
 		if sdl_text == nil {
 			log.panicf("Failed to create SDL text: %s", sdl.GetError())
 		}
-		GLOB.text_cache.cache[id] = sdl_text
+		GLOB.text_cache.cache[key] = sdl_text
+		return sdl_text
 	} else {
-		//TODO if IDs are always unique and never change the underlying text
-		// can get rid of this
-		if !sdl_ttf.SetTextString(sdl_text, txt, 0) {
+		if !sdl_ttf.SetTextString(existing, c_str, 0) {
 			log.panicf("Failed to update SDL text string: %s", sdl.GetError())
 		}
+		return existing
+	}
+}
+
+// ---------------------------------------------------------------------------------------------------------------------
+// ----- Text drawing ------------------
+// ---------------------------------------------------------------------------------------------------------------------
+
+// Draw text at a position with optional rotation and origin.
+//
+// When `id` is nil (the default), the text is created and destroyed each frame — simple and
+// leak-free, appropriate for HUDs and moderate UI (up to ~50 text elements per frame).
+//
+// When `id` is set, the TTF_Text object is cached across frames keyed by the provided u32.
+// This avoids per-frame HarfBuzz shaping and allocation, which matters for text-heavy apps
+// (editors, terminals, chat). The user is responsible for choosing unique IDs per logical text
+// element and calling `clear_text_cache` or `clear_text_cache_entry` when cached entries are
+// no longer needed. Custom text IDs occupy a separate namespace from Clay text IDs, so
+// collisions between the two are impossible.
+//
+// `origin` is in pixels from the text block's top-left corner (raylib convention).
+// The point whose local coords equal `origin` lands at `pos` in world space.
+// `rotation` is in degrees, counter-clockwise.
+text :: proc(
+	layer: ^Layer,
+	text_string: string,
+	position: [2]f32,
+	font_id: Font_Id,
+	font_size: u16 = 44,
+	color: Color = BLACK,
+	origin: [2]f32 = {0, 0},
+	rotation: f32 = 0,
+	id: Maybe(u32) = nil,
+	temp_allocator := context.temp_allocator,
+) {
+	c_str := strings.clone_to_cstring(text_string, temp_allocator)
+
+	sdl_text: ^sdl_ttf.Text
+	cached := false
+
+	if cache_id, ok := id.?; ok {
+		cached = true
+		sdl_text = cache_get_or_update(Cache_Key{cache_id, .Custom}, c_str, get_font(font_id, font_size))
+	} else {
+		sdl_text = sdl_ttf.CreateText(GLOB.text_cache.engine, get_font(font_id, font_size), c_str, 0)
+		if sdl_text == nil {
+			log.panicf("Failed to create SDL text: %s", sdl.GetError())
+		}
 	}
 
-	return Text{sdl_text, pos, color}
+	if needs_transform(origin, rotation) {
+		dpi_scale := GLOB.dpi_scaling
+		transform := build_pivot_rotation(position * dpi_scale, origin * dpi_scale, rotation)
+		prepare_text_transformed(layer, Text{sdl_text, {0, 0}, color}, transform)
+	} else {
+		prepare_text(layer, Text{sdl_text, position, color})
+	}
+
+	if !cached {
+		// Don't destroy now — the draw data (atlas texture, vertices) is still referenced
+		// by the batch buffers until end() submits to the GPU. Deferred to clear_global().
+		append(&GLOB.tmp_uncached_text, sdl_text)
+	}
 }
 
+// ---------------------------------------------------------------------------------------------------------------------
+// ----- Public text measurement -------
+// ---------------------------------------------------------------------------------------------------------------------
+
+// Measure a string in logical pixels (pre-DPI-scaling) using the same font backend as the renderer.
+measure_text :: proc(
+	text_string: string,
+	font_id: Font_Id,
+	font_size: u16 = 44,
+	allocator := context.temp_allocator,
+) -> [2]f32 {
+	c_str := strings.clone_to_cstring(text_string, allocator)
+	width, height: c.int
+	if !sdl_ttf.GetStringSize(get_font(font_id, font_size), c_str, 0, &width, &height) {
+		log.panicf("Failed to measure text: %s", sdl.GetError())
+	}
+	return {f32(width) / GLOB.dpi_scaling, f32(height) / GLOB.dpi_scaling}
+}
+
+// ---------------------------------------------------------------------------------------------------------------------
+// ----- Text anchor helpers -----------
+// ---------------------------------------------------------------------------------------------------------------------
+
+center_of_text :: proc(text_string: string, font_id: Font_Id, font_size: u16 = 44) -> [2]f32 {
+	size := measure_text(text_string, font_id, font_size)
+	return size * 0.5
+}
+
+top_left_of_text :: proc(text_string: string, font_id: Font_Id, font_size: u16 = 44) -> [2]f32 {
+	return {0, 0}
+}
+
+top_of_text :: proc(text_string: string, font_id: Font_Id, font_size: u16 = 44) -> [2]f32 {
+	size := measure_text(text_string, font_id, font_size)
+	return {size.x * 0.5, 0}
+}
+
+top_right_of_text :: proc(text_string: string, font_id: Font_Id, font_size: u16 = 44) -> [2]f32 {
+	size := measure_text(text_string, font_id, font_size)
+	return {size.x, 0}
+}
+
+left_of_text :: proc(text_string: string, font_id: Font_Id, font_size: u16 = 44) -> [2]f32 {
+	size := measure_text(text_string, font_id, font_size)
+	return {0, size.y * 0.5}
+}
+
+right_of_text :: proc(text_string: string, font_id: Font_Id, font_size: u16 = 44) -> [2]f32 {
+	size := measure_text(text_string, font_id, font_size)
+	return {size.x, size.y * 0.5}
+}
+
+bottom_left_of_text :: proc(text_string: string, font_id: Font_Id, font_size: u16 = 44) -> [2]f32 {
+	size := measure_text(text_string, font_id, font_size)
+	return {0, size.y}
+}
+
+bottom_of_text :: proc(text_string: string, font_id: Font_Id, font_size: u16 = 44) -> [2]f32 {
+	size := measure_text(text_string, font_id, font_size)
+	return {size.x * 0.5, size.y}
+}
+
+bottom_right_of_text :: proc(text_string: string, font_id: Font_Id, font_size: u16 = 44) -> [2]f32 {
+	size := measure_text(text_string, font_id, font_size)
+	return size
+}
+
+// ---------------------------------------------------------------------------------------------------------------------
+// ----- Cache management --------------
+// ---------------------------------------------------------------------------------------------------------------------
+
+// Destroy all cached text objects (both custom and Clay entries). Call on scene transitions,
+// view changes, or periodically in apps that produce many distinct cached text entries over time.
+// After calling this, subsequent text draws with an `id` will re-create their cache entries.
+clear_text_cache :: proc() {
+	for _, sdl_text in GLOB.text_cache.cache {
+		sdl_ttf.DestroyText(sdl_text)
+	}
+	clear(&GLOB.text_cache.cache)
+}
+
+// Destroy a specific cached custom text entry by its u32 id (the same value passed to the
+// `text` proc's `id` parameter). This only affects custom text entries — Clay text entries
+// are managed internally and are not addressable by the user.
+// No-op if the id is not in the cache.
+clear_text_cache_entry :: proc(id: u32) {
+	key := Cache_Key{id, .Custom}
+	sdl_text, ok := GLOB.text_cache.cache[key]
+	if ok {
+		sdl_ttf.DestroyText(sdl_text)
+		delete_key(&GLOB.text_cache.cache, key)
+	}
+}
+
+// ---------------------------------------------------------------------------------------------------------------------
+// ----- Internal cache lifecycle ------
+// ---------------------------------------------------------------------------------------------------------------------
+
 @(private, require_results)
 init_text_cache :: proc(
 	device: ^sdl.GPUDevice,
@@ -121,7 +290,7 @@ init_text_cache :: proc(
 
 	text_cache = Text_Cache {
 		engine = engine,
-		cache  = make(map[u32]^sdl_ttf.Text, allocator = allocator),
+		cache  = make(map[Cache_Key]^sdl_ttf.Text, allocator = allocator),
 	}
 
 	log.debug("Done initializing text cache")
@@ -132,6 +301,9 @@ destroy_text_cache :: proc() {
 	for _, font in GLOB.text_cache.sdl_fonts {
 		sdl_ttf.CloseFont(font)
 	}
+	for _, sdl_text in GLOB.text_cache.cache {
+		sdl_ttf.DestroyText(sdl_text)
+	}
 	delete(GLOB.text_cache.sdl_fonts)
 	delete(GLOB.text_cache.font_bytes)
 	delete(GLOB.text_cache.cache)