anchor.blot.im\logs\2026-01-02_18-59.md
anchor.blot.im\logs\2026-01-03_12-30.md
anchor.blot.im\logs\2026-01-03_15-30.md
anchor.blot.im\logs\anchor-phase-1.md
anchor.blot.im\logs\anchor-phase-10-part-1.md
anchor.blot.im\logs\anchor-phase-10-part-10.md
anchor.blot.im\logs\anchor-phase-10-part-2.md
anchor.blot.im\logs\anchor-phase-10-part-3.md
anchor.blot.im\logs\anchor-phase-10-part-4.md
anchor.blot.im\logs\anchor-phase-10-part-5.md
anchor.blot.im\logs\anchor-phase-10-part-6.md
anchor.blot.im\logs\anchor-phase-10-part-7.md
anchor.blot.im\logs\anchor-phase-10-part-8.md
anchor.blot.im\logs\anchor-phase-10-part-9.md
anchor.blot.im\logs\anchor-phase-2.md
anchor.blot.im\logs\anchor-phase-3-part-1.md
anchor.blot.im\logs\anchor-phase-3-part-2.md
anchor.blot.im\logs\anchor-phase-3-part-3.md
anchor.blot.im\logs\anchor-phase-3-part-4.md
anchor.blot.im\logs\anchor-phase-4.md
anchor.blot.im\logs\anchor-phase-5.md
anchor.blot.im\logs\anchor-phase-6.md
anchor.blot.im\logs\anchor-phase-7.md
anchor.blot.im\logs\anchor-phase-8.md
anchor.blot.im\logs\anchor-phase-9.md
anchor.blot.im\logs\anchor-review-1.md
anchor.blot.im\logs\anchor-review-2.md
anchor.blot.im\logs\audible-book-recommendations-2.md
anchor.blot.im\logs\audible-book-recommendations.md
anchor.blot.im\logs\anchor-phase-10-part-11.md
... [71 more lines]

anchor.blot.im\logs\ebb-visual-polish.md
anchor.blot.im\logs\orblike-visual-aural-consistency.md
anchor.blot.im\logs\orblike-map-gen-wall-visuals.md
anchor.blot.im\logs\orblike-wall-visuals-ability-editions.md
anchor.blot.im\logs\orblike-more-visuals-edition-system.md
a327ex-site\logs\ebb-visual-polish.md
a327ex-site\logs\orblike-map-gen-wall-visuals.md
a327ex-site\logs\orblike-more-visuals-edition-system.md
a327ex-site\logs\orblike-visual-aural-consistency.md
a327ex-site\logs\orblike-wall-visuals-ability-editions.md
a327ex-site\logs\visual-experiments-1.md
a327ex-site\logs\visual-experiments-2.md

a327ex-site\logs\2026-01-01-07-37.md
a327ex-site\logs\2026-01-01-19-37.md
a327ex-site\logs\2026-01-02-10-00.md
a327ex-site\logs\2026-01-02-18-59.md
a327ex-site\logs\2026-01-03-12-30.md
a327ex-site\logs\2026-01-03-15-30.md
a327ex-site\logs\2026-01-03-18-20.md
a327ex-site\logs\2026-01-04-12-15.md
a327ex-site\logs\anchor-app-design.md
a327ex-site\logs\anchor-app-phase-1-part-1.md
a327ex-site\logs\anchor-app-ui-discussion.md
a327ex-site\logs\anchor-arena-final-and-neovim-diff-tool.md
a327ex-site\logs\anchor-arena-test.md
a327ex-site\logs\anchor-phase-1.md
a327ex-site\logs\anchor-phase-10-part-1.md
a327ex-site\logs\anchor-phase-10-part-10.md
a327ex-site\logs\anchor-phase-10-part-11.md
a327ex-site\logs\anchor-phase-10-part-2.md
a327ex-site\logs\anchor-phase-10-part-3.md
a327ex-site\logs\anchor-phase-10-part-4.md
a327ex-site\logs\anchor-phase-10-part-5.md
a327ex-site\logs\anchor-phase-10-part-6.md
a327ex-site\logs\anchor-phase-10-part-7.md
a327ex-site\logs\anchor-phase-10-part-8.md
a327ex-site\logs\anchor-phase-10-part-9.md
a327ex-site\logs\anchor-phase-2.md
a327ex-site\logs\anchor-phase-3-part-1.md
a327ex-site\logs\anchor-phase-3-part-2.md
a327ex-site\logs\anchor-phase-3-part-3.md
a327ex-site\logs\anchor-phase-3-part-4.md
... [71 more lines]

File content (724.1KB) exceeds maximum allowed size (256KB). Use offset and limit parameters to read specific portions of the file, or search for specific content instead of reading the whole file.

1    Title: Visual Experiments 2
2   Date: 2026-05-29 21:38:44
3   
4   # Visual Experiments 2
5   
6   ## Summary
7   
8   Continuation of the `ricochet-template` visual-experiments work (follows "Visual Experiments 1"). Extended the pattern × color × dither effect system with **local gradient patterns**, a **second per-pattern knob**, a **SHAPE axis** (field rendered as a grid of shapes), a **`ramp` color recipe** + an in-app **ramp editor**, then pivoted to **icon research** — built an Iconify→PNG pipeline, compared Phosphor/Tabler/Remix in a flat themed UI screen, chose **Tabler**, and rebuilt that screen as a full-width Tabler **UI kit**. Also fixed and documented a recurring **sprite-centering** engine gotcha and a latent `layer_line` framework-copy bug.
9   
10  **Local gradient patterns (linear + radial):**
11  - Added two patterns sampled from the draw's own `vUV` (rect-local) instead of `vPos` (world/global): `linear_gradient` (kind 10) and `radial_gradient` (kind 11). The field is locked to the rectangle's own box, so it moves/scales with the rect.
12  - Aspect-corrected via the quad's pixel size: hoisted a `vec2 quad_px` in the shader `main()`, set per shape branch (rect = `vShape0.xy`, circle = `vec2(vShape0.x)`, sprite = `vec2(texSize)`, etc.). Widened `effect_field(world_pos)` → `effect_field(world_pos, uv, quad_px)`; both call sites updated.
13  - `linear_gradient`: `param` = ANGLE (0..1 → 0..2π), aspect-corrected; ramp normalized corner-to-corner along the direction. `radial_gradient`: `param` = SHAPE, blends a true circle (inscribed, short side) ↔ rect-fitting ellipse.
14  - Via AskUserQuestion, the "how fast the gradient goes" control was decided to mean **transition steepness**, which is mechanically the existing universal **CONTRAST** modifier — so no new uniform; SCALE + SPEED are inert for gradients.
15  
16  **Second per-pattern knob (`param2`):**
17  - Added a general second intrinsic knob (`u_pattern_param2`) mirroring `param`, with a *per-pattern* slider range (`param2_min`/`param2_max`). Linear → **OFFSET** ([-1,1], slides the f=0.5 midline along the angle); radial → **RANGE** (0.25..3, scales the circle's reach). Conditional 5th MODIFIERS slider shown only when `param2_label` is set; sidebar height bumped.
18  
19  **SHAPE axis (4th axis) — field as a grid of shapes:**
20  - New `shape` axis (`shape_kinds`, `u_shape_kind`, `u_shape_size`), `apply_shape` between `effect_field` and dither. Cells anchored to screen space (`gl_FragCoord`), the same space as the dither. Shared SHAPE + CELL SIZE controls (like color/dither).
21  - First pass had three circle variants (size/fill/hard). User: "Let's try all these options" → made them selectable shape kinds.
22  - **Warping bug:** circles warped into blobs/crescents because the field was sampled per-fragment (f varies across a cell). Fixed by sampling the field **once per cell at the cell center**, reconstructing world_pos/uv there from screen-space derivatives (`dFdx`/`dFdy`) — exact for the 2D affine camera. Introduced `effect_field_shaped`.
23  - **Mask × fill decouple:** user wanted each circle to carry the effect's internal variation, not a flat tone. Generalized to two roles: **coverage** (mask, from the per-cell `fc` → true circle boundary) and **color/fill** (per-fragment `ff`). `shape_mask()` isolated as the single extension point for future image-stencil shapes.
24  - Removed `circle (fill)` per user ("too close to the raw effect"); kept `circle` (size) + `circle_hard`.
25  
26  **SNKRX editions — conceptual placement (no code):**
27  - Discussed where SNKRX's `dynamic` / `polychrome` / `palette_cycle` / `silver` editions fit. Conclusion: they're **COLOR recipes**. SNKRX fused (shimmer field + color transform + animation) into one "edition"; our pipeline already split those. So an edition decomposes: shimmer → pattern, animation → pattern `speed`, dither → dither, the color transform → the color axis.
28  - Key payoff: animated color is **free** — feed an animated pattern into a hue/ramp recipe and it sweeps over time without the recipe knowing about time; spatial color comes from gradient patterns. Only field-independent uniform time-cycling would need a new color-stage input (better expressed as a pattern). Ricochet caveat: a continuous rainbow fights the limited palette → use a **palette-token ramp** (quantized "polychrome").
29  
30  **`ramp` color recipe:**
... [90 more lines]

1    # ricochet-template — orientation for new instances
2   
3   ## ⛔ WORKFLOW RULE — DESIGN BRIEF BEFORE CODE ⛔
4   
5   **This rule has no exceptions and is valid for every single
6   non-trivial change, every session, every file in this project.**
7   
8   Before calling `Write` or `Edit` on project code, post a **design
9   brief** in chat — a conceptual explanation of the approach, NOT the
10  code itself. Then wait. The user will usually reply with questions
11  and improvements; iterate the brief with them until you reach a
12  conclusion. Only then write the code.
13  
14  ### What a design brief contains
15  
16  1. **Approach.** A high-level explanation of how each feature works —
17     what it does, the strategy, the key data it holds, its lifecycle.
18  2. **Interfaces.** How the feature connects to existing features:
19     which existing functions / files / globals it touches or reuses,
20     and how. Include this whenever the feature is not self-contained.
21  3. **Functions.** For each new or changed function:
22     - its name,
23     - what it does,
24     - which other functions it calls — selective, not exhaustive,
25     - its general structure, when that isn't already obvious.
26  
27  The brief is a conceptual map, not a spec and not pseudocode. Aim for
28  "a reader could predict the shape of the code from this."
29  
30  ### After the brief is agreed
... [667 more lines]

1    --[[
2     effect.lua — three-axis visual effect system (pattern × color × dither).
3   
4     THE 30-SECOND MENTAL MODEL
5       Every draw routes through assets/draw_shader.frag. The shader takes
6       three orthogonal selectors and composes them as:
7   
8           f = pattern(world_pos, time)          -- spatial structure + animation
9           f = dither(f, pixel_pos)              -- quantize to discrete levels
10          rgb = color(base_rgb, f, palette)     -- map f → palette token(s)
11  
12      Pattern owns "what the field looks like" AND "whether it moves over
13      time." Color is a pure recipe that picks (or mixes) palette tokens
14      by f. Dither inserts the pixel-art stipple between them. The three
15      are independent — any combination is valid.
16  
17    PUBLIC API
18      effect_setup(opts)                       install draw shader, cache its
19                                               GL program ID, push the palette
20      effect_set(layer, spec)                  write effect uniforms for this
21                                               layer's subsequent draws
22      effect_clear(layer)                      pattern/color/dither → 0 (passthrough)
23      effect_draw(layer, spec, fn, ...)        scoped form
24      effect_write_palette()                   re-push u_palette to the shader
25                                               (called automatically by palette_init
26                                               and effect_setup)
27  
28      Cycle helpers (return the NEXT/PREV name in each axis's cycle):
29        effect_next_pattern(cur), effect_prev_pattern(cur)
30        effect_next_color(cur),   effect_prev_color(cur)
... [425 more lines]

1    // =============================================================================
2   // ricochet-template — unified draw shader (single fragment shader for the
3   // whole project, installed via set_draw_shader as the engine's default).
4   //
5   // THREE-AXIS MODEL
6   //   Every fragment is processed as:
7   //
8   //       f   = pattern(world_pos, time, scale, param)   // [0, 1]
9   //       f   = dither(f, pixel_pos)                     // {0, 1/n, …, 1}
10  //       rgb = color(base_rgb, f, palette[color_a], palette[color_b])
11  //
12  //   Pattern owns spatial structure AND animation. Color is a pure recipe
13  //   that picks/mixes palette tokens by f. Dither inserts the pixel-art
14  //   stipple between them. The three are independent — any combination is
15  //   valid. See effect.lua for the registered names per axis.
16  //
17  // VARYING INPUTS (from engine vertex shader; do not rename)
18  //   vPos       World-space fragment position. For camera-attached layers
19  //              this is camera-transformed world space.
20  //   vUV        0..1 UV inside the draw's quad (or sprite UV for sprites).
21  //   vColor     Per-vertex RGBA color (the `color` arg passed to layer_*).
22  //   vType      Shape dispatch: 0=rect, 1=circle, 2=sprite, 3=line/capsule,
23  //              4=triangle, 5=polygon, 6=rounded rect.
24  //   vShape0..4 Per-shape parameter bundle.
25  //   vAddColor  Per-vertex additive RGB offset (engine 'flash' channel).
26  //
27  // UNIFORMS YOU SET FROM LUA (via effect.lua's effect_set)
28  //   u_pattern_kind   int   1=organic 2=solid 3=plasma 4=waves 5=marble
29  //                          6=sine_grid 7=wood 8=truchet 9=contours
30  //                          10=linear_gradient 11=radial_gradient (LOCAL)
... [721 more lines]

1    --[[
2     ricochet-template — visual style + three-axis effect template.
3   
4     Entry point. Owns:
5       - configuration (resolution, render mode, monitor)
6       - boot order (palette → fonts → effect/shader → ui modules)
7       - layers (game, ui), camera
8       - current effect-spec state (pattern, color, dither, color_a/b, scale)
9       - sidebar control panel + keyboard input (both mutate the same state)
10      - test-scene draw (sample shapes through the effect pipeline)
11  
12    Delegates:
13      - palette.lua     — dark/light palette tables + palette_init
14      - effect.lua      — pattern × color × dither spec API + cycle helpers
15      - rectcut.lua     — RectCut layout primitive (cut/get/add/contract/...)
16      - ui_state.lua    — frame-local hot/active/focus + ui_claim_hot
17      - ui.lua          — five Ricochet-style widgets that mutate effect state
18  
19    Goal: visually evaluate any (pattern, color, dither, color_a, color_b,
20    scale) combination by clicking sidebar controls or pressing the hotkeys.
21    Both input modes update the same caller-owned state.
22  ]]
23  
24  require('anchor')({
25    width            = 1920,
26    height           = 1080,
27    title            = "ricochet-template",
28    scale            = 1,
29    filter           = "smooth",
30    render_uncapped  = true,
... [489 more lines]

1    --[[
2     showcase.lua — flat "UI KIT" screen (Tab toggles it from the pattern gallery).
3   
4     A live catalogue of themed UI elements built with the chosen icon pack
5     (TABLER) so the whole vocabulary can be eyeballed in the Ricochet theme.
6     Everything is FLAT: raw palette colors on `ui_layer`, NO effect pipeline.
7   
8     Icons are 44px white PNGs at assets/icons/tabler/<concept>.png, loaded into
9     the global `icons[concept]` table by main.lua and tinted to a palette token
10    on draw (white * token = token). Generated by tools/fetch_icons.mjs.
11  
12    Three columns of element groups (full screen width now that it's one pack):
13      HUD          resource badges, vitals bars, ability bar, life pips, wave bar
14      PROGRESSION  upgrade card, tooltip, leaderboard list
15      CONTROLS     tabs, toggle/checkbox/radio, stepper, stat block, keycaps, tags
16  
17    DRAW ORIGIN REMINDER: layer_image / layer_circle are CENTERED at (x,y);
18    layer_rectangle / layer_text are TOP-LEFT. (See the framework gotcha note.)
19  ]]
20  
21  local COL_W = 580
22  local COL_X = { 40, 670, 1300 }
23  local TOP   = 120
24  local HAIR  = 1
25  local ICON  = 44   -- matches the rasterized PNG size (tools/fetch_icons.mjs SIZE)
26  
27  -- ============================================================================
28  -- primitives (all draw to ui_layer, screen-space)
29  -- ============================================================================
30  
... [341 more lines]

Found 1 file
ricochet-template\anchor\layer.lua

E:/a327ex/ricochet-template/assets/:
total 932
drwxr-xr-x 1 adn 197121      0 May 29 20:29 ./
drwxr-xr-x 1 adn 197121      0 May 29 21:26 ../
-rw-r--r-- 1 adn 197121 598060 May 28 05:55 CascadiaCode-Regular.ttf
-rw-r--r-- 1 adn 197121 200500 May 28 13:02 IBMPlexSans-Regular.ttf
-rw-r--r-- 1 adn 197121 107948 May 28 13:02 Oswald-Medium.ttf
-rw-r--r-- 1 adn 197121  35530 May 29 10:04 draw_shader.frag
drwxr-xr-x 1 adn 197121      0 May 29 21:22 icons/

E:/a327ex/ricochet-template/assets/icons/:
total 4
drwxr-xr-x 1 adn 197121 0 May 29 21:22 ./
drwxr-xr-x 1 adn 197121 0 May 29 20:29 ../
drwxr-xr-x 1 adn 197121 0 May 29 21:22 tabler/

E:/a327ex/ricochet-template/tools/:
total 25
drwxr-xr-x 1 adn 197121    0 May 29 21:22 ./
drwxr-xr-x 1 adn 197121    0 May 29 21:26 ../
-rw-r--r-- 1 adn 197121 2117 May 29 21:05 compare_packs.mjs
-rw-r--r-- 1 adn 197121 2612 May 29 21:22 fetch_icons.mjs
drwxr-xr-x 1 adn 197121    0 May 29 20:29 node_modules/
-rw-r--r-- 1 adn 197121 7252 May 29 20:29 package-lock.json
-rw-r--r-- 1 adn 197121  247 May 29 20:28 package.json

70:  shader_set_texture = layer_shader_set_texture,
71-  get_texture = layer_get_texture,
72-  reset_effects = layer_reset_effects,
73-  clear = layer_clear,
74-  render = layer_render,
75-  draw_from = layer_draw_from,
76-  stencil_mask = layer_stencil_mask,
77-  stencil_test = layer_stencil_test,
78-  stencil_test_inverse = layer_stencil_test_inverse,
79-  stencil_off = layer_stencil_off,
80-}
81-
82---- Resolve layer state table or raw C handle (lightuserdata) for engine calls.
83-local function lyr_handle(lyr)
84-  if type(lyr) == 'table' then
85-    return lyr.handle
86-  end
87-  return lyr
88-end
89-
90---- Create a layer state table and optionally register in global `layers`.
--
219:function layer_shader_set_float(lyr, shader, name, value)
220-  eng.shader_set_float(lyr_handle(lyr), shader, name, value)
221-end
222-
223:function layer_shader_set_vec2(lyr, shader, name, x, y)
224-  eng.shader_set_vec2(lyr_handle(lyr), shader, name, x, y)
225-end
226-
... [29 more lines]

Binary file anchor/../anchor.exe matches
---- engine src ----
392:    COMMAND_SET_UNIFORM_TEXTURE, // Bind a texture to a sampler uniform
3943:static void layer_shader_set_texture(Layer* layer, GLuint shader, const char* name, GLuint tex_id, int unit) {
3950:    cmd->type = COMMAND_SET_UNIFORM_TEXTURE;
3988:    GLint tex_loc = glGetUniformLocation(shader, "u_texture");
4674:            cmd->type == COMMAND_SET_UNIFORM_TEXTURE) {
4693:                case COMMAND_SET_UNIFORM_TEXTURE: {
5810:// layer_shader_set_texture(layer, shader, name, texture_userdata, unit)
5811:static int l_layer_shader_set_texture(lua_State* L) {
5818:        layer_shader_set_texture(layer, shader, name, tex->id, unit);
5924:        GLint tex_loc = glGetUniformLocation(shader, "u_texture");
9472:    lua_register(L, "layer_shader_set_texture", l_layer_shader_set_texture);
9770:    "uniform sampler2D u_texture;\n"
9871:    "            ivec2 texSize = textureSize(u_texture, 0);\n"
9874:    "        vec4 texColor = texture(u_texture, sampleUV);\n"

3940 // Set a texture uniform on a shader (binds to texture unit 1+)
3941    // params[0] = texture unit index (1, 2, etc — 0 is reserved for the layer's own texture)
3942    // texture_id = the GL texture handle
3943    static void layer_shader_set_texture(Layer* layer, GLuint shader, const char* name, GLuint tex_id, int unit) {
3944        if (layer->command_count >= MAX_COMMAND_CAPACITY) return;
3945        GLint loc = glGetUniformLocation(shader, name);
3946        if (loc < 0) return;
3947    
3948        DrawCommand* cmd = &layer->commands[layer->command_count++];
3949        memset(cmd, 0, sizeof(DrawCommand));
3950        cmd->type = COMMAND_SET_UNIFORM_TEXTURE;
3951        cmd->shader_id = shader;
3952        cmd->uniform_location = (uint32_t)loc;
3953        cmd->texture_id = tex_id;
3954        cmd->params[0] = (float)unit;
3955    }
3956    
3957    // Execute shader application (ping-pong): read from current buffer, apply shader, write to alternate
3958    // Called during command processing when COMMAND_APPLY_SHADER is encountered
3959    static void execute_apply_shader(Layer* layer, GLuint shader) {
3960        // Ensure effect buffer exists
3961        layer_ensure_effect_buffer(layer);
3962    
3963        // Determine source and destination based on current state
3964        GLuint src_tex, dst_fbo;
3965        if (layer->textures_swapped) {
3966            src_tex = layer->effect_texture;
3967            dst_fbo = layer->fbo;
3968        } else {
3969            src_tex = layer->color_texture;
... [45 more lines]

4660     glUseProgram(shader_program);
4661        GLint aa_loc = glGetUniformLocation(shader_program, "u_aa_width");
4662        float aa_width = (layer->filter_mode == FILTER_SMOOTH) ? 1.0f : 0.0f;
4663        glUniform1f(aa_loc, aa_width);
4664    
4665        for (int i = 0; i < layer->command_count; i++) {
4666            const DrawCommand* cmd = &layer->commands[i];
4667    
4668            // Handle uniform setting commands
4669            // These need to temporarily switch programs, so flush batch first and restore after
4670            if (cmd->type == COMMAND_SET_UNIFORM_FLOAT ||
4671                cmd->type == COMMAND_SET_UNIFORM_VEC2 ||
4672                cmd->type == COMMAND_SET_UNIFORM_VEC4 ||
4673                cmd->type == COMMAND_SET_UNIFORM_INT ||
4674                cmd->type == COMMAND_SET_UNIFORM_TEXTURE) {
4675                // Flush any pending draws before switching programs
4676                batch_flush();
4677                current_batch_texture = 0;
4678    
4679                glUseProgram(cmd->shader_id);
4680                switch (cmd->type) {
4681                    case COMMAND_SET_UNIFORM_FLOAT:
4682                        glUniform1f((GLint)cmd->uniform_location, cmd->params[0]);
4683                        break;
4684                    case COMMAND_SET_UNIFORM_VEC2:
4685                        glUniform2f((GLint)cmd->uniform_location, cmd->params[0], cmd->params[1]);
4686                        break;
4687                    case COMMAND_SET_UNIFORM_VEC4:
4688                        glUniform4f((GLint)cmd->uniform_location, cmd->params[0], cmd->params[1], cmd->params[2], cmd->params[3]);
4689                        break;
... [60 more lines]

868:    glBindTexture(GL_TEXTURE_2D, tex->id);
874:    glBindTexture(GL_TEXTURE_2D, 0);
897:    glBindTexture(GL_TEXTURE_2D, tex->id);
903:    glBindTexture(GL_TEXTURE_2D, 0);
1210:        glBindTexture(GL_TEXTURE_2D, font->atlas_texture);
1217:        glBindTexture(GL_TEXTURE_2D, 0);
1922:    glBindTexture(GL_TEXTURE_2D, layer->color_texture);
1994:    glBindTexture(GL_TEXTURE_2D, layer->effect_texture);
2015:    glBindTexture(GL_TEXTURE_2D, 0);
2364:static GLuint current_batch_texture = 0;  // Currently bound texture for batching
3992:    glActiveTexture(GL_TEXTURE0);
3993:    glBindTexture(GL_TEXTURE_2D, src_tex);
3997:        glActiveTexture(GL_TEXTURE1);
3998:        glBindTexture(GL_TEXTURE_2D, layer->extra_texture);
4000:        glActiveTexture(GL_TEXTURE0);
4010:        glActiveTexture(GL_TEXTURE1);
4011:        glBindTexture(GL_TEXTURE_2D, 0);
4012:        glActiveTexture(GL_TEXTURE0);
4024:static void batch_flush(void) {
4028:    if (current_batch_texture != 0) {
4029:        glActiveTexture(GL_TEXTURE0);
4030:        glBindTexture(GL_TEXTURE_2D, current_batch_texture);
4490:static void process_sprite(const DrawCommand* cmd) {
4497:    if (current_batch_texture != cmd->texture_id && batch_vertex_count > 0) {
4498:        batch_flush();
4500:    current_batch_texture = cmd->texture_id;
4545:    if (current_batch_texture != cmd->texture_id && batch_vertex_count > 0) {
4546:        batch_flush();
4548:    current_batch_texture = cmd->texture_id;
4579:static void process_spritesheet_frame(const DrawCommand* cmd) {
... [30 more lines]

4024 static void batch_flush(void) {
4025        if (batch_vertex_count == 0) return;
4026    
4027        // Bind texture if we have one (for sprites)
4028        if (current_batch_texture != 0) {
4029            glActiveTexture(GL_TEXTURE0);
4030            glBindTexture(GL_TEXTURE_2D, current_batch_texture);
4031        }
4032    
4033        glBindVertexArray(vao);
4034        glBindBuffer(GL_ARRAY_BUFFER, vbo);
4035        glBufferSubData(GL_ARRAY_BUFFER, 0,
4036                        batch_vertex_count * VERTEX_FLOATS * sizeof(float),
4037                        batch_vertices);
4038        glDrawArrays(GL_TRIANGLES, 0, batch_vertex_count);
4039        glBindVertexArray(0);
4040    
4041        batch_vertex_count = 0;
4042        draw_calls++;
4043    }
4044    
4045    // Process a rectangle command (SDF-based, UV-space approach)
4046    // The SDF is computed in local quad space using UV coordinates.
4047    // This handles rotation correctly because UV interpolation implicitly
4048    // provides the inverse rotation.
4049    static void process_rectangle(const DrawCommand* cmd) {
4050        float x = cmd->params[0];
4051        float y = cmd->params[1];
4052        float w = cmd->params[2];
4053        float h = cmd->params[3];
... [25 more lines]

airborne.lua
area.lua
beam.lua
edition.lua
findings.md
flicker.lua
fx.lua
geometry.lua
gradient.lua
init.lua
palette.lua
player.lua
projectile.lua
prop.lua
seeker.lua
sound_tuner.lua
sound_tuning.lua
stats.lua
symbols.lua
ui/
wall.lua
=== grep effect/edition/dither/shader in snkrx ===
snkrx/area.lua
snkrx/beam.lua
snkrx/edition.lua
snkrx/findings.md
snkrx/flicker.lua
snkrx/fx.lua
snkrx/geometry.lua
snkrx/gradient.lua
... [15 more lines]

=== main.lua require/anchor config ===
--[[
  snkrx-template / main.lua — clean reference scene.

  This is your file. It is a complete, minimal scene: a player you can
  move (WASD / arrows), the five-layer render stack, a camera, physics,
  the update / draw skeleton, AND the UI render pipeline — all wired
  and ready, but firing no attacks and drawing no UI widgets. Read it
  top to bottom; it is the contract you build on.

  TWO TOOLKITS, ONE REQUIRE.  `require('snkrx')` pulls in BOTH:

    snkrx/<carrier>.lua     attack toolkit — projectile / area / beam
                            / flicker / prop, the edition draw-shader
                            stack, fx, sounds, example entities.
    snkrx/ui/<module>.lua   UI toolkit — panel / button / slot / bar /
                            tabs / checkbox / radio / slider /
                            list_row / card / text_input / modal /
                            bracket_tag / stat_chip / pips / ...

  Both are plain, readable Lua, not opaque modules — open the files.

  ADDING AN ATTACK.  Every carrier already has its object list and its
  update / draw loop wired below. To add an attack:
    1. bind a key in the INPUT section,
    2. call the carrier's spawn_* helper on that press inside update()
       — see the ATTACKS comment block there.
  Nothing else: no new list, no new loop.

  DRAWING UI.  The UI toolkit is immediate-mode: call widget functions
... [108 more lines]

66:require('anchor')({
67:  width  = 480,
68:  height = 270,
70:  scale  = 3,
74:-- The SNKRX attack toolkit. Must come AFTER require('anchor') above —
=== edition draw_shader.frag field/dither size ===
main.lua:67:  width  = 480,
main.lua:68:  height = 270,
snkrx/edition.lua:20:                                                (+ u_field_scale, + u_dither_mode)
snkrx/edition.lua:69:    modes (30, 31) ignore u_dither_mode.
snkrx/edition.lua:72:    are computed directly from gl_FragCoord — no const-array matrices, so
snkrx/edition.lua:190:  layer_shader_set_float(layer, edition_draw_shader, 'u_field_scale', field_scale or 0.15)
snkrx/edition.lua:191:  layer_shader_set_float(layer, edition_draw_shader, 'u_dither_mode', dither_mode or 0)

Your questions have been answered: "Image-as-SHAPE needs a mask texture bound to the rect draw, but the engine doesn't currently support that for the draw-shader path (only for post-process). How should I wire it?"="Patch the engine", "For 'effect over a real image,' what should drive the field f?"="Need to test options visually to see what's better.", "'Make everything work on 480×270' — the gallery + sidebar are hand-placed for 1920×1080. What's the scope?"="We'll do this later manually, piece-by-piece, don't worry about it now.". You can now continue with these answers in mind.

=== image.lua ===
--[[
  image — thin wrapper for GPU texture handles.

  Usage:
    images.player = image_load('player', 'assets/player.png')
    layer_image(game_layer, images.player, 100, 100)

  An image is a plain table with .handle, .width, .height.
]]

image = class()

function image:new(handle)
  self.handle = handle
  self.width = texture_get_width(handle)
  self.height = texture_get_height(handle)
end

-- Load a texture from a file, wrap it, and add to the global `images` table.
function image_load(name, path)
  local handle = texture_load(path)
  local img = image(handle)
  if images then images[name] = img end
  return img
end
=== l_layer_shader_set_texture + l_image_load in engine ===
    int value = (int)luaL_checkinteger(L, 4);
    layer_shader_set_int(layer, shader, name, value);
    return 0;
... [18 more lines]

[
  {
    "type": "text",
    "text": "Chapter marked: \"Image-as-SHAPE\". Continue your current work."
  }
]

4668         // Handle uniform setting commands
4669            // These need to temporarily switch programs, so flush batch first and restore after
4670            if (cmd->type == COMMAND_SET_UNIFORM_FLOAT ||
4671                cmd->type == COMMAND_SET_UNIFORM_VEC2 ||
4672                cmd->type == COMMAND_SET_UNIFORM_VEC4 ||
4673                cmd->type == COMMAND_SET_UNIFORM_INT ||
4674                cmd->type == COMMAND_SET_UNIFORM_TEXTURE) {
4675                // Flush any pending draws before switching programs
4676                batch_flush();
4677                current_batch_texture = 0;
4678    
4679                glUseProgram(cmd->shader_id);
4680                switch (cmd->type) {
4681                    case COMMAND_SET_UNIFORM_FLOAT:
4682                        glUniform1f((GLint)cmd->uniform_location, cmd->params[0]);
4683                        break;
4684                    case COMMAND_SET_UNIFORM_VEC2:
4685                        glUniform2f((GLint)cmd->uniform_location, cmd->params[0], cmd->params[1]);
4686                        break;
4687                    case COMMAND_SET_UNIFORM_VEC4:
4688                        glUniform4f((GLint)cmd->uniform_location, cmd->params[0], cmd->params[1], cmd->params[2], cmd->params[3]);
4689                        break;
4690                    case COMMAND_SET_UNIFORM_INT:
4691                        glUniform1i((GLint)cmd->uniform_location, (int)cmd->params[0]);
4692                        break;
4693                    case COMMAND_SET_UNIFORM_TEXTURE: {
4694                        // Store for binding during execute_apply_shader
4695                        layer->extra_texture = cmd->texture_id;
4696                        layer->extra_texture_loc = (GLint)cmd->uniform_location;
4697                        layer->has_extra_texture = true;
... [8 more lines]

The file E:\a327ex\Anchor2\engine\src\anchor.c has been updated successfully. (file state is current in your context — no need to Read it back)

Microsoft Windows [Version 10.0.26200.8457]
(c) Microsoft Corporation. All rights reserved.

E:\a327ex\Anchor2\engine>

=== build.bat ===
@echo off
setlocal

cd /d "%~dp0"

:: Find and run vcvarsall.bat for Visual Studio
if exist "C:\Program Files\Microsoft Visual Studio\18\Community\VC\Auxiliary\Build\vcvarsall.bat" (
    call "C:\Program Files\Microsoft Visual Studio\18\Community\VC\Auxiliary\Build\vcvarsall.bat" x64 >nul 2>&1
) else if exist "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvarsall.bat" (
    call "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvarsall.bat" x64 >nul 2>&1
) else if exist "C:\Program Files\Microsoft Visual Studio\2022\Professional\VC\Auxiliary\Build\vcvarsall.bat" (
    call "C:\Program Files\Microsoft Visual Studio\2022\Professional\VC\Auxiliary\Build\vcvarsall.bat" x64 >nul 2>&1
) else if exist "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\Build\vcvarsall.bat" (
    call "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\Build\vcvarsall.bat" x64 >nul 2>&1
) else (
    echo ERROR: Could not find Visual Studio
    exit /b 1
)

if not exist build mkdir build

:: Build Lua library if it doesn't exist
if not exist lib\lua.lib (
    echo Building Lua library...
    if not exist build\lua_obj mkdir build\lua_obj
    cl.exe /nologo /O2 /W3 /c /I"include/lua" ^
        include/lua/lapi.c include/lua/lauxlib.c include/lua/lbaselib.c ^
        include/lua/lcode.c include/lua/lcorolib.c include/lua/lctype.c ^
        include/lua/ldblib.c include/lua/ldebug.c include/lua/ldo.c ^
... [79 more lines]

Exit code 1
'build.bat' is not recognized as an internal or external command,
operable program or batch file.
---- EXIT: 1 ----

LastWriteTime      Length
-------------      ------
24/05/26 11:41:06 4185088

anchor.c
include/stb\stb_image_write.h(776): warning C4996: 'sprintf': This function or variable may be unsafe. Consider using sprintf_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(290): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(1095): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(1312): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(2759): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(2753): warning C4267: 'initializing': conversion from 'size_t' to 'int', possible loss of data
src/anchor.c(3150): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(3428): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(3434): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(3544): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(3553): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(3647): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(3649): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(4907): warning C4996: 'strdup': The POSIX name for this item is deprecated. Instead, use the ISO C and C++ conformant name: _strdup. See online help for details.
src/anchor.c(6026): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(6911): warning C4244: 'initializing': conversion from 'lua_Number' to 'float', possible loss of data
src/anchor.c(6912): warning C4244: 'initializing': conversion from 'lua_Number' to 'float', possible loss of data
src/anchor.c(8685): warning C4244: '=': conversion from 'lua_Integer' to 'int', possible loss of data
src/anchor.c(8686): warning C4244: '=': conversion from 'lua_Integer' to 'int', possible loss of data
src/anchor.c(8692): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(8806): warning C4996: 'strncpy': This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(10144): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(10148): warning C4996: 'strcat': This function or variable may be unsafe. Consider using strcat_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(10153): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(10157): warning C4996: 'strcat': This function or variable may be unsafe. Consider using strcat_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(10163): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(10167): warning C4996: 'strcat': This function or variable may be unsafe. Consider using strcat_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(10173): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
src/anchor.c(10177): warning C4996: 'strcat': This function or variable may be unsafe. Consider using strcat_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
... [13 more lines]

LastWriteTime      Length
-------------      ------
30/05/26 14:04:55 4185088

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: main.lua
OK: effect.lua

=== shape_mask refs (shader) ===
557:float shape_mask(int kind, float dist, float fc, vec2 cell_uv) {
588:    coverage = shape_mask(u_shape_kind, length(q) / 0.5, fc, cell_uv);
=== u_shape_tex refs ===
96:uniform sampler2D u_shape_tex;     // tiled mask for the 'sprite' shape kind; bound to
563:        return textureLod(u_shape_tex, vec2(cell_uv.x, 1.0 - cell_uv.y), 0.0).a;

85:ui_row_h          = 32     -- height of one label row (PATTERN, COLOR, etc.)
90:ui_swatch_size    = 22     -- side of one palette-token swatch
92:ui_section_gap    = 12     -- vertical gap between groups of rows
93:ui_panel_inset    = 16     -- inner padding inside the panel
94:ui_title_h        = 28     -- title bar height

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: main.lua

1    // Fetch the Tabler icon set used by the showcase, via the Iconify API, as white
2   // SVGs, thin the stroke (Tabler is stroke-based) toward the theme's 1px hairline
3   // UI, rasterize with resvg, and write to ../assets/icons/tabler/<concept>.png.
4   //
5   // Run from this directory:  node fetch_icons.mjs
6   //
7   // Tabler was chosen over Phosphor/Remix (most icons, best game coverage of the
8   // clean line packs, and a freely adjustable stroke). Icons are rasterized WHITE
9   // on transparent so the engine can tint them to any palette token.
10  
11  import { Resvg } from '@resvg/resvg-js';
12  import { mkdir, writeFile } from 'node:fs/promises';
13  import { dirname } from 'node:path';
14  
15  const SIZE = 44;
16  const OUT = '../assets/icons/tabler';
17  const STROKE = '0.75';   // Tabler ships stroke-width="2"; thin it to sit with the hairline UI
18  
19  // concept -> ordered candidate Tabler names (first that resolves wins).
20  const concepts = {
21    heart:   ['heart'],
22    star:    ['star'],
23    shield:  ['shield'],
24    sword:   ['sword'],
25    flame:   ['flame'],
26    bolt:    ['bolt'],
27    trophy:  ['trophy'],
28    coin:    ['coin'],
29    skull:   ['skull'],
30    gear:    ['settings'],
... [44 more lines]

The file E:\a327ex\ricochet-template\tools\fetch_icons.mjs has been updated successfully. (file state is current in your context — no need to Read it back)

OK    heart <- tabler:heart
  +    heart_filled <- tabler:heart-filled
OK    star <- tabler:star
  +    star_filled <- tabler:star-filled
OK    shield <- tabler:shield
  +    shield_filled <- tabler:shield-filled
OK    sword <- tabler:sword
OK    flame <- tabler:flame
  +    flame_filled <- tabler:flame-filled
OK    bolt <- tabler:bolt
  +    bolt_filled <- tabler:bolt-filled
OK    trophy <- tabler:trophy
  +    trophy_filled <- tabler:trophy-filled
OK    coin <- tabler:coin
  +    coin_filled <- tabler:coin-filled
OK    skull <- tabler:skull
OK    gear <- tabler:settings
  +    gear_filled <- tabler:settings-filled
OK    target <- tabler:target
OK    bomb <- tabler:bomb
  +    bomb_filled <- tabler:bomb-filled
OK    key <- tabler:key
  +    key_filled <- tabler:key-filled
OK    flask <- tabler:flask
  +    flask_filled <- tabler:flask-filled
OK    user <- tabler:user
  +    user_filled <- tabler:user-filled
OK    map <- tabler:map
OK    bag <- tabler:backpack
OK    bell <- tabler:bell
... [23 more lines]

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: main.lua
OK: effect.lua
=== verify every icon_names PNG exists ===
46/46 present, 0 missing

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: main.lua

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: main.lua

Your questions have been answered: "What test images should I source into assets/images/ for the image-as-CONTENT screen?"="I'll drop in my own". You can now continue with these answers in mind.

=== any dirent/FindFirstFile/readdir in engine ===
=== lua registers mentioning file/dir ===
9453:lua_register(L, "sound_get_path"
9455:lua_register(L, "file_write_string"
9474:lua_register(L, "shader_load_file"

assets/images/ does not exist yet

created: assets/images/

-rw-r--r-- 1 adn 197121 3383629 May 30 15:01 boy.png
-rw-r--r-- 1 adn 197121 1845595 May 30 15:00 girl.png
=== PNG header info (width/height/bit-depth/colortype) ===
girl.png: 2600x1500  bitdepth=8  colortype=2 (RGB)
boy.png: 2048x1990  bitdepth=8  colortype=2 (RGB)

[
  {
    "type": "image",
    "source": {
      "type": "base64",
      "data": "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",
      "media_type": "image/jpeg"
    }
  }
]

[
  {
    "type": "image",
    "source": {
      "type": "base64",
      "data": "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",
      "media_type": "image/jpeg"
    }
  }
]

Your questions have been answered: "Your test images are opaque, so plain PATTERN mode just fills the whole box. How should PATTERN mode treat them?"="Ah, I see, you can find a generic image online to be used as stencil to test the pattern mode.". You can now continue with these answers in mind.

OK   wolf <- game-icons:wolf-head
OK   eagle <- game-icons:eagle-emblem
OK   spartan <- game-icons:spartan-helmet
OK   dragon <- game-icons:dragon-head
OK   raven <- game-icons:raven
OK   flower <- game-icons:rose
OK   butterfly <- ph:butterfly-fill
OK   lion <- game-icons:lion
total 180
drwxr-xr-x 1 adn 197121     0 May 30 15:06 ./
drwxr-xr-x 1 adn 197121     0 May 30 15:06 ../
-rw-r--r-- 1 adn 197121 13786 May 30 15:06 butterfly.png
-rw-r--r-- 1 adn 197121 21150 May 30 15:06 dragon.png
-rw-r--r-- 1 adn 197121 25880 May 30 15:06 eagle.png
-rw-r--r-- 1 adn 197121 23081 May 30 15:06 flower.png
-rw-r--r-- 1 adn 197121 18496 May 30 15:06 lion.png
-rw-r--r-- 1 adn 197121 12558 May 30 15:06 raven.png
-rw-r--r-- 1 adn 197121 24347 May 30 15:06 spartan.png
-rw-r--r-- 1 adn 197121 19609 May 30 15:06 wolf.png

previews written

[
  {
    "type": "image",
    "source": {
      "type": "base64",
      "data": "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",
      "media_type": "image/png"
    }
  }
]

[
  {
    "type": "image",
    "source": {
      "type": "base64",
      "data": "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",
      "media_type": "image/png"
    }
  }
]

[
  {
    "type": "image",
    "source": {
      "type": "base64",
      "data": "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",
      "media_type": "image/png"
    }
  }
]

--- assets/images/ now: ---
total 5137
drwxr-xr-x 1 adn 197121       0 May 30 15:08 ./
drwxr-xr-x 1 adn 197121       0 May 30 14:49 ../
-rw-r--r-- 1 adn 197121     258 May 30 15:08 CREDITS.txt
-rw-r--r-- 1 adn 197121 3383629 May 30 15:01 boy.png
-rw-r--r-- 1 adn 197121 1845595 May 30 15:00 girl.png
-rw-r--r-- 1 adn 197121   19609 May 30 15:08 wolf.png

=== layer image/texture fns in framework ===
169:function layer_image(lyr, img, x, y, color, flash)
174:function layer_texture(lyr, tex, x, y, color)
178:function layer_spritesheet(lyr, sheet, frame, x, y, color, flash)
=== signatures (with bodies) ===
169:function layer_image(lyr, img, x, y, color, flash)
170-  eng.draw_texture(lyr_handle(lyr), img.handle, x, y, color or 0xFFFFFFFF, flash or 0)
171-end
172-
--
174:function layer_texture(lyr, tex, x, y, color)
175-  eng.draw_texture(lyr_handle(lyr), tex, x, y, color or 0xFFFFFFFF, 0)
176-end
177-
=== engine l_layer_image / texture draw args ===

=== draw_texture-related Lua bindings ===
9418:lua_register(L, "layer_draw_texture"
9424:lua_register(L, "texture_load"
9425:lua_register(L, "texture_create"
9426:lua_register(L, "texture_unload"
9427:lua_register(L, "texture_get_width"
9428:lua_register(L, "texture_get_height"
9430:lua_register(L, "spritesheet_load"
9431:lua_register(L, "spritesheet_get_frame_width"
9432:lua_register(L, "spritesheet_get_frame_height"
9433:lua_register(L, "spritesheet_get_total_frames"
9434:lua_register(L, "layer_draw_spritesheet_frame"
9485:lua_register(L, "layer_shader_set_texture"
9488:lua_register(L, "layer_get_texture"

=== l_draw_texture signature/impl ===
5637:static int l_layer_draw_texture(lua_State* L) {
9418:    lua_register(L, "layer_draw_texture", l_layer_draw_texture);
=== framework draw_texture wrapper ===
E:/a327ex/ricochet-template/anchor/layer.lua:58:  draw_texture = layer_draw_texture,
E:/a327ex/ricochet-template/anchor/layer.lua:170:  eng.draw_texture(lyr_handle(lyr), img.handle, x, y, color or 0xFFFFFFFF, flash or 0)
E:/a327ex/ricochet-template/anchor/layer.lua:175:  eng.draw_texture(lyr_handle(lyr), tex, x, y, color or 0xFFFFFFFF, 0)

5637 static int l_layer_draw_texture(lua_State* L) {
5638        Layer* layer = (Layer*)lua_touserdata(L, 1);
5639        Texture* tex = (Texture*)lua_touserdata(L, 2);
5640        float x = (float)luaL_checknumber(L, 3);
5641        float y = (float)luaL_checknumber(L, 4);
5642        uint32_t color = (uint32_t)luaL_optinteger(L, 5, 0xFFFFFFFF);  // Default white (no tint)
5643        uint32_t flash = (uint32_t)luaL_optinteger(L, 6, 0x00000000);  // Default black (no flash)
5644        layer_add_image(layer, tex, x, y, color, flash);
5645        return 0;
5646    }
5647    
5648    static int l_layer_set_blend_mode(lua_State* L) {
5649        Layer* layer = (Layer*)lua_touserdata(L, 1);
5650        const char* mode = luaL_checkstring(L, 2);
5651        if (strcmp(mode, "alpha") == 0) {
5652            layer_set_blend_mode(layer, BLEND_ALPHA);
5653        } else if (strcmp(mode, "additive") == 0) {
5654            layer_set_blend_mode(layer, BLEND_ADDITIVE);
5655        } else {
5656            return luaL_error(L, "Invalid blend mode: %s (use 'alpha' or 'additive')", mode);
5657        }
5658        return 0;
5659    }
5660    
5661    // Stencil Lua bindings
5662    static int l_layer_stencil_mask(lua_State* L) {
5663        Layer* layer = (Layer*)lua_touserdata(L, 1);
5664        layer_stencil_mask(layer);
5665        return 0;
5666    }
... [30 more lines]

1    --[[
2     camera — procedural 2D camera with follow, bounds, and integrated shake.
3   
4     Usage:
5       main_camera = camera_new(width, height)
6   
7       -- In update:
8       camera_update(main_camera, dt)
9   
10      -- In draw:
11      camera_attach(main_camera, game_layer)
12      layer_circle(game_layer, 100, 100, 20, red())
13      -- ... more draws to game_layer
14      camera_detach(main_camera, game_layer)
15  
16      -- Follow a target:
17      camera_follow(main_camera, p1)
18      camera_follow(main_camera, p1, 0.9, 0.3)   -- 90% distance in 0.3s
19      camera_follow(main_camera, nil)             -- stop following
20  
21      -- Bounds:
22      camera_set_bounds(main_camera, 0, map_w, 0, map_h)
23  
24    Shake is an integrated sub-structure accessible as camera.shake.
25    Use the shake_* functions on it:
26      shake_push(main_camera.shake, angle, amount)
27      shake_shake(main_camera.shake, 10, 0.3)
28      shake_trauma(main_camera.shake, 0.5, 0.3)
29  
30    Design notes:
... [185 more lines]

198:function layer_push(lyr, x, y, r, sx, sy)
199-  eng.push(lyr_handle(lyr), x, y, r, sx, sy)
200-end
--
202:function layer_pop(lyr)
203-  eng.pop(lyr_handle(lyr))
204-end

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

238  local GAL_RECT_H      = 230
239 
240 -- Which gallery rect the modifier sliders edit. Click a rect to select it.
241 gallery_selected = 1
242 
243 -- Step through an icon-name list for the sprite icon picker. dir = +1 / -1;
244 -- list defaults to the full icon_names (the field mode passes icon_names_filled).
245 local function cycle_icon(cur, dir, list)
246   list = list or icon_names
247   for i, v in ipairs(list) do
248     if v == cur then return list[((i - 1 + dir) % #list) + 1] end
249   end
250   return list[1]
251 end
252 
253 -- True for the icon-mask shape kinds (bg + field), which share the ICON picker,
254 -- the SPACING slider, and the per-frame mask-texture bind.
255 local function shape_is_sprite(s) return s == 'sprite' or s == 'sprite_field' end
256 
257 -- Step through content_names for the IMAGE-screen picker. dir = +1 / -1.
258 local function cycle_content_image(cur, dir)
259   for i, v in ipairs(content_names) do
260     if v == cur then return content_names[((i - 1 + dir) % #content_names) + 1] end
261   end
262   return content_names[1]
263 end
264 
265 -- Screen rect for gallery entry i (row-major in the GAL grid). Shared by the
266 -- selection hit-test (ui_update), the rect draw, and the highlight (draw).
267 local function gallery_rect(i)
... [40 more lines]

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

325    -- COLOR / DITHER cycler rows. (PATTERN + SCALE rows removed — in gallery
326   -- mode each rectangle bakes its own pattern + tuned scale; the sidebar now
327   -- only drives the SHARED look applied across all gallery rects.)
328   row = rect_cut_top(body, ui_row_h)
329   local f = ui_field{ rect = row, id = 'color', label = 'COLOR', value = current_color }
330   if f.prev_clicked then current_color = effect_prev_color(current_color) end
331   if f.next_clicked then current_color = effect_next_color(current_color) end
332 
333   row = rect_cut_top(body, ui_row_h)
334   f = ui_field{ rect = row, id = 'dither', label = 'DITHER',
335                 value = effect_dither_label(current_dither) }
336   if f.prev_clicked then current_dither = effect_prev_dither(current_dither) end
337   if f.next_clicked then current_dither = effect_next_dither(current_dither) end
338 
339   -- SHAPE cycler — re-express the field as a grid of dots. Shared across all
340   -- gallery rects. The SHAPE SIZE slider only appears when a shape is active.
341   row = rect_cut_top(body, ui_row_h)
342   f = ui_field{ rect = row, id = 'shape', label = 'SHAPE',
343                 value = effect_shape_label(current_shape) }
344   if f.prev_clicked then current_shape = effect_prev_shape(current_shape) end
345   if f.next_clicked then current_shape = effect_next_shape(current_shape) end
346 
347   -- Sprite-only controls (both icon-mask kinds): which ICON the mask uses, and
348   -- the SPACING gap between tiled icons. The chosen icon's texture is bound as
349   -- the per-cell mask in draw().
350   if shape_is_sprite(current_shape) then
351     -- field mode works only with filled icons (an outline scaled down by the
352     -- field becomes a thin ring) → restrict the picker to icon_names_filled and
353     -- snap the current selection to a filled icon, preferring its own filled
354     -- counterpart (heart → heart_filled) when one exists.
... [100 more lines]

454    row = rect_cut_top(body, ui_row_h + 20)
455   local ms = ui_slider{ rect = row, id = 'mod_scale', label = 'SCALE',
456                         value = sel.scale, vmin = 0.05, vmax = 1.2, fmt = '%.3f' }
457   if ms.changed then sel.scale = ms.value end
458 
459   row = rect_cut_top(body, ui_row_h + 20)
460   local msp = ui_slider{ rect = row, id = 'mod_speed', label = 'SPEED',
461                          value = sel.speed, vmin = 0, vmax = 4, fmt = '%.2f' }
462   if msp.changed then sel.speed = msp.value end
463 
464   row = rect_cut_top(body, ui_row_h + 20)
465   local mco = ui_slider{ rect = row, id = 'mod_contrast', label = 'CONTRAST',
466                          value = sel.contrast, vmin = 0, vmax = 2.5, fmt = '%.2f' }
467   if mco.changed then sel.contrast = mco.value end
468 
469   row = rect_cut_top(body, ui_row_h + 20)
470   local mpa = ui_slider{ rect = row, id = 'mod_param', label = sel.param_label,
471                          value = sel.param, vmin = 0, vmax = 1, fmt = '%.2f' }
472   if mpa.changed then sel.param = mpa.value end
473 
474   -- Second intrinsic knob — only patterns that define one (the gradients:
475   -- OFFSET / RANGE) get this row; it uses the pattern's own min/max range.
476   if sel.param2_label then
477     row = rect_cut_top(body, ui_row_h + 20)
478     local mp2 = ui_slider{ rect = row, id = 'mod_param2', label = sel.param2_label,
479                            value = sel.param2, vmin = sel.param2_min or 0,
480                            vmax = sel.param2_max or 1, fmt = '%.2f' }
481     if mp2.changed then sel.param2 = mp2.value end
482   end
483 end
... [10 more lines]

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: main.lua
OK: effect.lua
OK: showcase.lua

646  
647     if (vType < 0.5) {
648         // Rectangle
649         vec2 quad_size = vShape0.xy;
650         quad_px = quad_size;
651         vec2 local_p = vUV * quad_size;
652         vec2 center = quad_size * 0.5;
653         vec2 half_size = vShape0.zw;
654         stroke = vShape1.x;
655         if (u_aa_width == 0.0) { local_p = floor(local_p) + 0.5; }
656         d = sdf_rect(local_p, center, half_size);
657     } else if (vType < 1.5) {
658         // Circle
659         float quad_size = vShape0.x;
660         quad_px = vec2(quad_size);
661         vec2 local_p = vUV * quad_size;
662         vec2 center = vec2(quad_size * 0.5);
663         float radius = vShape0.z;
664         stroke = vShape0.w;
665         if (u_aa_width == 0.0) { radius = floor(radius + 0.5); }
666         d = sdf_circle(local_p, center, radius);
667     } else if (vType < 2.5) {
668         // Sprite
669         ivec2 texSize = textureSize(u_texture, 0);
670         vec2 snappedUV = (floor(vUV * vec2(texSize)) + 0.5) / vec2(texSize);
671         vec4 texColor = texture(u_texture, snappedUV);
672         float sprite_alpha = texColor.a * vColor.a;
673         if (sprite_alpha <= 0.0) discard;
674         vec3 col = texColor.rgb * vColor.rgb + vAddColor;
675         if (u_color_kind > 0) {
... [22 more lines]

697                      } else {
698                         col = ccol;
699                         sprite_alpha *= cd;                              // transparent backing
700                     }
701                 }
702             }
703         }
704         FragColor = vec4(col, sprite_alpha);
705         return;
706     } else if (vType < 3.5) {
707         // Line / capsule
708         vec2 quad_size = vShape0.xy;
709         quad_px = quad_size;
710         vec2 local_p = vUV * quad_size;
711         vec2 center = quad_size * 0.5;
712         float half_len = vShape0.z;

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: main.lua

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: main.lua
OK: effect.lua

678          vec2 local_p = vUV * quad_size;
679         vec2 center = vec2(quad_size * 0.5);
680         float radius = vShape0.z;
681         stroke = vShape0.w;
682         if (u_aa_width == 0.0) { radius = floor(radius + 0.5); }
683         d = sdf_circle(local_p, center, radius);
684     } else if (vType < 2.5) {
685         // Sprite
686         ivec2 texSize = textureSize(u_texture, 0);
687         vec2 snappedUV = (floor(vUV * vec2(texSize)) + 0.5) / vec2(texSize);
688         vec4 texColor = texture(u_texture, snappedUV);
689         float sprite_alpha = texColor.a * vColor.a;
690         if (sprite_alpha <= 0.0) discard;
691         vec3 col = texColor.rgb * vColor.rgb + vAddColor;
692         if (u_color_kind > 0) {
693             // The field comes from effect_field (the image's own luminance when
694             // u_image_field is set — optionally rippled by the moving pattern —
695             // else the pattern) and flows through the SAME shape / dither / color /
696             // backing pipeline as the gallery. So SHAPE works on images too
697             // (luminance + shape = halftone), and the image's alpha stays the base
698             // coverage that shape coverage multiplies into.
699             float coverage;
700             float f  = effect_field_shaped(vec2(texSize), coverage);
701             float fd = apply_dither(f, u_dither_kind, gl_FragCoord.xy);
702             vec3 ccol = apply_color(u_color_kind, col, fd, u_palette[u_color_a].rgb, u_palette[u_color_b].rgb);
703             if (u_shape_kind == 0) {
704                 col = ccol;
705             } else {
706                 float cd = apply_dither(coverage, u_dither_kind, gl_FragCoord.xy);
707                 if (u_shape_backing == 1) {
... [4 more lines]

711                      sprite_alpha *= cd;                              // transparent backing
712                 }
713             }
714         }
715         FragColor = vec4(col, sprite_alpha);
716         return;
717     } else if (vType < 3.5) {
718         // Line / capsule
719         vec2 quad_size = vShape0.xy;
720         quad_px = quad_size;

556  //     the color. For an image shape, `m` would be the sprite's alpha at the
557 //     cell-local UV; nothing downstream changes.
558 //   * COLOR — *what hue*. The field value handed to dither + the color recipe,
559 //     at FULL strength (never multiplied by the mask). circle (size) uses the
560 //     per-FRAGMENT value `ff` (color varies inside the dot); circle (hard) uses
561 //     the per-CELL value `fc` (one flat color per dot).
562 //
563 // Compositing happens in main() per u_shape_backing:
564 //   backing 0 (transparent) → final alpha *= coverage  (gaps show the page)
565 //   backing 1 (token)       → mix(palette[color_a], colored, coverage)
566 //
567 // Cells are u_shape_size px square, anchored to screen space (gl_FragCoord),
568 // the same space as the dither. dist = normalized distance from the cell
569 // center (0 → 1 at the inscribed-circle edge, ~1.41 in the corners).
570 // =============================================================================
571 
572 // Coverage mask of the shape at the fragment, in [0,1]. `dist` is the
573 // normalized distance from the cell center; `cell_uv` is the 0..1 position
574 // within the cell; `fc` is the per-cell field (sizes the circle variants).
575 // THIS is the single extension point for new shapes — add a branch returning
576 // the new shape's coverage; the rest of the pipeline is unchanged.
577 float shape_mask(int kind, float dist, float fc, vec2 cell_uv) {
578     if (kind == 3 || kind == 4) {
579         // sprite: the chosen icon's silhouette (its alpha), tiled one-per-cell.
580         // Flip Y (gl_FragCoord is bottom-up, texture origin top-left); textureLod
581         // (.,0) avoids cell-edge mip seams.
582         //   fit  = icon footprint as a fraction of the cell — SPACING insets it
583         //   kind 3 (bg)    : full footprint; the effect fills/colors a constant icon
584         //   kind 4 (field) : footprint also scaled by the per-cell field fc, so the
585         //                    icon grows with the effect and vanishes where it's dark
... [22 more lines]

606  float effect_field_shaped(vec2 quad_px, out float coverage) {
607     float ff = effect_field(vPos, vUV, quad_px);                // per-fragment field
608     coverage = 1.0;
609     if (u_shape_kind == 0) return ff;
610     // Grid pitch (cell-to-cell distance). For sprite kinds, SPACING enlarges the
611     // pitch (more gap between icons) while shape_mask insets the icon by the same
612     // factor — the two cancel, so the icon's absolute size stays u_shape_size and
613     // only the gap grows. Circles ignore spacing (their gaps come from fc).
614     float size = max(u_shape_size, 1.0);
615     if (u_shape_kind >= 3) size /= max(1.0 - u_shape_spacing, 0.05);
616     vec2 cc  = (floor(gl_FragCoord.xy / size) + 0.5) * size;    // cell center, screen px
617     vec2 dpx = cc - gl_FragCoord.xy;                            // fragment → cell center
618     vec2 wc  = vPos + dFdx(vPos) * dpx.x + dFdy(vPos) * dpx.y;  // world pos at cell center
619     vec2 uvc = vUV  + dFdx(vUV)  * dpx.x + dFdy(vUV)  * dpx.y;  // uv at cell center
620     float fc = effect_field(wc, uvc, quad_px);                 // per-cell field
621     vec2 cell_uv = fract(gl_FragCoord.xy / size);              // 0..1 within the cell
622     vec2 q   = cell_uv - 0.5;
623     coverage = shape_mask(u_shape_kind, length(q) / 0.5, fc, cell_uv);
624     return (u_shape_kind == 2) ? fc : ff;   // hard: flat per-cell color; size/sprite: per-fragment color
625 }

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: effect.lua
=== shape_mask call sites in shader (should be 1, new sig) ===
583:float shape_mask(int kind, float fc, vec2 cell_uv) {
641:    // pitch (more gap between icons) while shape_mask insets the icon by the same
652:    coverage = shape_mask(u_shape_kind, fc, cell_uv);

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: effect.lua

578  // The geometric shapes share one idea: a "gauge" metric m(cell_uv) that is 0 at
579 // the cell center and reaches 1 at a cell mid-side, with the shape boundary at
580 // m == fc. So every shape grows with the per-cell field exactly like the circle
581 // — only the distance metric differs (Euclidean → circle, Chebyshev → square,
582 // Manhattan → diamond, hex gauge → hexagon). Ring + cross are handled specially.
583 float shape_mask(int kind, float fc, vec2 cell_uv) {
584     if (kind == 3 || kind == 4) {
585         // sprite: the chosen icon's silhouette (its alpha), tiled one-per-cell.
586         // Flip Y (gl_FragCoord is bottom-up, texture origin top-left); textureLod
587         // (.,0) avoids cell-edge mip seams.
588         //   fit  = icon footprint as a fraction of the cell — SPACING insets it
589         //   kind 3 (bg)    : full footprint; the effect fills/colors a constant icon
590         //   kind 4 (field) : footprint also scaled by the per-cell field fc, so the
591         //                    icon grows with the effect and vanishes where it's dark
592         //                    (the circle-analog — the effect IS the icon grid)
593         vec2 uv = vec2(cell_uv.x, 1.0 - cell_uv.y);
594         float fit = 1.0 - u_shape_spacing;
595         float s   = (kind == 4) ? fc * fit : fit;
596         s = max(s, 1e-4);
597         uv = (uv - 0.5) / s + 0.5;
598         if (uv.x < 0.0 || uv.x > 1.0 || uv.y < 0.0 || uv.y > 1.0) return 0.0;
599         return textureLod(u_shape_tex, uv, 0.0).a;
600     }
601 
602     float e = 0.04;                       // anti-alias edge width
603     vec2 p = (cell_uv - 0.5) * 2.0;       // [-1,1]^2; |component| = 1 at a cell mid-side
604     vec2 a = abs(p);
605 
606     // RING — annulus of the circle (hole = 50% of the outer radius).
607     if (kind == 8) {
... [22 more lines]

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\assets\draw_shader.frag has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: main.lua
OK: effect.lua

108:function layer_rectangle(lyr, x, y, w, h, color)
109-  eng.rectangle(lyr_handle(lyr), x, y, w, h, color)
110-end
--
112:function layer_circle(lyr, x, y, radius, color)
113-  eng.circle(lyr_handle(lyr), x, y, radius, color)
114-end
=== how engine draw_rectangle reads color ===
4954:static int l_layer_rectangle(lua_State* L) {
4966:static int l_layer_rectangle_line(lua_State* L) {
4979:static int l_layer_rectangle_gradient_h(lua_State* L) {
4992:static int l_layer_rectangle_gradient_v(lua_State* L) {
9401:    lua_register(L, "layer_rectangle", l_layer_rectangle);
9402:    lua_register(L, "layer_rectangle_line", l_layer_rectangle_line);
9403:    lua_register(L, "layer_rectangle_gradient_h", l_layer_rectangle_gradient_h);
9404:    lua_register(L, "layer_rectangle_gradient_v", l_layer_rectangle_gradient_v);

4954 static int l_layer_rectangle(lua_State* L) {
4955        Layer* layer = (Layer*)lua_touserdata(L, 1);
4956        float x = (float)luaL_checknumber(L, 2);
4957        float y = (float)luaL_checknumber(L, 3);
4958        float w = (float)luaL_checknumber(L, 4);
4959        float h = (float)luaL_checknumber(L, 5);
4960        uint32_t color = (uint32_t)luaL_checkinteger(L, 6);
4961        layer_add_rectangle(layer, x, y, w, h, 0.0f, color); // stroke=0 for filled
4962        return 0;
4963    }
4964    
4965    // layer_rectangle_line(layer, x, y, w, h, color, line_width?) -- rectangle outline
4966    static int l_layer_rectangle_line(lua_State* L) {
4967        Layer* layer = (Layer*)lua_touserdata(L, 1);

378    Performance: enqueues ~14 uniform writes per call (plus the ramp token list
379   when color='ramp'). Cheap (a few bytes each).
380 ]]
381 function effect_set(layer, spec)
382   spec = spec or {}
383   local p  = pattern_kinds[spec.pattern or 'solid'] or 2
384   local c  = color_kinds[spec.color or 'none'] or 0
385   local d  = dither_kinds[spec.dither or 'off'] or 0
386   local sh = shape_kinds[spec.shape or 'none'] or 0
387   local ca = resolve_color(spec.color_a or 'ink')
388   local cb = resolve_color(spec.color_b or 'field')
389   local scale    = spec.pattern_scale or effect_default_pattern_scale
390   local param    = spec.pattern_param or 0
391   local param2   = spec.pattern_param2 or 0
392   -- Universal modifiers. Default 1.0 = neutral (speed 1 = normal animation,
393   -- contrast 1 = field unchanged). MUST default to 1, not 0 — a 0 here would
394   -- freeze animation / flatten the field to mid-grey.
395   local speed    = spec.speed or 1.0
396   local contrast = spec.contrast or 1.0
397   local shape_sz = spec.shape_size or 24
398   local shape_sp = spec.shape_spacing or 0
399 
400   layer_shader_set_int  (layer, effect_draw_shader, 'u_pattern_kind',     p)
401   layer_shader_set_int  (layer, effect_draw_shader, 'u_color_kind',       c)
402   layer_shader_set_int  (layer, effect_draw_shader, 'u_dither_kind',      d)
403   layer_shader_set_int  (layer, effect_draw_shader, 'u_shape_kind',       sh)
404   layer_shader_set_float(layer, effect_draw_shader, 'u_shape_size',       shape_sz)
405   layer_shader_set_float(layer, effect_draw_shader, 'u_shape_spacing',    shape_sp)
406   layer_shader_set_float(layer, effect_draw_shader, 'u_shape_rotation',   spec.shape_rotation or 0)
407   -- shape gaps: 'transparent' (0, gaps show through) or 'token' (1, gaps = color_a)

407    -- shape gaps: 'transparent' (0, gaps show through) or 'token' (1, gaps = color_a)
408   layer_shader_set_int  (layer, effect_draw_shader, 'u_shape_backing',
409                          (spec.shape_backing == 'token') and 1 or 0)
410   layer_shader_set_int  (layer, effect_draw_shader, 'u_color_a',          ca)
411   layer_shader_set_int  (layer, effect_draw_shader, 'u_color_b',          cb)
412   -- color='ramp': push the ordered token list (indices + count). Only when the
413   -- recipe is active — other recipes leave the (unused) ramp uniforms as-is.
414   if c == 3 then
415     local ramp = spec.ramp or effect_ramp
416     local rn = math.min(#ramp, 8)
417     layer_shader_set_int(layer, effect_draw_shader, 'u_ramp_count', rn)
418     for i = 1, rn do
419       layer_shader_set_int(layer, effect_draw_shader, 'u_ramp_tokens[' .. (i - 1) .. ']', resolve_color(ramp[i]))
420     end

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

421    -- recipe is active — other recipes leave the (unused) ramp uniforms as-is.
422   if c == 3 then
423     local ramp = spec.ramp or effect_ramp
424     local rn = math.min(#ramp, 8)
425     layer_shader_set_int(layer, effect_draw_shader, 'u_ramp_count', rn)
426     for i = 1, rn do
427       layer_shader_set_int(layer, effect_draw_shader, 'u_ramp_tokens[' .. (i - 1) .. ']', resolve_color(ramp[i]))
428     end
429   end
430   layer_shader_set_float(layer, effect_draw_shader, 'u_pattern_scale',    scale)
431   layer_shader_set_float(layer, effect_draw_shader, 'u_pattern_param',    param)
432   layer_shader_set_float(layer, effect_draw_shader, 'u_pattern_param2',   param2)
433   layer_shader_set_float(layer, effect_draw_shader, 'u_pattern_speed',    speed)
434   layer_shader_set_float(layer, effect_draw_shader, 'u_pattern_contrast', contrast)
435   -- image_field: sprite draws only. 1 = the image's own luminance drives the
436   -- field (dither/recolor a real image); 0 = the pattern drives it (default).
437   layer_shader_set_int  (layer, effect_draw_shader, 'u_image_field',
438                          spec.image_field and 1 or 0)
439   -- image_pattern_amount: luminance mode only. How much the animated pattern
440   -- ripples the image's field before dither (0 = static, →1 = flowing).
441   layer_shader_set_float(layer, effect_draw_shader, 'u_image_pattern_amount',
442                          spec.image_pattern_amount or 0)
443   layer_shader_set_float(layer, effect_draw_shader, 'u_time',             time)
444 end
445 
446 --[[
447   effect_clear(layer)
448   Reset pattern/color/dither on this layer to passthrough. The other
449   uniforms (color_a/b, scale, param) are left as-is — they only have an
450   effect when a non-zero kind reads them.
... [10 more lines]

459  --[[
460   effect_draw(layer, spec, fn, ...)
461   Scoped form: set the effect, run fn(...), then reset to passthrough.
462   Forwards extra arguments to fn so closures can stay light.
463 ]]
464 function effect_draw(layer, spec, fn, ...)
465   effect_set(layer, spec)
466   fn(...)
467   effect_clear(layer)
468 end
469 
470 -- =============================================================================
471 -- CYCLING — wrap-around step through each axis's named cycle.
472 -- Pass nil to get the first entry; pass an unknown name to get the first too.
473 -- =============================================================================
474 

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

485               palette.ink())
486 
487   row = rect_cut_top(body, ui_row_h + 20)
488   local ms = ui_slider{ rect = row, id = 'mod_scale', label = 'SCALE',
489                         value = sel.scale, vmin = 0.05, vmax = 1.2, fmt = '%.3f' }
490   if ms.changed then sel.scale = ms.value end
491 
492   row = rect_cut_top(body, ui_row_h + 20)
493   local msp = ui_slider{ rect = row, id = 'mod_speed', label = 'SPEED',
494                          value = sel.speed, vmin = 0, vmax = 4, fmt = '%.2f' }
495   if msp.changed then sel.speed = msp.value end
496 
497   row = rect_cut_top(body, ui_row_h + 20)
498   local mco = ui_slider{ rect = row, id = 'mod_contrast', label = 'CONTRAST',
499                          value = sel.contrast, vmin = 0, vmax = 2.5, fmt = '%.2f' }
500   if mco.changed then sel.contrast = mco.value end
501 
502   row = rect_cut_top(body, ui_row_h + 20)
503   local mpa = ui_slider{ rect = row, id = 'mod_param', label = sel.param_label,
504                          value = sel.param, vmin = 0, vmax = 1, fmt = '%.2f' }
505   if mpa.changed then sel.param = mpa.value end
506 
507   -- Second intrinsic knob — only patterns that define one (the gradients:
508   -- OFFSET / RANGE) get this row; it uses the pattern's own min/max range.
509   if sel.param2_label then
510     row = rect_cut_top(body, ui_row_h + 20)
511     local mp2 = ui_slider{ rect = row, id = 'mod_param2', label = sel.param2_label,
512                            value = sel.param2, vmin = sel.param2_min or 0,
513                            vmax = sel.param2_max or 1, fmt = '%.2f' }
514     if mp2.changed then sel.param2 = mp2.value end
... [22 more lines]

536    local cx, cy = ax + aw/2, ay + ah/2
537 
538   -- Both modes share the selected gallery pattern (LUMINANCE uses it to ripple
539   -- the dither; PATTERN uses it as the field), the shared color/dither, and the
540   -- shared SHAPE controls (so the image can be halftoned / icon-filled).
541   local g = gallery[gallery_selected]
542   local spec = {
543     pattern = g.pattern, pattern_scale = g.scale, speed = g.speed,
544     pattern_param = g.param, pattern_param2 = g.param2,
545     color = current_color, dither = current_dither,
546     color_a = current_color_a, color_b = current_color_b,
547     shape = current_shape, shape_size = current_shape_size,
548     shape_spacing = current_shape_spacing, shape_rotation = current_shape_rotation,
549     shape_backing = current_shape_backing,
550   }
551   if current_image_mode == 'luminance' then
552     spec.image_field = true
553     spec.contrast = current_image_contrast
554     spec.image_pattern_amount = current_image_pattern_amount
555   else
556     spec.contrast = g.contrast
557   end
558   effect_set(game_layer, spec)
559   -- Bind the icon mask for sprite shapes (same as the gallery), on unit 2.
560   if shape_is_sprite(current_shape) and icons[current_shape_icon] then
561     layer_shader_set_texture(game_layer, effect_draw_shader, 'u_shape_tex',
562                              icons[current_shape_icon].handle, 2)
563   end
564   layer_push(game_layer, cx, cy, 0, sc, sc)   -- translate + scale the image draw
565   layer_image(game_layer, img, 0, 0)          -- centered at the pushed origin
... [10 more lines]

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

615  
616   -- The UI runs HERE, not in update(). Under render_uncapped on a monitor
617   -- faster than the 120Hz fixed-timestep update, render frames can fire
618   -- without a preceding update step; anything queued only in update() gets
619   -- consumed by layer_render and then renders empty (blank FBO) on those
620   -- frames → flicker. Queueing the UI in draw() (which runs every render
621   -- frame) avoids that entirely. ui_update reads the edge globals captured
622   -- in update() and may mutate current_* — run before the game shapes so a
623   -- click applies to the scene the same frame.
624   ui_state_begin_frame()
625   ui_update()
626 
627   -- IMAGE screen: content is the chosen test image through the effect (no camera
628   -- — the image is screen-space via layer_push). Composite, consume edges, done.
629   if current_screen == 'image' then
630     image_draw()
631     layer_render(game_layer)
632     layer_render(ui_layer)
633     layer_draw(game_layer)
634     layer_draw(ui_layer)
635     ui_mouse_pressed  = false
636     ui_mouse_released = false
637     return
638   end
639 
640   -- Background fill in the active palette's field token. Drawn WITHOUT the
641   -- effect pipeline (color='none' default state at top of game_layer), so
642   -- the background stays a clean flat token regardless of the spec.
643   layer_rectangle(game_layer, 0, 0, width, height, palette.field())
644 
... [22 more lines]

666        contrast      = g.contrast,
667       pattern_param = g.param,
668       pattern_param2 = g.param2,
669       color         = current_color,
670       color_a       = current_color_a,
671       color_b       = current_color_b,
672       dither        = current_dither,
673       shape         = current_shape,
674       shape_size    = current_shape_size,
675       shape_spacing = current_shape_spacing,
676       shape_rotation = current_shape_rotation,
677       shape_backing = current_shape_backing,
678     })
679     layer_rectangle(game_layer, r.x, r.y, GAL_RECT_W, GAL_RECT_H, palette.ink())
680   end
681 
682   effect_clear(game_layer)
683 
684   camera_detach(main_camera, game_layer)
685 
686   -- Gallery labels on ui_layer (no effect pipeline). Drawn after detach;
687   -- ui_layer is screen-space and the camera is identity here, so these
688   -- align under their rects. "<name>  (<scale>)" per cell.
689   for i, g in ipairs(gallery) do
690     local r = gallery_rect(i)
691     layer_text(ui_layer, g.label .. '  (' .. string.format('%.2f', g.scale) .. ')',

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

File created successfully at: E:\a327ex\ricochet-template\api_showcase.lua (file state is current in your context — no need to Read it back)

OK: main.lua
OK: effect.lua
OK: api_showcase.lua

Your questions have been answered: "The editor will have an IMAGE target, which makes the separate IMAGE screen redundant. Keep both, or fold it in?"="Fold into editor, drop it", "Include a small live code readout under the preview (the effect_* call your edits build), or keep the editor purely visual?"="No, editor only". You can now continue with these answers in mind.

290    -- MODIFIERS section below reflects the new selection this same frame.
291   if current_screen == 'gallery' and ui_mouse_pressed then
292     local mx, my = mouse_position()
293     for i = 1, #gallery do
294       if ui_point_in_rect(mx, my, gallery_rect(i)) then gallery_selected = i end
295     end
296   end
297 
298   local panel_r = rect_new(SIDEBAR_X, SIDEBAR_Y, SIDEBAR_W, SIDEBAR_H)
299   ui_panel{ rect = panel_r, title = 'effect control', corner_brackets = true }
300 
301   -- Carve out the panel body: title bar at top, inset on the remaining area.
302   local body = rect_new(panel_r.x, panel_r.y, panel_r.w, panel_r.h)
303   rect_cut_top(body, ui_title_h)                         -- skip the title bar
304   body = rect_contract(body, ui_panel_inset)             -- inner padding
305 
306   -- PALETTE — dark/light segmented toggle.
307   local row = rect_cut_top(body, ui_row_h)
308   local seg = ui_segment{
309     rect = row, id = 'palette', label = 'PALETTE',
310     options = { 'DARK', 'LIGHT' },
311     index   = palette_mode == 'dark' and 1 or 2,
312   }
313   if seg.changed then
314     palette_init(seg.index == 1 and 'dark' or 'light')
315   end
316 
317   -- IMAGE-screen-only controls: field-source MODE + which test IMAGE. The COLOR
318   -- / DITHER / COLOR A·B / ramp rows below are shared and re-skin the image live.
319   if current_screen == 'image' then
... [145 more lines]

464      if rc.next_clicked and #effect_ramp < 8 then
465       table.insert(effect_ramp, ramp_selected + 1, effect_ramp[ramp_selected])
466       ramp_selected = ramp_selected + 1
467     end
468     if rc.prev_clicked and #effect_ramp > 2 then
469       table.remove(effect_ramp, ramp_selected)
470       if ramp_selected > #effect_ramp then ramp_selected = #effect_ramp end
471     end
472 
473     row = rect_cut_top(body, swatch_row_h)
474     local rs = ui_swatch_row{ rect = row, id = 'ramp_stops', label = 'RAMP',
475                               tokens = effect_ramp, selected_index = ramp_selected }
476     if rs.index then ramp_selected = rs.index end
477 
478     row = rect_cut_top(body, swatch_row_h)
479     local rp = ui_swatch_row{ rect = row, id = 'ramp_pick', label = 'STOP',
480                               tokens = palette_token_names,
481                               selected = effect_ramp[ramp_selected] }
482     if rp.changed then effect_ramp[ramp_selected] = rp.selected end
483   end
484 
485   -- ---- MODIFIERS for the selected gallery rect (GALLERY screen only) ----
486   -- Per-pattern scale / speed / contrast + the pattern's intrinsic knob.
487   -- Each gallery entry stores its own values; these sliders edit whichever
488   -- rect is selected (click a rect to switch which one they drive).
489   if current_screen ~= 'image' then
490   local sel = gallery[gallery_selected]
491   rect_cut_top(body, ui_section_gap)
492   local mhead = rect_cut_top(body, ui_row_h)
493   layer_text(ui_layer, 'MODIFIERS  ' .. sel.label, ui_font_label,
... [45 more lines]

File created successfully at: E:\a327ex\ricochet-template\ui_controls.lua (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

360    if not img then return end
361 
362   local ax, ay = GAL_X0, GAL_Y0
363   local aw, ah = width - ax - 40, height - ay - 80
364 
365   local g = gallery[current_image_pattern_idx]
366   local spec = {
367     pattern = g.pattern, pattern_scale = g.scale, speed = g.speed,
368     pattern_param = g.param, pattern_param2 = g.param2,
369     color = current_color, dither = current_dither,
370     color_a = current_color_a, color_b = current_color_b,
371     shape = current_shape, shape_size = current_shape_size,
372     shape_spacing = current_shape_spacing, shape_rotation = current_shape_rotation,
373     shape_backing = current_shape_backing,
374     shape_icon = shape_is_sprite(current_shape) and icons[current_shape_icon] or nil,
375   }
376   if current_image_mode == 'luminance' then
377     spec.image_field = true
378     spec.contrast = current_image_contrast
379     spec.image_pattern_amount = current_image_pattern_amount
380   else
381     spec.contrast = g.contrast
382   end
383   effect_image(game_layer, img, ax, ay, aw, ah, spec)
384 
385   -- Label (ui_layer, screen-space) above the image.
386   layer_text(ui_layer,
387     string.upper(current_image) .. '  ·  ' .. string.upper(current_image_mode) .. '  (' .. g.label .. ')',
388     ui_font_label, ax, ay - 30, palette.ink())
389 end
... [30 more lines]

344                             vmax = sel.param2_max or 1, fmt = '%.2f' }
345     if mp2.changed then sel.param2 = mp2.value end
346   end
347 end
348 
349 -- -----------------------------------------------------------------------------
350 -- image_draw — IMAGE screen content: the chosen test image fit into the content
351 -- area, processed through the effect via the single-call effect_image. LUMINANCE
352 -- → the image's own brightness is the field (dither/recolor it, optionally
353 -- rippled by the moving pattern); PATTERN → the chosen pattern fills it, stenciled
354 -- by the image's alpha. The image PATTERN picker selects which gallery pattern
355 -- drives both (its entry supplies the tuned scale / param).
356 -- -----------------------------------------------------------------------------
357 local function image_draw()
358   layer_rectangle(game_layer, 0, 0, width, height, palette.field())
359   local img = content_imgs[current_image]
360   if not img then return end
361 

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

400      ui_mouse_pressed  = false
401     ui_mouse_released = false
402     return
403   end
404 
405   -- API reference screen: a grid of example effect_* calls, each rendered (on
406   -- game_layer, through the pipeline) next to its generated source (on ui_layer).
407   if current_screen == 'api' then
408     api_showcase_draw()
409     layer_render(game_layer)
410     layer_render(ui_layer)
411     layer_draw(game_layer)
412     layer_draw(ui_layer)
413     ui_mouse_pressed  = false
414     ui_mouse_released = false
415     return
416   end
417 
418   -- The UI runs HERE, not in update(). Under render_uncapped on a monitor
419   -- faster than the 120Hz fixed-timestep update, render frames can fire
420   -- without a preceding update step; anything queued only in update() gets
421   -- consumed by layer_render and then renders empty (blank FBO) on those
422   -- frames → flicker. Queueing the UI in draw() (which runs every render
423   -- frame) avoids that entirely. ui_update reads the edge globals captured
424   -- in update() and may mutate current_* — run before the game shapes so a
425   -- click applies to the scene the same frame.
426   ui_state_begin_frame()
427   ui_update()
428 
429   -- IMAGE screen: content is the chosen test image through the effect (no camera
... [30 more lines]

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

File created successfully at: E:\a327ex\ricochet-template\editor.lua (file state is current in your context — no need to Read it back)

deleted api_showcase.lua
OK: main.lua
OK: effect.lua
OK: ui_controls.lua
OK: editor.lua
OK: showcase.lua

=== dangling refs (should be empty) ===
editor.lua:74:  if editor_target == 'image' and current_image_mode == 'luminance' then
editor.lua:107:  if editor_target == 'image' then
editor.lua:198:  if editor_target == 'image' then
=== current_screen usages ===
169:current_screen = 'gallery'   -- 'gallery' | 'showcase' | 'editor'
385:    if current_screen == 'gallery' then current_screen = 'showcase'
386:    elseif current_screen == 'showcase' then current_screen = 'editor'
387:    else current_screen = 'gallery' end
395:  if current_screen == 'showcase' then
416:  if current_screen == 'editor' then
=== confirm shared fns referenced ===
main.lua:301:  ui_shape_block(body)
main.lua:303:  ui_color_ab_block(body)
main.lua:304:  ui_ramp_block(body)
main.lua:417:    editor_ui()
main.lua:418:    editor_preview()
editor.lua:85:function editor_ui()
editor.lua:167:  ui_shape_block(body)
editor.lua:169:  ui_color_ab_block(body)
editor.lua:170:  ui_ramp_block(body)
editor.lua:174:function editor_preview()

278    local panel_r = rect_new(SIDEBAR_X, SIDEBAR_Y, SIDEBAR_W, SIDEBAR_H)
279   ui_panel{ rect = panel_r, title = 'effect control', corner_brackets = true }
280 
281   -- Carve out the panel body: title bar at top, inset on the remaining area.
282   local body = rect_new(panel_r.x, panel_r.y, panel_r.w, panel_r.h)
283   rect_cut_top(body, ui_title_h)                         -- skip the title bar
284   body = rect_contract(body, ui_panel_inset)             -- inner padding
285 
286   -- PALETTE — dark/light segmented toggle.
287   local row = rect_cut_top(body, ui_row_h)
288   local seg = ui_segment{
289     rect = row, id = 'palette', label = 'PALETTE',
290     options = { 'DARK', 'LIGHT' },
291     index   = palette_mode == 'dark' and 1 or 2,
292   }
293   if seg.changed then
294     palette_init(seg.index == 1 and 'dark' or 'light')
295   end
296 
297   -- Shared look controls (color recipe + tokens + dither + shape + ramp), drawn
298   -- by the shared blocks in ui_controls.lua (reused by the editor screen too).
299   ui_color_recipe_row(body)
300   ui_dither_row(body)
301   ui_shape_block(body)
302   rect_cut_top(body, ui_section_gap)
303   ui_color_ab_block(body)
304   ui_ramp_block(body)
305 
306   -- ---- MODIFIERS for the selected gallery rect ----
307   -- Per-pattern scale / speed / contrast + the pattern's intrinsic knob.
... [42 more lines]

=== current_spec (dead?) ===
185:local function current_spec()
=== files present ===
editor.lua
effect.lua
main.lua
palette.lua
rectcut.lua
showcase.lua
ui.lua
ui_controls.lua
ui_state.lua
=== editor/ui_controls required? ===
24:require('anchor')({
34:require('palette')
35:require('effect')
36:require('rectcut')
37:require('ui_state')
38:require('ui')
39:require('ui_controls')
40:require('showcase')
41:require('editor')

183  ui_mouse_released = false
184 
185 local function current_spec()
186   return {
187     pattern       = current_pattern,
188     color         = current_color,
189     dither        = current_dither,
190     color_a       = current_color_a,
191     color_b       = current_color_b,
192     pattern_scale = current_scale,
193   }
194 end
195 
196 -- -----------------------------------------------------------------------------

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: main.lua

46   ## What this is
47  
48  A starter Anchor 2 project that ships **Ricochet's visual identity** plus
49  a clean four-axis effect system (pattern × color × dither × shape). Fork the
50  folder, drop your game code into `main.lua`, prototype. The template
51  itself ships no gameplay — that's the forking dev's job.
52  
53  The full visual identity (palette philosophy, dotted motif, typography,
54  shape language, motion principles) lives in `Ricochet/design.md`. This
55  template implements the palette + the effect pipeline only; everything
56  else in design.md is guidance for how a forked project should look.
57  
58  Layout:
59  - **`main.lua`** — the test scene: a **pattern gallery** (one rect per
60    pattern, the `gallery` table) + a **click-to-select inspector** (the
61    sidebar edits the selected rect's modifiers). Owns boot order, layers,
62    the per-frame UI contract, and the keyboard hotkeys.
63  - **`palette.lua`** — `palette_dark` + `palette_light`, 13 named tokens
64    per palette. `palette_init('dark' | 'light')` selects the active one.
65  - **`effect.lua`** — the four-axis effect API: pattern/color/dither/shape
66    registries, `effect_set(layer, spec)`, cycle helpers.
67  - **`rectcut.lua`** — RectCut layout primitive (cut/get/add/contract/
68    center/split). Verbatim from snkrx-template's `snkrx/ui/rect.lua`.
69  - **`ui_state.lua`** — frame-local hot/active/focus tracking +
70    `ui_claim_hot`. Slimmed from snkrx-template's `snkrx/ui/state.lua`
71    (text-input + modal-lock bits removed).
72  - **`ui.lua`** — the five Ricochet-style widgets that build the sidebar
73    control panel (`ui_panel`, `ui_field`, `ui_segment`, `ui_swatch_row`,
74    `ui_slider`) plus internal `ui_icon_button`.
75  - **`assets/draw_shader.frag`** — the unified fragment shader. Replaces
... [45 more lines]

The file E:\a327ex\ricochet-template\.claude\CLAUDE.md has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\.claude\CLAUDE.md has been updated successfully. (file state is current in your context — no need to Read it back)

122  1. **480×270 support.** Make the effects (and eventually the dev UI) read at the
123    SNKRX render target, where 1 logical px = 1 dither cell. The shader is already
124    `gl_FragCoord`-anchored, so it's mostly a verify + tune pass, but the user
125    wants to do it by hand.
126 2. **Replace the snkrx-template effect system** (its `edition.lua` + dither
127    stack) with this one, once 480×270 is settled. This is the end goal of the
128    whole template.
129 3. **Flow-field shape rotation** (the Orblike/Invoker "wall icon" effect): a
130    *per-cell* rotation angle from a field, vs. today's single global ROTATION.
131    Feasible; bigger. The hook is the rotation math at the top of `shape_mask`.
132 
133 **When this toolkit is used in a REAL game:** the showcase UI will want
134 restyling to match the Ricochet design sheets (`Ricochet/reference/`) more
135 closely in *feel*. (User flagged this explicitly.)
136 
137 ## The effect system (pattern × color × dither × shape)
138 
139 Every draw routes through `assets/draw_shader.frag`. The shader composes:
140 
141 ```
142 f   = pattern(world_pos, time*speed, scale, param, param2)   // [0,1]
143 f   = contrast_remap(f)                              // spread/compress around 0.5
144 f   = shape(f, pixel_pos, shape_size)                // carve into a dot grid (or passthrough)
145 f   = dither(f, pixel_pos)                           // {0, 1/n, …, 1}
146 rgb = color(base_rgb, f, palette[color_a], palette[color_b])
147 ```
148 
149 Four orthogonal selectors (pattern / color / dither / shape) plus three
150 universal **modifiers** (scale / speed / contrast) and **two per-pattern
151 intrinsic knobs** (`param` / `param2`). Pattern owns spatial structure AND

151  intrinsic knobs** (`param` / `param2`). Pattern owns spatial structure AND
152 animation; shape optionally re-expresses the field as a dot grid; color is a
153 pure recipe that picks/mixes palette tokens by f; dither inserts the
154 pixel-art stipple between them.
155 
156 ### Patterns (`pattern_kinds` in `effect.lua`; dispatch in the shader's `pattern_field`)
157 
158 Two base kinds + seven world-anchored "gallery" patterns + two **LOCAL**
159 gradient patterns. Each gallery pattern has ONE intrinsic `param` knob whose
160 meaning differs per pattern; its DEFAULT value reproduces the canonical look.
161 
162 | kind | name            | character                        | param (intrinsic knob)    | param2 (2nd knob)        |

The file E:\a327ex\ricochet-template\.claude\CLAUDE.md has been updated successfully. (file state is current in your context — no need to Read it back)

250      trap). Use DITHER `off` for a smooth ramp, or a dither with **>= N levels**
251     (`bayer4_8`, `bayer8_16`, `noise_8`) to keep the middle stops.
252 
253 ### Dither (`dither_kinds` — 34 modes)
254 Full port of snkrx-template's dither catalog: a 2-level group (bayer4_2,
255 bayer8_2, noise_2) plus multi-level Bayer / white-noise / cluster-dot /
256 diagonal-line / scanline / triangle / hearts families. The complete list +
257 human labels are in `dither_kinds` and `dither_pretty_names` in `effect.lua`.
258 The level count is baked into each name (`bayer4_4` = 4×4 matrix, 4 levels);
259 2-level = hard A/B, higher = smoother `mix` transitions.
260 
261 ### Shape (`shape_kinds` — circle dot grid)
262 A stage between `effect_field` and dither that re-expresses the field as a grid
263 of shapes. It still dithers + colors normally afterward. `none` is passthrough.
264 Cells are `shape_size` px square, anchored to screen space (`gl_FragCoord`, the
265 same space as dither).
266 
267 **The general model (`shape_mask` + `effect_field_shaped`), built to extend to
268 image/sprite shapes later. The field has TWO roles, kept on SEPARATE channels
269 — conflating them is what made gaps + dot-edges collapse to `color_ramp`'s stop
270 0 (the "everything goes red" trap):**
271 - **Coverage** — *where* a dot is. The mask `m`, a disc whose size comes from
272   the **per-cell** value `fc` (sampled once at the cell center via
273   `effect_field_shaped`, reconstructing world/uv there from `dFdx`/`dFdy` —
274   exact for the 2D affine camera). Per-cell sampling keeps the boundary a
275   **true circle**. Coverage drives compositing (alpha / blend), **not** color.
276   `shape_mask()` is the single extension point: an image shape returns
277   `texture(u_shape_tex, cell_uv).a` there; nothing downstream changes.
278 - **Color** — *what hue*. The field value fed to dither + the color recipe at
279   **full strength** (never multiplied by the mask). `circle` uses the
... [38 more lines]

317    color_a = 'ink',         -- palette token name
318   color_b = 'field',       -- palette token name (used by mix)
319   ramp    = {'red','orange','yellow','green','blue','pink'},  -- token list, used by color='ramp'
320   pattern_scale  = 0.4,    -- feature size
321   speed          = 1.0,    -- animation rate (default 1)
322   contrast       = 1.0,    -- mid-range spread (default 1 = unchanged)
323   pattern_param  = 0.5,    -- the active pattern's intrinsic knob, [0,1]
324   pattern_param2 = 0,      -- 2nd intrinsic knob (gradient OFFSET / RANGE)
325 })
326 -- ... draws here ...
327 effect_clear(layer)        -- or the scoped effect_draw(layer, spec, fn)
328 ```
329 
330 Every field optional; omitted → documented default. **speed/contrast
331 default to 1.0, NOT 0** (0 would freeze animation / flatten the field to
332 mid-grey). Default spec is `color='none'`, so a bare `effect_set` is a
333 no-op until you set a color recipe.
334 
335 ---
336 
337 ## Pattern exploration history (don't re-tread)
338 
339 The pattern set was found by trial across the "Visual Experiments 1"
340 session. A new instance should NOT re-propose the rejected ideas without a
341 fresh reason — they were built and cut.
342 
343 **Rejected outright (built, looked wrong, removed):**
344 - **worley / cellular** — only read as cells in a narrow scale window;

The file E:\a327ex\ricochet-template\.claude\CLAUDE.md has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\.claude\CLAUDE.md has been updated successfully. (file state is current in your context — no need to Read it back)

386  ---
387 
388 ## Palette
389 
390 13 named tokens per palette mode. Light mode uses Ricochet's "ink on warm
391 paper" colors; dark mode uses the "blueprint room with the lights off"
392 counterparts (see `Ricochet/design.md` § Dark mode for the role mapping).
393 
394 Keys (shared across both palettes):
395 ```
396 neutrals:  field  panel  recess  ink  mid_light  mid_grey  deep_grey
397 accents:   blue   orange green   pink yellow     red
398 ```
399 
400 Light mode collapses panel→field and deep_grey→recess (no separate
401 elevation surface defined in design.md). Forked games that need
402 explicit elevation in light mode override the table.
403 
404 `palette.lua` exposes:
405 - `palette_dark`, `palette_light` — the two tables
406 - `palette` — the currently-active table (set by `palette_init`)
407 - `palette_mode` — `'dark'` or `'light'`
408 - `palette_token_names`, `palette_token_index`, `palette_token_count`
409 
410 For raw layer_* draws, call `palette.ink()` etc. (the parens trigger the
411 color metatable's __call → rgba tuple). For shader-pipeline draws, pass
412 the token NAME via `color_a`/`color_b` in the spec.
413 
414 ---
415 
... [30 more lines]

445  tooltip, leaderboard list), CONTROLS (tabs, toggle/checkbox/radio, stepper, stat
446 block, keycaps, bracket tags). Each is a small local helper taking `(x, y, …)`
447 and returning the next `y`.
448 
449 **Icon pack: Tabler** (chosen over Phosphor/Remix — most icons + best game
450 coverage of the clean line packs + freely adjustable stroke). Icons are 44px
451 white PNGs at `assets/icons/tabler/<concept>.png`, loaded flat into
452 `icons[concept]` and tinted to a palette token. Regenerate / extend the set via
453 `tools/fetch_icons.mjs` (Iconify API → resvg; `STROKE = 0.75` for the hairline
454 look). `tools/compare_packs.mjs` is the pack-comparison helper (counts + game
455 coverage).
456 
457 NOTE: `F1`/`F2` (pattern) and `[`/`]` (scale) still mutate
458 `current_pattern` / `current_scale`, but the gallery ignores those globals
459 (every rect is per-pattern, per-scale via the inspector), so they're inert
460 in gallery mode. Color / dither / tokens / palette ARE shared, so their
461 keys work.
462 
463 ---
464 
465 ## UI control panel
466 
467 A left-anchored sidebar (380×700, top-left at (32, 40)). The gallery sits
468 to its right. The sidebar has two regions:
469 - **Shared look controls** (apply to every gallery rect): PALETTE segment,
470   COLOR recipe, DITHER, SHAPE (+ CELL SIZE slider, shown only when SHAPE ≠
471   off), COLOR A, COLOR B.
472 - **MODIFIERS** (apply only to the currently-selected gallery rect):
473   SCALE / SPEED / CONTRAST sliders + the selected pattern's intrinsic-knob
474   slider(s) (labeled per pattern, e.g. WARP / VEIN / LINES; gradients add a

The file E:\a327ex\ricochet-template\.claude\CLAUDE.md has been updated successfully. (file state is current in your context — no need to Read it back)

461  
462 ## UI control panel
463 
464 A left-anchored sidebar (380×700, top-left at (32, 40)). The gallery sits
465 to its right. The sidebar has two regions:
466 - **Shared look controls** (apply to every gallery rect): PALETTE segment,
467   COLOR recipe, DITHER, SHAPE (+ CELL SIZE slider, shown only when SHAPE ≠
468   off), COLOR A, COLOR B.
469 - **MODIFIERS** (apply only to the currently-selected gallery rect):
470   SCALE / SPEED / CONTRAST sliders + the selected pattern's intrinsic-knob
471   slider(s) (labeled per pattern, e.g. WARP / VEIN / LINES; gradients add a
472   2nd knob OFFSET / RANGE). Click a rect to choose which one these edit.
473 
474 ### The widgets
475 
476 All five draw to `ui_layer`. Each is a function taking one opts table;
477 state is caller-owned (pass `value` / `selected` / `index`; widget
478 reports `clicked` / `changed` / new value/index; caller updates its
479 own globals).
480 
481 - **`ui_panel(opts)`** — hairline-bordered container with optional
482   title bar and optional corner brackets (small L shapes outside each
483   corner, per Ricochet/design.md's chamber treatment).
484 - **`ui_field(opts)`** — `LABEL  value  [<] [>]` row. The cycler used
485   for COLOR and DITHER.
486 - **`ui_segment(opts)`** — N-cell segmented control. Active cell is
487   drawn with an INVERTED fill (ink rectangle, field-colored label) —
488   matches the reference sheets' button-active idiom. The palette-mode
489   toggle uses it.
490 - **`ui_swatch_row(opts)`** — palette-token picker: label above, row

The file E:\a327ex\ricochet-template\.claude\CLAUDE.md has been updated successfully. (file state is current in your context — no need to Read it back)

640  ### Draw origin: images CENTER, rectangles/text TOP-LEFT
641 `layer_image` / `layer_texture` / `layer_circle` draw **centered** at `(x, y)`;
642 `layer_rectangle` / `layer_text` draw from the **top-left** `(x, y)`. Do NOT
643 subtract half the image size to "center" an image — it already centers, so that
644 double-offsets it up-left by half (the showcase icon bug). To center an icon in
645 a top-left box `{x, y, w, h}`: `layer_image(lyr, img, x + w/2, y + h/2, col)`.
646 See `showcase.lua`'s `icon()` helper.
647 
648 ### Color in draws vs in specs
649 - **Raw layer_* draws** (background fill, HUD text, anything that should
650   bypass the pipeline) → pass a color object directly, with parens:
651   `layer_rectangle(game_layer, 0, 0, w, h, palette.field())`.
652 - **Pipeline draws** (anything that should be re-colored by the effect) →
653   pass token NAMES in the spec: `color_a = 'ink'`. The base color of the
654   draw is then irrelevant (use any visible token so layout-only debugging
655   remains possible).
656 
657 ### `effect_clear` after a bracketed set
658 The shader uniforms persist on the layer until rewritten. Use
659 `effect_clear(layer)` after a bracketed pipeline draw, or use the scoped
660 `effect_draw(layer, spec, fn, ...)` form which clears for you.
661 
662 ### Don't run `./run.bat` yourself
663 The user runs and tests. After framework changes, syntax-check Lua with:
664 ```
665 E:/a327ex/Anchor/reference/love-compare/tools/lua54.exe -e "loadfile('main.lua')"
666 ```
667 
668 ---
669 
... [45 more lines]

The file E:\a327ex\ricochet-template\.claude\CLAUDE.md has been updated successfully. (file state is current in your context — no need to Read it back)

The file E:\a327ex\ricochet-template\.claude\CLAUDE.md has been updated successfully. (file state is current in your context — no need to Read it back)

714  
715 ---
716 
717 ## File map
718 
719 ```
720 ricochet-template/
721 ├── main.lua                    # pattern gallery + inspector sidebar + keys
722 ├── palette.lua                 # 2 palettes × 13 tokens; palette_init
723 ├── effect.lua                  # 4-axis spec API (pattern/color/dither/shape) + cycle helpers
724 ├── rectcut.lua                 # RectCut layout primitive
725 ├── ui_state.lua                # frame-local hot/active/focus + ui_claim_hot
726 ├── ui.lua                      # 5 Ricochet-style widgets + helpers
727 ├── anchor/                     # Anchor 2 framework
728 ├── anchor.exe                  # built engine
729 ├── assets/
730 │   ├── draw_shader.frag        # unified pattern+shape+dither+color shader
731 │   ├── Oswald-Medium.ttf       # UI titles + field labels
732 │   ├── IBMPlexSans-Regular.ttf # UI labels (alt)
733 │   └── CascadiaCode-Regular.ttf # UI value column + numeric readouts
734 ├── run.bat
735 └── .claude/CLAUDE.md           # this file
736 ```
737 
738 ---
739 
740 ## Working style
741 
742 Same conventions as Anchor 2 / Ricochet / Invoker:
743 
... [5 more lines]

The file E:\a327ex\ricochet-template\.claude\CLAUDE.md has been updated successfully. (file state is current in your context — no need to Read it back)

626:## Critical conventions
627-
628:### Boot order
629-1. `require('anchor')(...)` — engine init.
630-2. `require('palette')` + `require('effect')` — module loads.
631-3. `palette_init('dark')` — populates `palette`.
632-4. `font_register(...)` — fonts for HUD/text.
633-5. `effect_setup({ draw_shader = '...' })` — installs shader AND pushes
634-   the active palette to `u_palette[]`.
635-
636-Steps 3 and 5 can be reversed — `palette_init` checks if
637-`effect_write_palette` exists and calls it if so, otherwise `effect_setup`
638-picks up the current palette when it runs. Either order works.
639-
640-### Draw origin: images CENTER, rectangles/text TOP-LEFT
641-`layer_image` / `layer_texture` / `layer_circle` draw **centered** at `(x, y)`;
642-`layer_rectangle` / `layer_text` draw from the **top-left** `(x, y)`. Do NOT
643-subtract half the image size to "center" an image — it already centers, so that
644-double-offsets it up-left by half (the showcase icon bug). To center an icon in
645-a top-left box `{x, y, w, h}`: `layer_image(lyr, img, x + w/2, y + h/2, col)`.
646-See `showcase.lua`'s `icon()` helper.

The file E:\a327ex\ricochet-template\.claude\CLAUDE.md has been updated successfully. (file state is current in your context — no need to Read it back)

1    --[[
2     effect.lua — three-axis visual effect system (pattern × color × dither).
3   
4     THE 30-SECOND MENTAL MODEL
5       Every draw routes through assets/draw_shader.frag. The shader takes
6       three orthogonal selectors and composes them as:
7   
8           f = pattern(world_pos, time)          -- spatial structure + animation
9           f = dither(f, pixel_pos)              -- quantize to discrete levels
10          rgb = color(base_rgb, f, palette)     -- map f → palette token(s)
11  
12      Pattern owns "what the field looks like" AND "whether it moves over
13      time." Color is a pure recipe that picks (or mixes) palette tokens
14      by f. Dither inserts the pixel-art stipple between them. The three
15      are independent — any combination is valid.
16  
17    PUBLIC API
18      effect_setup(opts)                       install draw shader, cache its
19                                               GL program ID, push the palette
20      effect_set(layer, spec)                  write effect uniforms for this
21                                               layer's subsequent draws
22      effect_clear(layer)                      pattern/color/dither → 0 (passthrough)
23      effect_draw(layer, spec, fn, ...)        scoped form
24      effect_write_palette()                   re-push u_palette to the shader
25                                               (called automatically by palette_init
26                                               and effect_setup)
27  
28      Cycle helpers (return the NEXT/PREV name in each axis's cycle):
29        effect_next_pattern(cur), effect_prev_pattern(cur)
30        effect_next_color(cur),   effect_prev_color(cur)
... [5 more lines]

The file E:\a327ex\ricochet-template\effect.lua has been updated successfully. (file state is current in your context — no need to Read it back)

1    --[[
2     ricochet-template — visual style + three-axis effect template.
3   
4     Entry point. Owns:
5       - configuration (resolution, render mode, monitor)
6       - boot order (palette → fonts → effect/shader → ui modules)
7       - layers (game, ui), camera
8       - current effect-spec state (pattern, color, dither, color_a/b, scale)
9       - sidebar control panel + keyboard input (both mutate the same state)
10      - test-scene draw (sample shapes through the effect pipeline)
11  
12    Delegates:
13      - palette.lua     — dark/light palette tables + palette_init
14      - effect.lua      — pattern × color × dither spec API + cycle helpers
15      - rectcut.lua     — RectCut layout primitive (cut/get/add/contract/...)
16      - ui_state.lua    — frame-local hot/active/focus + ui_claim_hot
17      - ui.lua          — five Ricochet-style widgets that mutate effect state
18  
19    Goal: visually evaluate any (pattern, color, dither, color_a, color_b,
20    scale) combination by clicking sidebar controls or pressing the hotkeys.
21    Both input modes update the same caller-owned state.
22  ]]
23  

The file E:\a327ex\ricochet-template\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

OK: main.lua
OK: effect.lua

File created successfully at: C:\Users\adn\.claude\projects\E--a327ex\memory\reference_anchor2_aux_sampler_draws.md (file state is current in your context — no need to Read it back)

1    # Memory
2   
3   ## Code Style - Emoji Ball Battles (Lua)
4   
5   ### Operator spacing convention
6   - `*` and `/`: no spaces (e.g., `a*b`, `x/y`)
7   - `+` and `-`: spaces (e.g., `a + b`, `x - y`)
8   
9   Example:
10  ```lua
11  result = a*b + c/d - e
12  velocity = speed*math.cos(angle) + offset
13  ```
14  
15  ## Conversion History
16  
17  - **Emoji Ball Battles**: Converted from YueScript to Lua (2026-02-24). Archived `.yue` files in `emoji-ball-battles/reference/yue/`. Framework files copied from `Anchor/framework/anchor/`.
18  - **Invoker** (pre-reset): Converted from YueScript to Lua (2026-02-25). Archived `.yue` files now live in `Invoker-old/reference/yue/` after the 2026-05-27 reset to snkrx-template. The new Invoker has no YueScript history.
19  
20  ## Naming Conventions
21  
22  - Always capitalize "Invoker" (capital O) — it's a proper name, not a directory path reference.
23  
24  ## Critical Warnings
25  - [Invoker runs on Anchor 2](feedback_invoker_anchor_divergence.md) — use Anchor2/docs/, not Anchor/docs/ (v1); Invoker/anchor/ is the actual ground truth
26  - [Anchor 2 mouse buttons follow LÖVE2D](reference_anchor2_mouse_buttons.md) — `mouse:2` = RMB, `mouse:3` = MMB (engine swaps 2↔3 at SDL boundary as of 2026-05-17)
27  - [Anchor 2 input edge events are update-only](reference_anchor2_input_in_draw.md) — `mouse_is_pressed/released` and `key_is_pressed/released` always return false from `draw()`; run UI logic (hit-testing, state machines, layer command queueing) in `update()` and reduce `draw()` to `layer_render + layer_draw`
28  - [Anchor 2 sprites/images draw CENTERED at (x,y)](reference_anchor2_sprite_centered.md) — `layer_image`/`layer_texture`/`layer_circle` center at (x,y); `layer_rectangle`/`layer_text` are TOP-LEFT. Don't subtract half-size to "center" an image — it already centers, so that double-offsets it up-left by half. To center an icon in a top-left box {x,y,w,h}: `layer_image(lyr, img, x + w/2, y + h/2, col)`
29  
30  ## Workflow
... [10 more lines]

The file C:\Users\adn\.claude\projects\E--a327ex\memory\MEMORY.md has been updated successfully. (file state is current in your context — no need to Read it back)

Launching skill: end-session

2026-05-30T20:25:57.330Z 919f238e-264d-437b-8ab7-fc807e5302b7 [claude] <-- MOST RECENT
   C:\Users\adn\.claude\projects\E--a327ex\919f238e-264d-437b-8ab7-fc807e5302b7.jsonl

2026-05-30T20:00:42.308Z aca39077-9a40-4d7e-ac77-c4c05ac51979 [claude]
   C:\Users\adn\.claude\projects\E--a327ex\aca39077-9a40-4d7e-ac77-c4c05ac51979.jsonl

2026-05-30T18:17:12.676Z 06f5cfc2-4245-4927-9a2d-d7f372358eff [claude]
   C:\Users\adn\.claude\projects\E--a327ex\06f5cfc2-4245-4927-9a2d-d7f372358eff.jsonl

2026-05-30T00:45:02.609Z 8794b075-651a-4460-a6f1-0b750c98fa26 [claude]
   C:\Users\adn\.claude\projects\E--a327ex\8794b075-651a-4460-a6f1-0b750c98fa26.jsonl

2026-05-30T00:41:53.417Z ced0b502-30f5-45ac-8a67-dc25bf7aae84 [claude]
   C:\Users\adn\.claude\projects\E--a327ex\ced0b502-30f5-45ac-8a67-dc25bf7aae84.jsonl

=== end timestamp ===
2026-05-30 17:26:34

=== ricochet-template git? ===
=== Anchor2 git? ===
true
origin  git@github.com:a327ex/Anchor2.git (fetch)
origin  git@github.com:a327ex/Anchor2.git (push)

is-work-tree: fatal: not a git repository (or any of the parent directories): .git
git-dir: fatal: not a git repository (or any of the parent directories): .git
remote:
fatal: not a git repository (or any of the parent directories): .git
=== top of status ===
fatal: not a git repository (or any of the parent directories): .git