kitty/kitty/shaders.c
Kovid Goyal 87d06354dd
Custom shaders: Fix focus driven groups ending in the wrong state after hide/unhide
Shader animation events are accumulated per OS window and only consumed
when the window is next rendered. A focus-out followed by a focus-in with
no rendered frame in between, as happens when the window is hidden and
shown again, or a brief focus steal within one repaint interval, left both
bits pending. The state machine then treated them as simultaneous, so the
focus-out hard stop prevented os-window-focus-in groups from starting and
groups started by focus-out were turned on while focused.

Now a focus change replaces any pending opposite focus change, so only the
final focus state is reported. Genuinely simultaneous events, such as a
click that moves focus away, still give focus-out priority.

Closes #10871
2026-11-15 08:39:25 +05:30

4184 lines
199 KiB
C

/*
* shaders.c
* Copyright (C) 2017 Kovid Goyal <kovid at kovidgoyal.net>
*
* Distributed under terms of the GPL3 license.
*/
#include "fonts.h"
#include "gl.h"
#include "cleanup.h"
#include "colors.h"
#include <stddef.h>
#include <string.h>
#include "animation.h"
#include "animation-parse.h"
#include "text-cache.h"
#include "tupleobject.h"
#include "window_logo.h"
#include "srgb_gamma.h"
#include "state.h"
enum {
CELL_PROGRAM,
CELL_FG_PROGRAM,
CELL_BG_PROGRAM,
CELL_PROGRAM_SENTINEL,
BORDERS_PROGRAM,
GRAPHICS_PROGRAM,
GRAPHICS_PREMULT_PROGRAM,
GRAPHICS_ALPHA_MASK_PROGRAM,
GRAPHICS_INDEXED_PROGRAM,
GRAPHICS_INDEXED_UNSCALED_PROGRAM,
BGIMAGE_PROGRAM,
TINT_PROGRAM,
TRAIL_PROGRAM,
BLIT_PROGRAM,
SCREENSHOT_PROGRAM,
ROUNDED_RECT_PROGRAM,
PADDING_PROGRAM,
CUSTOM_END_PROGRAM,
NUM_PROGRAMS
};
enum { SPRITE_MAP_UNIT, GRAPHICS_UNIT, SPRITE_DECORATIONS_MAP_UNIT, CUSTOM_END_TEXTURE_A_UNIT, CUSTOM_END_TEXTURE_B_UNIT, CUSTOM_END_TEXTURE_PERSIST_UNIT };
typedef struct UIRenderData {
unsigned screen_width, screen_height, cell_width, cell_height, screen_left, screen_top, full_framebuffer_width, full_framebuffer_height;
Window *window;
Screen *screen;
OSWindow *os_window;
GraphicsRenderData grd;
WindowLogoRenderData *window_logo;
float bg_alpha, inactive_text_alpha;
bool has_background_image;
color_type background_color; // RGB only
monotonic_t now;
} UIRenderData;
static inline float
row_offset_for_screen(const Screen *screen) {
if (!pixel_scroll_enabled(screen) || !screen->cell_size.height) return 0.f;
return -1.f + (float)(rendered_pixel_scroll_offset_y(screen) / (double)screen->cell_size.height);
}
// Sprites {{{
typedef struct {
int xnum, ynum, x, y, z, last_num_of_layers, last_ynum;
GLuint texture_id;
GLint max_texture_size, max_array_texture_layers;
struct decorations_map {
GLuint texture_id;
unsigned width, height;
size_t count;
} decorations_map;
} SpriteMap;
static const SpriteMap NEW_SPRITE_MAP = {.xnum = 1, .ynum = 1, .last_num_of_layers = 1, .last_ynum = -1};
static GLint max_texture_size = 0, max_array_texture_layers = 0;
static GLfloat
srgb_color(uint8_t color) {
return srgb_lut[color];
}
static void
color_vec3(GLint location, color_type color) {
glUniform3f(location, srgb_lut[(color >> 16) & 0xFF], srgb_lut[(color >> 8) & 0xFF], srgb_lut[color & 0xFF]);
}
static void
color_vec4_premult(GLint location, color_type color, GLfloat alpha) {
glUniform4f(location, srgb_lut[(color >> 16) & 0xFF] * alpha, srgb_lut[(color >> 8) & 0xFF] * alpha, srgb_lut[color & 0xFF] * alpha, alpha);
}
static void
color_vec4(GLint location, color_type color, GLfloat alpha) {
glUniform4f(location, srgb_lut[(color >> 16) & 0xFF], srgb_lut[(color >> 8) & 0xFF], srgb_lut[color & 0xFF], alpha);
}
// Write a color into a std140 float4 whose fourth component is unused. Uses
// the same table as the shaders so the result is identical to converting the
// color in the shader.
static void
linear_color_vec4(GLfloat *dest, color_type color) {
dest[0] = srgb_lut[(color >> 16) & 0xFF];
dest[1] = srgb_lut[(color >> 8) & 0xFF];
dest[2] = srgb_lut[color & 0xFF];
dest[3] = 0.f;
}
static void
clear_current_framebuffer(void) {
glClearColor(0, 0, 0, 0);
glClear(GL_COLOR_BUFFER_BIT);
}
// Must be called as soon as there is a current OpenGL context, see
// initialize_gpu(). Font groups are created before that happens and size their
// sprite layout using these limits, so the sooner they are known the better.
static void
query_gpu_texture_limits(void) {
if (max_texture_size) return;
glGetIntegerv(GL_MAX_TEXTURE_SIZE, &(max_texture_size));
glGetIntegerv(GL_MAX_ARRAY_TEXTURE_LAYERS, &(max_array_texture_layers));
#ifdef __APPLE__
// Since on Apple we could have multiple GPUs, with different capabilities,
// upper bound the values according to the data from https://developer.apple.com/graphicsimaging/opengl/capabilities/
max_texture_size = MIN(8192, max_texture_size);
max_array_texture_layers = MIN(512, max_array_texture_layers);
#endif
sprite_tracker_set_limits(max_texture_size, max_array_texture_layers);
}
SPRITE_MAP_HANDLE
alloc_sprite_map(void) {
query_gpu_texture_limits();
SpriteMap *ans = calloc(1, sizeof(SpriteMap));
if (!ans) fatal("Out of memory allocating a sprite map");
*ans = NEW_SPRITE_MAP;
ans->max_texture_size = max_texture_size;
ans->max_array_texture_layers = max_array_texture_layers;
return (SPRITE_MAP_HANDLE)ans;
}
void
free_sprite_data(FONTS_DATA_HANDLE fg) {
SpriteMap *sprite_map = (SpriteMap *)fg->sprite_map;
if (sprite_map) {
if (sprite_map->texture_id) free_texture(&sprite_map->texture_id);
if (sprite_map->decorations_map.texture_id) free_texture(&sprite_map->decorations_map.texture_id);
free(sprite_map);
fg->sprite_map = NULL;
}
}
static void
discard_pending_gl_errors(void) {
// glad's error checking callback is installed only with --debug-rendering,
// so in normal runs errors queue up silently. Drain the queue so that a
// subsequent glGetError() reports only what we are interested in.
for (unsigned i = 0; i < 16 && glGetError() != GL_NO_ERROR; i++);
}
static bool
copy_texture_via_framebuffer(GLuint old_texture, GLuint new_texture, GLenum texture_type, GLint width, GLint height, GLint layers) {
GLint prev_read_fbo;
glGetIntegerv(GL_READ_FRAMEBUFFER_BINDING, &prev_read_fbo);
discard_pending_gl_errors();
GLuint fbo;
glGenFramebuffers(1, &fbo);
glBindFramebuffer(GL_READ_FRAMEBUFFER, fbo);
bool ok = true;
for (GLint z = 0; z < layers && ok; z++) {
if (texture_type == GL_TEXTURE_2D_ARRAY) glFramebufferTextureLayer(GL_READ_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, old_texture, 0, z);
else glFramebufferTexture2D(GL_READ_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_2D, old_texture, 0);
ok = glCheckFramebufferStatus(GL_READ_FRAMEBUFFER) == GL_FRAMEBUFFER_COMPLETE;
if (!ok) break;
glBindTexture(texture_type, new_texture);
if (texture_type == GL_TEXTURE_2D_ARRAY) glCopyTexSubImage3D(texture_type, 0, 0, 0, z, 0, 0, width, height);
else glCopyTexSubImage2D(texture_type, 0, 0, 0, 0, 0, width, height);
// A complete framebuffer does not mean the copy itself was accepted,
// for instance if the two internal formats are not copy compatible.
// Without this check we would claim success and leave the new texture
// blank, losing every already rendered sprite.
ok = glGetError() == GL_NO_ERROR;
}
glBindFramebuffer(GL_READ_FRAMEBUFFER, prev_read_fbo);
glDeleteFramebuffers(1, &fbo);
if (!ok) glBindTexture(texture_type, old_texture);
return ok;
}
static void
copy_32bit_texture(GLuint old_texture, GLuint new_texture, GLenum texture_type) {
// requires new texture to be at least as big as old texture. Assumes textures are 32bits per pixel
GLint width, height, layers;
glBindTexture(texture_type, old_texture);
glGetTexLevelParameteriv(texture_type, 0, GL_TEXTURE_WIDTH, &width);
glGetTexLevelParameteriv(texture_type, 0, GL_TEXTURE_HEIGHT, &height);
glGetTexLevelParameteriv(texture_type, 0, GL_TEXTURE_DEPTH, &layers);
if (GLAD_GL_ARB_copy_image) {
glCopyImageSubData(old_texture, texture_type, 0, 0, 0, 0, new_texture, texture_type, 0, 0, 0, 0, width, height, layers);
return;
}
if (copy_texture_via_framebuffer(old_texture, new_texture, texture_type, width, height, layers)) return;
static bool copy_image_warned = false;
// Neither ARB_copy_image nor a framebuffer copy is usable, do a slow roundtrip copy
if (!copy_image_warned) {
copy_image_warned = true;
log_error("WARNING: Your system's OpenGL implementation cannot copy between textures on the GPU, falling back to a slower implementation");
}
GLint internal_format;
glGetTexLevelParameteriv(texture_type, 0, GL_TEXTURE_INTERNAL_FORMAT, &internal_format);
GLenum format, type;
switch (internal_format) {
case GL_R8UI:
case GL_R8I:
case GL_R16UI:
case GL_R16I:
case GL_R32UI:
case GL_R32I:
case GL_RG8UI:
case GL_RG8I:
case GL_RG16UI:
case GL_RG16I:
case GL_RG32UI:
case GL_RG32I:
case GL_RGB8UI:
case GL_RGB8I:
case GL_RGB16UI:
case GL_RGB16I:
case GL_RGB32UI:
case GL_RGB32I:
case GL_RGBA8UI:
case GL_RGBA8I:
case GL_RGBA16UI:
case GL_RGBA16I:
case GL_RGBA32UI:
case GL_RGBA32I:
format = GL_RED_INTEGER;
type = GL_UNSIGNED_INT;
break;
default:
format = GL_RGBA;
type = GL_UNSIGNED_INT_8_8_8_8;
break;
}
glPixelStorei(GL_UNPACK_ALIGNMENT, 4);
RAII_ALLOC(uint8_t, pixels, malloc((size_t)width * height * layers * 4u));
if (!pixels) fatal("Out of memory");
glGetTexImage(texture_type, 0, format, type, pixels);
glBindTexture(texture_type, new_texture);
glPixelStorei(GL_PACK_ALIGNMENT, 4);
if (texture_type == GL_TEXTURE_2D_ARRAY) glTexSubImage3D(GL_TEXTURE_2D_ARRAY, 0, 0, 0, 0, width, height, layers, format, type, pixels);
else glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width, height, format, type, pixels);
}
static GLuint
setup_new_sprites_texture(GLenum texture_type) {
GLuint tex;
glGenTextures(1, &tex);
glBindTexture(texture_type, tex);
// We use GL_NEAREST otherwise glyphs that touch the edge of the cell
// often show a border between cells
glTexParameteri(texture_type, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glTexParameteri(texture_type, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(texture_type, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(texture_type, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
return tex;
}
static bool
decorations_map_new_size(size_t count, size_t current_capacity, size_t max_size, GLint *width, GLint *height) {
const size_t max_capacity = max_size * max_size;
if (count >= max_capacity) return false;
size_t capacity = MAX(MAX(count + 256, current_capacity * 2), (size_t)4096);
capacity = MIN(capacity, max_capacity);
if (capacity <= max_size) {
*width = capacity;
*height = 1;
} else {
*width = max_size;
*height = (capacity + max_size - 1) / max_size;
}
return true;
}
static void
realloc_sprite_decorations_texture_if_needed(FONTS_DATA_HANDLE fg) {
#define dm (sm->decorations_map)
SpriteMap *sm = (SpriteMap *)fg->sprite_map;
size_t current_capacity = (size_t)dm.width * dm.height;
if (dm.count < current_capacity && dm.texture_id) return;
GLint width, height;
if (!decorations_map_new_size(dm.count, current_capacity, sm->max_texture_size, &width, &height))
fatal("Max texture size too small for sprite decorations map, maybe switch to using a GL_TEXTURE_2D_ARRAY");
const GLenum texture_type = GL_TEXTURE_2D;
GLuint tex = setup_new_sprites_texture(texture_type);
glTexImage2D(texture_type, 0, GL_R32UI, width, height, 0, GL_RED_INTEGER, GL_UNSIGNED_INT, NULL);
if (dm.texture_id) { // copy data from old texture
copy_32bit_texture(dm.texture_id, tex, texture_type);
free_texture(&dm.texture_id);
}
glBindTexture(texture_type, 0);
dm.texture_id = tex;
dm.width = width;
dm.height = height;
#undef dm
}
static void
realloc_sprite_texture(FONTS_DATA_HANDLE fg) {
unsigned int xnum, ynum, z, znum, width, height;
sprite_tracker_current_layout(fg, &xnum, &ynum, &z);
znum = z + 1;
SpriteMap *sprite_map = (SpriteMap *)fg->sprite_map;
width = xnum * fg->fcm.cell_width;
height = ynum * (fg->fcm.cell_height + 1);
const GLenum texture_type = GL_TEXTURE_2D_ARRAY;
GLuint tex = setup_new_sprites_texture(texture_type);
glTexStorage3D(texture_type, 1, GL_SRGB8_ALPHA8, width, height, znum);
if (sprite_map->texture_id) { // copy old texture data into new texture
copy_32bit_texture(sprite_map->texture_id, tex, texture_type);
free_texture(&sprite_map->texture_id);
}
glBindTexture(texture_type, 0);
sprite_map->last_num_of_layers = znum;
sprite_map->last_ynum = ynum;
sprite_map->texture_id = tex;
}
static void
ensure_sprite_map(FONTS_DATA_HANDLE fg) {
SpriteMap *sprite_map = (SpriteMap *)fg->sprite_map;
if (!sprite_map->texture_id) realloc_sprite_texture(fg);
if (!sprite_map->decorations_map.texture_id) realloc_sprite_decorations_texture_if_needed(fg);
// We have to rebind since we don't know if the texture was ever bound
// in the context of the current OSWindow
glActiveTexture(GL_TEXTURE0 + SPRITE_DECORATIONS_MAP_UNIT);
glBindTexture(GL_TEXTURE_2D, sprite_map->decorations_map.texture_id);
glActiveTexture(GL_TEXTURE0 + SPRITE_MAP_UNIT);
glBindTexture(GL_TEXTURE_2D_ARRAY, sprite_map->texture_id);
}
void
send_sprite_to_gpu(FONTS_DATA_HANDLE fg, sprite_index idx, pixel *buf, sprite_index decoration_idx) {
SpriteMap *sprite_map = (SpriteMap *)fg->sprite_map;
unsigned int xnum, ynum, znum, x, y, z;
#define dm (sprite_map->decorations_map)
if (idx >= dm.count) dm.count = idx + 1;
realloc_sprite_decorations_texture_if_needed(fg);
div_t d = div(idx, dm.width);
x = d.rem;
y = d.quot;
glActiveTexture(GL_TEXTURE0 + SPRITE_DECORATIONS_MAP_UNIT);
glBindTexture(GL_TEXTURE_2D, dm.texture_id);
glPixelStorei(GL_UNPACK_ALIGNMENT, 4);
glTexSubImage2D(GL_TEXTURE_2D, 0, x, y, 1, 1, GL_RED_INTEGER, GL_UNSIGNED_INT, &decoration_idx);
#undef dm
sprite_tracker_current_layout(fg, &xnum, &ynum, &znum);
if ((int)znum >= sprite_map->last_num_of_layers || (znum == 0 && (int)ynum > sprite_map->last_ynum)) {
realloc_sprite_texture(fg);
sprite_tracker_current_layout(fg, &xnum, &ynum, &znum);
}
glActiveTexture(GL_TEXTURE0 + SPRITE_MAP_UNIT);
glBindTexture(GL_TEXTURE_2D_ARRAY, sprite_map->texture_id);
glPixelStorei(GL_UNPACK_ALIGNMENT, 4);
sprite_index_to_pos(idx, xnum, ynum, &x, &y, &z);
x *= fg->fcm.cell_width;
y *= (fg->fcm.cell_height + 1);
glTexSubImage3D(GL_TEXTURE_2D_ARRAY, 0, x, y, z, fg->fcm.cell_width, fg->fcm.cell_height + 1, 1, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8, buf);
}
void
send_image_to_gpu(GLuint *tex_id, const void *data, GLsizei width, GLsizei height, bool is_opaque, bool is_4byte_aligned, bool linear, RepeatStrategy repeat) {
if (!(*tex_id)) { glGenTextures(1, tex_id); }
glBindTexture(GL_TEXTURE_2D, *tex_id);
glPixelStorei(GL_UNPACK_ALIGNMENT, is_4byte_aligned ? 4 : 1);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, linear ? GL_LINEAR : GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, linear ? GL_LINEAR : GL_NEAREST);
RepeatStrategy r;
switch (repeat) {
case REPEAT_MIRROR: r = GL_MIRRORED_REPEAT; break;
case REPEAT_CLAMP: {
static const GLfloat border_color[4] = {0};
glTexParameterfv(GL_TEXTURE_2D, GL_TEXTURE_BORDER_COLOR, border_color);
r = GL_CLAMP_TO_BORDER;
break;
}
default: r = GL_REPEAT;
}
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, r);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, r);
glTexImage2D(GL_TEXTURE_2D, 0, GL_SRGB_ALPHA, width, height, 0, is_opaque ? GL_RGB : GL_RGBA, GL_UNSIGNED_BYTE, data);
}
// Indexed color images are stored as integer textures with the palette index
// in the red channel and, if bytes_per_pixel is 2, alpha in the green channel.
// Integer textures cannot be filtered, the shader does its own filtering after
// the palette lookup.
void
send_indexed_image_to_gpu(GLuint *tex_id, const void *data, GLsizei width, GLsizei height, unsigned bytes_per_pixel, bool is_4byte_aligned) {
if (!(*tex_id)) { glGenTextures(1, tex_id); }
glBindTexture(GL_TEXTURE_2D, *tex_id);
glPixelStorei(GL_UNPACK_ALIGNMENT, is_4byte_aligned ? 4 : 1);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
const bool has_alpha = bytes_per_pixel == 2;
glTexImage2D(GL_TEXTURE_2D, 0, has_alpha ? GL_RG8UI : GL_R8UI, width, height, 0, has_alpha ? GL_RG_INTEGER : GL_RED_INTEGER, GL_UNSIGNED_BYTE, data);
}
// }}}
// Rounded rect {{{
static double
thickness_as_float(const OSWindow *os_window, unsigned level) {
level = MIN(level, arraysz(OPT(box_drawing_scale)));
double pts = OPT(box_drawing_scale)[level];
double dpi = (os_window->fonts_data->logical_dpi_x + os_window->fonts_data->logical_dpi_y) / 2.0;
return line_thickness_scale_for_font_size(os_window->fonts_data) * pts * dpi / 72.0;
}
static void
draw_rounded_rect(
Viewport rect,
unsigned framebuffer_height,
float thickness,
unsigned corner_radius_px,
color_type srgb_color,
color_type srgb_background,
float bg_alpha,
color_type srgb_outer,
float outer_alpha) {
bind_program(ROUNDED_RECT_PROGRAM);
color_vec4(program_uniform_location(ROUNDED_RECT_PROGRAM, "color"), srgb_color, 1.f);
color_vec4(program_uniform_location(ROUNDED_RECT_PROGRAM, "background_color"), srgb_background, bg_alpha);
color_vec4(program_uniform_location(ROUNDED_RECT_PROGRAM, "outer_color"), srgb_outer, outer_alpha);
// y co-ord has to be changed to co-ord system with origin at bottom left
float y = (float)framebuffer_height - (float)(rect.top + rect.height);
glUniform4f(program_uniform_location(ROUNDED_RECT_PROGRAM, "rect"), rect.left, y, rect.width, rect.height);
glUniform2f(program_uniform_location(ROUNDED_RECT_PROGRAM, "params"), thickness, corner_radius_px);
save_viewport_using_top_left_origin(rect.left, rect.top, rect.width, rect.height, framebuffer_height);
draw_quad(true, 0);
restore_viewport();
}
// }}}
// Cell {{{
enum { CELL_RENDER_DATA_BINDING_POINT = 0, COLOR_TABLE_BINDING_POINT = 1, BORDER_COLORS_BINDING_POINT = 2, CUSTOM_END_DATA_BINDING_POINT = 3 };
enum { BORDER_COLORS_GLOBAL_BUFFER };
// VAOs used only to hold buffers for UBOs that are shared amongst programs/windows,
// their vertex attribute/array facilities are unused.
static ssize_t shader_globals_vao_idx = -1;
static ssize_t custom_end_vao_idx = -1;
typedef enum { CUSTOM_SHADER_TEXTURE_DEFAULT, CUSTOM_SHADER_TEXTURE_A, CUSTOM_SHADER_TEXTURE_B, CUSTOM_SHADER_TEXTURE_PERSIST } NamedTexture;
typedef struct CustomShaderGroup {
struct {
float x, y;
} viewport_pos;
struct {
float w, h;
} viewport_size;
NamedTexture output_texture;
ShaderAnimEventMask animation_start_events; // events that start the animation; 0 = none
ShaderAnimEventMask animation_toggle_events; // events that start the animation when stopped and vice versa; 0 = none
ShaderAnimEventMask animation_end_events; // events that stop the animation; 0 = none
monotonic_t animation_end_duration; // nanoseconds; 0 = no time limit, < 0 = use cursor_stop_blinking_after
monotonic_t animation_duration; // nanoseconds for animation_progress to reach one; < 0 = same as the stop duration
monotonic_t animation_reverse; // nanoseconds to play the animation in reverse when stopped; 0 = no, < 0 = animation_duration
unsigned animation_loops; // times to play the animation forwards and in reverse when reversing
monotonic_t animation_step; // nanoseconds between animation samples
Animation *animation_curve; // parsed easing curve; NULL = no animation
bool animation_curve_is_shared; // true if animation_curve is owned by an earlier group
bool attached; // true = always active when the preceding group is active
} CustomShaderGroup;
typedef struct CustomShaderPipeline {
unsigned textures; // bit mask of named textures from above enum
CustomShaderGroup groups[MAX_CUSTOM_SHADER_GROUPS];
size_t num_groups;
bool active;
bool bypass_builtin_trail; // true when any group subscribes to cursor-trail-move
ShaderAnimEventMask all_animation_start_events; // union of the start and toggle events of all groups
bool has_pointer_map; // true when a shader defines pointer_map() to move pointer events
uint32_t pointer_map_groups; // bitmask of the groups with shaders whose pointer_map() is used
// Names (without the user: prefix) of the user events the pipeline uses,
// user event i is bit SHADER_ANIM_FIRST_USER_EVENT + i
char user_events[MAX_SHADER_ANIM_USER_EVENTS][MAX_SHADER_ANIM_USER_EVENT_NAME_LEN + 1];
size_t num_user_events;
} CustomShaderPipeline;
static struct {
CustomShaderPipeline end;
size_t count;
unsigned generation; // incremented every time the end pipeline changes
} custom_shaders;
static struct {
GLuint texture_id, framebuffer_id;
} pointer_map_target;
static void
write_uint_array_to_ubo(void *dest, const GLuint *src, size_t count, const ArrayInformation *ai) {
GLint stride = MAX((GLint)sizeof(GLuint), ai->stride);
uint8_t *buf = (uint8_t *)dest + ai->offset;
for (size_t i = 0; i < count; i++, buf += stride) memcpy(buf, src + i, sizeof(GLuint));
}
// The gamma LUT is a compile time constant that lives in the default uniform
// block of every program that needs it, rather than in a UBO shared between
// them. Uniform values are part of program object state, which the GL
// implementation must maintain, while the data store of a buffer object is
// ours and some drivers silently discard it, notably macOS waking from sleep
// or reconfiguring its displays. Every color in every affected program then
// resolves to garbage with nothing to ever restore it. Uploading it costs a
// single GL call per program, so this can simply be called again whenever we
// have reason to believe GPU state was lost.
// See https://github.com/kovidgoyal/kitty/issues/10571
void
upload_gamma_lut(void) {
#define U(program) \
{ \
bind_program(program); \
GLint loc = program_uniform_location(program, "gamma_lut"); \
if (loc >= 0) glUniform1fv(loc, arraysz(srgb_lut), srgb_lut); \
}
for (int i = CELL_PROGRAM; i < CELL_PROGRAM_SENTINEL; i++) U(i);
U(BORDERS_PROGRAM);
U(PADDING_PROGRAM);
U(GRAPHICS_INDEXED_PROGRAM);
U(GRAPHICS_INDEXED_UNSCALED_PROGRAM);
#undef U
unbind_program();
}
static void
init_cell_program(void) {
for (int i = CELL_PROGRAM; i < CELL_PROGRAM_SENTINEL; i++) {
UniformBlock crd = program_uniform_block(i, "CellRenderData");
glUniformBlockBinding(program_id(i), crd.index, CELL_RENDER_DATA_BINDING_POINT);
UniformBlock ct = program_uniform_block(i, "ColorTable");
glUniformBlockBinding(program_id(i), ct.index, COLOR_TABLE_BINDING_POINT);
}
// Sanity check to ensure the attribute location binding worked
#define C(p, name, expected) \
{ \
int aloc = attrib_location(p, #name); \
if (aloc != expected && aloc != -1) fatal("The attribute location for %s is %d != %d in program: %d", #name, aloc, expected, p); \
}
for (int p = CELL_PROGRAM; p < CELL_PROGRAM_SENTINEL; p++) {
C(p, colors, 0);
C(p, sprite_idx, 1);
C(p, is_selected, 2);
C(p, decorations_sprite_map, 3);
}
#undef C
UniformBlock border_colors = program_uniform_block(BORDERS_PROGRAM, "Colors");
glUniformBlockBinding(program_id(BORDERS_PROGRAM), border_colors.index, BORDER_COLORS_BINDING_POINT);
// The padding program shares the cell background computation and so uses the
// same uniform blocks bound to the same binding points as the cell programs.
{
UniformBlock crd = program_uniform_block(PADDING_PROGRAM, "CellRenderData");
glUniformBlockBinding(program_id(PADDING_PROGRAM), crd.index, CELL_RENDER_DATA_BINDING_POINT);
UniformBlock ct = program_uniform_block(PADDING_PROGRAM, "ColorTable");
glUniformBlockBinding(program_id(PADDING_PROGRAM), ct.index, COLOR_TABLE_BINDING_POINT);
}
#define C(name, expected) \
{ \
int aloc = attrib_location(PADDING_PROGRAM, #name); \
if (aloc != expected && aloc != -1) fatal("The attribute location for %s is %d != %d in the padding program", #name, aloc, expected); \
}
C(colors, 0);
C(sprite_idx, 1);
C(is_selected, 2);
#undef C
// Indexed color images get their colors from the color table of the window they are in
for (int i = GRAPHICS_INDEXED_PROGRAM; i <= GRAPHICS_INDEXED_UNSCALED_PROGRAM; i++) {
UniformBlock ct = program_uniform_block(i, "ColorTable");
glUniformBlockBinding(program_id(i), ct.index, COLOR_TABLE_BINDING_POINT);
}
if (shader_globals_vao_idx == -1) {
shader_globals_vao_idx = create_vao();
add_buffer_to_vao(shader_globals_vao_idx, GL_UNIFORM_BUFFER);
// The border colors change on every draw, but are shared amongst all border VAOs (only one is ever drawn at a time)
alloc_vao_buffer(shader_globals_vao_idx, border_colors.size, BORDER_COLORS_GLOBAL_BUFFER, GL_STREAM_DRAW);
}
upload_gamma_lut();
bind_shader_globals_to_current_context();
}
// Whether the group has events of its own that turn it on, as opposed to
// running on every frame or only following its predecessor via attach
static inline bool
group_has_own_animation(const CustomShaderGroup *cg) {
return (cg->animation_start_events | cg->animation_toggle_events) != 0;
}
// Whether the group runs only when the animation state machine says it is active
static inline bool
group_is_animated(const CustomShaderGroup *cg) {
return cg->attached || group_has_own_animation(cg);
}
// How long after starting the group stops by itself, zero if it does not
static inline monotonic_t
group_stop_duration(const CustomShaderGroup *cg) {
return cg->animation_end_duration < 0 ? OPT(cursor_stop_blinking_after) : cg->animation_end_duration;
}
// How long animation_progress takes to rise to one, zero if it stays at zero
static inline monotonic_t
group_ramp_duration(const CustomShaderGroup *cg) {
return cg->animation_duration < 0 ? group_stop_duration(cg) : cg->animation_duration;
}
// How long animation_progress takes to fall from one to zero after the group
// is stopped, zero if the group turns off immediately instead
static inline monotonic_t
group_reverse_duration(const CustomShaderGroup *cg) {
return cg->animation_reverse < 0 ? group_ramp_duration(cg) : cg->animation_reverse;
}
// The animation_progress of an active group before applying the easing curve
static double
group_linear_progress(const CustomShaderGroup *cg, const ShaderGroupAnimState *s, monotonic_t now) {
double ans = 0;
if (s->reverse_ends_at) {
const monotonic_t d = group_reverse_duration(cg);
if (d > 0) ans = (double)(s->reverse_ends_at - now) / (double)d;
} else {
const monotonic_t d = group_ramp_duration(cg);
if (d > 0) ans = (double)(now - s->ramp_started_at) / (double)d;
}
return MAX(0., MIN(ans, 1.));
}
// The animation_progress of every group in the current pipeline. Attached
// groups without events of their own share the progress of the group they are
// attached to, so that multi-pass effects animate in sync.
static void
custom_shader_group_progress(const OSWindow *os_window, monotonic_t now, float *ans) {
for (size_t g = 0; g < custom_shaders.end.num_groups; g++) {
const CustomShaderGroup *cg = custom_shaders.end.groups + g;
const ShaderGroupAnimState *s = os_window->shader_group_anim + g;
ans[g] = 0.f;
if (!group_is_animated(cg) || !s->active) continue;
if (g > 0 && cg->attached && !group_has_own_animation(cg)) {
ans[g] = ans[g - 1];
continue;
}
const double t = group_linear_progress(cg, s, now);
ans[g] = (float)(cg->animation_curve ? apply_easing_curve(cg->animation_curve, t, group_ramp_duration(cg)) : t);
}
}
static void
start_group_animation(const CustomShaderGroup *cg, ShaderGroupAnimState *s, monotonic_t now) {
// Restarting while playing in reverse or looping continues from the
// current progress rather than jumping back to zero
if (s->active && (s->reverse_ends_at || s->looping)) {
s->ramp_started_at = now - (monotonic_t)(group_linear_progress(cg, s, now) * (double)group_ramp_duration(cg));
} else s->ramp_started_at = now;
s->active = true;
s->started_at = now;
s->reverse_ends_at = 0;
s->loops_left = 0;
s->looping = false;
}
static void
stop_group_animation(const CustomShaderGroup *cg, ShaderGroupAnimState *s, monotonic_t now) {
// Playing in reverse takes as long as falling from the current progress
// to zero, so a group stopped part way through rising does not jump to one
const monotonic_t reverse_for = (monotonic_t)(group_linear_progress(cg, s, now) * (double)group_reverse_duration(cg));
if (reverse_for > 0) {
s->reverse_ends_at = now + reverse_for;
s->loops_left = cg->animation_loops > 1 ? cg->animation_loops - 1 : 0;
} else s->active = false;
}
// Moves a group that is playing in reverse or looping on to whichever phase it
// should be in at now, turning it off once it has finished. The phases are
// advanced from the time the previous one ended rather than from now, so that
// loops do not drift when the state machine runs late.
static void
advance_stopped_group_animation(const CustomShaderGroup *cg, ShaderGroupAnimState *s, monotonic_t now, size_t i) {
while (true) {
if (s->reverse_ends_at && now >= s->reverse_ends_at) {
if (s->loops_left) {
s->loops_left--;
s->ramp_started_at = s->reverse_ends_at;
s->reverse_ends_at = 0;
s->looping = true;
} else {
s->active = false;
s->reverse_ends_at = 0;
debug_rendering("Custom shader group %zu finished playing in reverse\n", i);
break;
}
} else if (s->looping && now >= s->ramp_started_at + group_ramp_duration(cg)) {
// Loops only start after a reverse, which needs both durations to be non-zero, so this terminates
s->looping = false;
s->reverse_ends_at = s->ramp_started_at + group_ramp_duration(cg) + group_reverse_duration(cg);
} else break;
}
}
// Non-animated, non-attached groups are active from the moment the pipeline
// loads; animated groups start inactive and attached groups inherit the state
// of their predecessor.
static bool
group_is_initially_active(const CustomShaderGroup *cg, bool predecessor_is_active) {
return cg->attached ? predecessor_is_active : !group_has_own_animation(cg);
}
static bool
pipeline_has_initially_active_group(void) {
bool prev = false;
for (size_t i = 0; i < custom_shaders.end.num_groups; i++) {
prev = group_is_initially_active(custom_shaders.end.groups + i, prev);
if (prev) return true;
}
return false;
}
static void
init_custom_programs(void) {
custom_shaders.count = 0;
for (int i = CUSTOM_END_PROGRAM; i < NUM_PROGRAMS; i++) {
Program *p = program_ptr(i);
if (p && p->id) {
custom_shaders.count++;
if (i == CUSTOM_END_PROGRAM) {
bind_program(i);
glUniform1i(program_uniform_location(i, "backbuffer"), GRAPHICS_UNIT);
#define set_optional_sampler(name, unit) \
{ \
GLint loc = try_program_uniform_location(i, name); \
if (loc >= 0) glUniform1i(loc, unit); \
}
set_optional_sampler("a", CUSTOM_END_TEXTURE_A_UNIT);
set_optional_sampler("b", CUSTOM_END_TEXTURE_B_UNIT);
set_optional_sampler("persist", CUSTOM_END_TEXTURE_PERSIST_UNIT);
#undef set_optional_sampler
GLint group_loc = try_program_uniform_location(i, "group");
if (group_loc >= 0) glUniform1i(group_loc, 0);
else if (custom_shaders.end.num_groups > 1) {
// Every group is dispatched by setting this uniform, so without it
// all of them render as group zero, which for a multi pass pipeline
// means the intermediate passes are drawn to the screen.
log_error(
"The custom shader pipeline has %zu groups but its group uniform was optimized away by the"
" GPU driver. Every group will render as group zero.",
custom_shaders.end.num_groups);
}
UniformBlock ubd = program_uniform_block(i, "KittyCustomShaderData");
glUniformBlockBinding(program_id(i), ubd.index, CUSTOM_END_DATA_BINDING_POINT);
if (custom_end_vao_idx == -1) {
custom_end_vao_idx = create_vao();
add_buffer_to_vao(custom_end_vao_idx, GL_UNIFORM_BUFFER);
alloc_vao_buffer(custom_end_vao_idx, ubd.size, 0, GL_STREAM_DRAW);
}
}
}
}
// Free global named textures that are no longer declared in the pipeline
const bool drop_a = !(custom_shaders.end.textures & (1u << CUSTOM_SHADER_TEXTURE_A));
const bool drop_b = !(custom_shaders.end.textures & (1u << CUSTOM_SHADER_TEXTURE_B));
if (drop_a) {
if (global_state.layers_render_texture.texture_a_id) free_texture(&global_state.layers_render_texture.texture_a_id);
if (global_state.layers_render_texture.texture_a_fbo_id) free_framebuffer(&global_state.layers_render_texture.texture_a_fbo_id);
}
if (drop_b) {
if (global_state.layers_render_texture.texture_b_id) free_texture(&global_state.layers_render_texture.texture_b_id);
if (global_state.layers_render_texture.texture_b_fbo_id) free_framebuffer(&global_state.layers_render_texture.texture_b_fbo_id);
}
// persist is per-window and freed lazily in start_os_window_rendering
for (size_t w = 0; w < global_state.num_os_windows; w++) {
OSWindow *osw = global_state.os_windows + w;
// Per window framebuffers must be attached afresh to the textures that
// replace the ones freed above, should a later pipeline declare them again
if (drop_a) osw->indirect_output.fbo_a_generation = 0;
if (drop_b) osw->indirect_output.fbo_b_generation = 0;
init_shader_animation_state(osw);
// Synthesize focus events for windows that already have focus so that
// animations triggered by os-window-focus-in start on the first render
// after shader compilation (the focus-in event fired before shaders were
// compiled and was consumed without being processed).
if (osw->is_focused) record_os_window_focus_change_for_shaders(osw, true);
}
debug_rendering(
"Custom shaders: %zu program(s), %zu group(s), textures_mask=0x%x, initially_active=%d\n",
custom_shaders.count,
custom_shaders.end.num_groups,
custom_shaders.end.textures,
custom_shaders.count > 0 && pipeline_has_initially_active_group());
}
// Reset the per-OS-window animation state to what it should be immediately
// after the current pipeline was loaded. Called both when the pipeline changes
// and when an OS window is created, so that the MONOTONIC_T_MAX sentinels in
// ShaderAnimState are never left at their zero initialized values -- a zero
// min_step/next_end_at reads as "an animation is running and its deadline has
// already passed", which would make the renderer redraw on every single tick.
void
init_shader_animation_state(OSWindow *osw) {
zero_at_ptr_count(osw->shader_group_anim, custom_shaders.end.num_groups);
bool prev_init = false, any_initially_active = false;
for (size_t i = 0; i < custom_shaders.end.num_groups; i++) {
const CustomShaderGroup *cg = &custom_shaders.end.groups[i];
const bool this_init = group_is_initially_active(cg, prev_init);
// Only attached groups are tracked by the state machine when initially
// active; always-active non-attached groups never enter it at all.
if (cg->attached && this_init) osw->shader_group_anim[i].active = true;
if (this_init) any_initially_active = true;
prev_init = this_init;
}
osw->shader_anim.has_active_shaders = custom_shaders.count > 0 && any_initially_active;
osw->shader_anim.min_step = MONOTONIC_T_MAX;
osw->shader_anim.next_end_at = MONOTONIC_T_MAX;
}
static const char *
shader_anim_event_mask_str(ShaderAnimEventMask mask) {
static char buf[256];
char *p = buf;
char *const end = buf + sizeof(buf) - 1;
static const struct {
ShaderAnimEventMask bit;
const char *name;
} events[] = {
{1u << SHADER_ANIM_EVENT_POINTER_LEFT_BUTTON_PRESS, "pointer-left-button-press"},
{1u << SHADER_ANIM_EVENT_OS_WINDOW_FOCUS_IN, "os-window-focus-in"},
{1u << SHADER_ANIM_EVENT_OS_WINDOW_FOCUS_OUT, "os-window-focus-out"},
{1u << SHADER_ANIM_EVENT_WINDOW_FOCUS_IN, "window-focus-in"},
{1u << SHADER_ANIM_EVENT_WINDOW_FOCUS_OUT, "window-focus-out"},
{1u << SHADER_ANIM_EVENT_TAB_CHANGE, "tab-change"},
{1u << SHADER_ANIM_EVENT_BELL_IN_WINDOW, "bell-in-window"},
{1u << SHADER_ANIM_EVENT_USER_ACTIVITY, "user-activity"},
{1u << SHADER_ANIM_EVENT_USER_IDLE, "user-idle"},
{1u << SHADER_ANIM_EVENT_CURSOR_TRAIL_MOVE, "cursor-trail-move"},
{1u << SHADER_ANIM_EVENT_CURSOR_TRAIL_STOP, "cursor-trail-stop"},
{1u << SHADER_ANIM_EVENT_COMMAND_START, "command-start"},
{1u << SHADER_ANIM_EVENT_COMMAND_FINISH, "command-finish"},
{1u << SHADER_ANIM_EVENT_ALTERNATE_SCREEN_ENTER, "alternate-screen-enter"},
{1u << SHADER_ANIM_EVENT_ALTERNATE_SCREEN_EXIT, "alternate-screen-exit"},
};
bool first = true;
for (size_t i = 0; i < sizeof(events) / sizeof(events[0]); i++) {
if (!(mask & events[i].bit)) continue;
if (!first && p < end) *p++ = '|';
size_t len = strlen(events[i].name);
if (p + len >= end) break;
memcpy(p, events[i].name, len);
p += len;
first = false;
}
for (size_t i = 0; i < custom_shaders.end.num_user_events; i++) {
if (!(mask & (1ull << (SHADER_ANIM_FIRST_USER_EVENT + i)))) continue;
if (!first && p < end) *p++ = '|';
int n = snprintf(p, end - p + 1, "user:%s", custom_shaders.end.user_events[i]);
if (n < 0 || p + n >= end) break;
p += n;
first = false;
}
if (first) {
memcpy(buf, "(none)", 7);
} else {
*p = '\0';
}
return buf;
}
monotonic_t
update_custom_shader_animations(ShaderAnimEventMask event_mask, monotonic_t now, OSWindow *os_window) {
if (!custom_shaders.count) {
os_window->shader_anim = (ShaderAnimState){.min_step = MONOTONIC_T_MAX, .next_end_at = MONOTONIC_T_MAX};
return MONOTONIC_T_MAX;
}
// Fast path: no events and no duration-bounded animation expiring this frame.
// Cached state is still valid — skip group iteration entirely.
if (!event_mask && now < os_window->shader_anim.next_end_at) return os_window->shader_anim.min_step;
// Slow path: update per-group state and recompute cached values in one pass.
CustomShaderPipeline *p = &custom_shaders.end;
bool any_active = false;
monotonic_t min_step = MONOTONIC_T_MAX;
monotonic_t next_end = MONOTONIC_T_MAX;
bool prev_active = false;
for (size_t i = 0; i < p->num_groups; i++) {
const CustomShaderGroup *cg = &p->groups[i];
bool this_active;
if (!group_is_animated(cg)) {
// Always-active, non-attached group — never enters the state machine.
any_active = true;
this_active = true;
min_step = MIN(min_step, cg->animation_step);
} else {
const monotonic_t eff_dur = group_stop_duration(cg);
ShaderGroupAnimState *s = os_window->shader_group_anim + i;
if (group_has_own_animation(cg)) {
advance_stopped_group_animation(cg, s, now, i);
// Whether the group is on, as opposed to off or playing in reverse or looping after being stopped
const bool was_on = s->active && !s->reverse_ends_at && !s->looping;
// Focus-out events are hard stops: if a focus-out stop condition is present in
// the same tick as a start event (e.g. user-activity fired by the click that
// moved focus away), the focus-out wins and the animation is not restarted.
const ShaderAnimEventMask focus_out_mask = (1u << SHADER_ANIM_EVENT_OS_WINDOW_FOCUS_OUT) | (1u << SHADER_ANIM_EVENT_WINDOW_FOCUS_OUT);
const bool hard_stop = (cg->animation_end_events & event_mask & focus_out_mask) != 0;
// A simultaneous start and end lets start win, except for focus-out
// events which act as hard stops.
bool turn_on = false, turn_off = false;
if (was_on) {
turn_off = (cg->animation_end_events && (event_mask & cg->animation_end_events)) || (eff_dur > 0 && now - s->started_at >= eff_dur);
}
if ((event_mask & cg->animation_start_events) && !hard_stop) turn_on = true;
// A toggle flips the state the group had before this tick and so
// takes precedence over simultaneous start and stop events.
if (event_mask & cg->animation_toggle_events) {
turn_on = !was_on && !hard_stop;
turn_off = was_on;
}
if (turn_on) {
start_group_animation(cg, s, now);
debug_rendering("Custom shader group %zu animation started (events=%s)\n", i, shader_anim_event_mask_str(event_mask));
} else if (turn_off) {
stop_group_animation(cg, s, now);
debug_rendering(
"Custom shader group %zu animation stopped%s (events=%s)\n",
i,
s->active ? ", playing in reverse" : "",
shader_anim_event_mask_str(event_mask));
}
}
// Attach: follow the previous group's effective active state.
if (cg->attached) {
if (prev_active && !s->active) {
s->active = true;
// Sync animation timing with the previous group when possible.
monotonic_t prev_started = (i > 0) ? os_window->shader_group_anim[i - 1].started_at : 0;
s->started_at = prev_started ? prev_started : now;
s->ramp_started_at = s->started_at;
s->reverse_ends_at = 0;
s->loops_left = 0;
s->looping = false;
debug_rendering("Custom shader group %zu activated via attach\n", i);
} else if (!prev_active && !group_has_own_animation(cg) && s->active) {
// No own events: mirror the previous group's deactivation.
s->active = false;
}
}
this_active = s->active;
if (this_active) {
any_active = true;
min_step = MIN(min_step, cg->animation_step);
// Only track next_end for groups with own stop conditions; attached-only groups
// deactivate when their predecessor does (already tracked by the predecessor).
if (group_has_own_animation(cg)) {
if (s->reverse_ends_at) next_end = MIN(next_end, s->reverse_ends_at);
else if (s->looping) next_end = MIN(next_end, s->ramp_started_at + group_ramp_duration(cg));
else if (eff_dur > 0) next_end = MIN(next_end, s->started_at + eff_dur);
}
}
}
prev_active = this_active;
}
bool prev_has_active = os_window->shader_anim.has_active_shaders;
if (prev_has_active != any_active) debug_rendering("has_active_shaders: %d -> %d\n", prev_has_active, any_active);
os_window->shader_anim.has_active_shaders = any_active;
os_window->shader_anim.min_step = min_step;
os_window->shader_anim.next_end_at = next_end;
return min_step;
}
// The bit for the user event with the specified name (without the user:
// prefix) in the current pipeline, zero if the pipeline does not use it
static ShaderAnimEventMask
user_event_bit(const char *name) {
for (size_t i = 0; i < custom_shaders.end.num_user_events; i++) {
if (strcmp(custom_shaders.end.user_events[i], name) == 0) return 1ull << (SHADER_ANIM_FIRST_USER_EVENT + i);
}
return 0;
}
// Queue the user event with the specified name for the OS window with the
// specified id, or for all OS windows when the id is zero. Returns false when
// the current pipeline does not use the event.
static bool
fire_custom_shader_user_event(const char *name, id_type os_window_id) {
const ShaderAnimEventMask bit = user_event_bit(name);
if (!bit) {
debug_rendering("Ignoring the custom shader event user:%s as the current pipeline does not use it\n", name);
return false;
}
bool queued = false;
for (size_t i = 0; i < global_state.num_os_windows; i++) {
OSWindow *osw = global_state.os_windows + i;
if (os_window_id && osw->id != os_window_id) continue;
osw->shader_anim_event_registry |= bit;
queued = true;
}
if (queued) wakeup_main_loop();
return true;
}
// Decide whether the custom shader layer needs a frame drawn this tick, given
// the cached state from the previous tick and the state just recomputed by
// update_custom_shader_animations(). Kept separate from the renderer so it can
// be unit tested -- see test_custom_shader_redraw_logic.
bool
custom_shader_needs_render(const ShaderAnimState *before, const ShaderAnimState *after, ShaderAnimEventMask event_mask, monotonic_t now) {
// Groups turned on or off: draw once so the change becomes visible. This is
// what produces the final frame that clears a finished effect.
if (before->has_active_shaders != after->has_active_shaders) return true;
// A duration bounded animation expires on this tick.
if (now >= before->next_end_at) return true;
// Something was animating last tick, so it either wants its next frame or
// has just stopped and wants one last one.
if (before->min_step < MONOTONIC_T_MAX) return true;
if (!after->has_active_shaders) return false;
// Animated groups need periodic frames. Static groups (animation_step 0)
// only need one when an event may have changed what they draw; everything
// else that changes the screen requests a redraw through other paths.
return event_mask != 0 || after->min_step < MONOTONIC_T_MAX;
}
void
bind_shader_globals_to_current_context(void) {
if (shader_globals_vao_idx == -1) return;
bind_vao_uniform_buffer(shader_globals_vao_idx, BORDER_COLORS_GLOBAL_BUFFER, BORDER_COLORS_BINDING_POINT);
}
#define CELL_BUFFERS enum { cell_data_buffer, selection_buffer, uniform_buffer, color_table_buffer, padding_cell_data_buffer, padding_selection_buffer };
ssize_t
create_cell_vao(void) {
ssize_t vao_idx = create_vao();
#define A(name, size, dtype, offset, stride) \
add_attribute_to_vao( \
vao_idx, program_attribute_location(CELL_PROGRAM, name), /*size=*/size, /*dtype=*/dtype, /*stride=*/stride, /*offset=*/offset, /*divisor=*/1);
#define A1(name, size, dtype, offset) A(name, size, dtype, (void *)(offsetof(GPUCell, offset)), sizeof(GPUCell))
add_buffer_to_vao(vao_idx, GL_ARRAY_BUFFER);
A1("sprite_idx", 2, GL_UNSIGNED_INT, sprite_idx);
A1("colors", 3, GL_UNSIGNED_INT, fg);
add_buffer_to_vao(vao_idx, GL_ARRAY_BUFFER);
A("is_selected", 1, GL_UNSIGNED_BYTE, NULL, 0);
size_t bufnum = add_buffer_to_vao(vao_idx, GL_UNIFORM_BUFFER);
alloc_vao_buffer(vao_idx, program_uniform_block(CELL_PROGRAM, "CellRenderData").size, bufnum, GL_STREAM_DRAW);
size_t ctbufnum = add_buffer_to_vao(vao_idx, GL_UNIFORM_BUFFER);
alloc_vao_buffer(vao_idx, program_uniform_block(CELL_PROGRAM, "ColorTable").size, ctbufnum, GL_STATIC_DRAW);
add_buffer_to_vao(vao_idx, GL_ARRAY_BUFFER); // padding_cell_data_buffer (slot 4)
add_buffer_to_vao(vao_idx, GL_ARRAY_BUFFER); // padding_selection_buffer (slot 5)
return vao_idx;
#undef A
#undef A1
}
#define IS_SPECIAL_COLOR(name) \
(screen->color_profile->overridden.name.type == COLOR_IS_SPECIAL || \
(screen->color_profile->overridden.name.type == COLOR_NOT_SET && screen->color_profile->configured.name.type == COLOR_IS_SPECIAL))
static float
cursor_color_luminance(color_type color) {
// Match color_to_vec() and contrast_ratio() in background.slang.
return 0.2126f * srgb_lut[(color >> 16) & 0xff] + 0.7152f * srgb_lut[(color >> 8) & 0xff] + 0.0722f * srgb_lut[color & 0xff];
}
static float
cursor_color_contrast(color_type a, color_type b) {
const float la = cursor_color_luminance(a), lb = cursor_color_luminance(b);
return (MAX(la, lb) + 0.05f) / (MIN(la, lb) + 0.05f);
}
static void
pick_cursor_color(color_type cell_fg, color_type cell_bg, color_type *cursor_fg, color_type *cursor_bg, color_type default_fg, color_type default_bg) {
*cursor_fg = cell_bg;
*cursor_bg = cell_fg;
const float cell_contrast = cursor_color_contrast(cell_fg, cell_bg);
if (cell_contrast < 2.5) {
if (cursor_color_contrast(default_fg, default_bg) > cell_contrast) {
*cursor_fg = default_bg;
*cursor_bg = default_fg;
}
}
}
static uint8_t
dimmed_color_component(uint8_t fg, uint8_t bg, float opacity) {
// DIM blends text with the cell background in linear color space.
const float linear = (1.f - opacity) * srgb_lut[bg] + opacity * srgb_lut[fg];
const float srgb = linear <= 0.0031308f ? 12.92f * linear : 1.055f * powf(linear, 1.f / 2.4f) - 0.055f;
return (uint8_t)lroundf(255.f * srgb);
}
static color_type
dimmed_cursor_color(color_type fg, color_type bg, float opacity) {
if (opacity >= 1.f) return fg;
return ((color_type)dimmed_color_component((fg >> 16) & 0xff, (bg >> 16) & 0xff, opacity) << 16) |
((color_type)dimmed_color_component((fg >> 8) & 0xff, (bg >> 8) & 0xff, opacity) << 8) |
(color_type)dimmed_color_component(fg & 0xff, bg & 0xff, opacity);
}
static bool
has_bgimage(OSWindow *w) {
return background_image_for_os_window(w) != NULL;
}
// The value resolve_dynamic_color() in background.slang needs for the color
// types that are an actual color. The other two types resolve to colors that
// are only known per-cell, so the shader ignores this value for them.
static color_type
dynamic_color_for_gpu(const ColorProfile *cp, color_type val) {
const DynamicColor dc = {.val = val}, zero = {0};
if (dc.type != COLOR_IS_RGB && dc.type != COLOR_IS_INDEX) return 0;
return colorprofile_to_color(cp, dc, zero).rgb;
}
static color_type
cell_update_uniform_block(
ssize_t vao_idx,
Screen *screen,
int uniform_buffer,
int color_table_buf,
const CursorRenderInfo *cursor,
OSWindow *os_window,
float inactive_text_alpha,
float bg_alpha) {
struct GPUCellRenderData {
// Colors that are constant for the whole draw, pre-converted to linear
// space so the vertex shader does not do a gamma LUT lookup for them
// once per vertex. These must come first and the fourth component of
// each is unused padding, see CellRenderData in background.slang.
GLfloat default_fg_lin[4], default_bg_lin[4], highlight_fg_lin[4], highlight_bg_lin[4];
GLfloat main_cursor_fg_lin[4], main_cursor_bg_lin[4], url_color_lin[4];
GLfloat extra_cursor_fg_lin[4], extra_cursor_bg_lin[4];
GLfloat use_cell_bg_for_selection_fg, use_cell_fg_for_selection_color, use_cell_for_selection_bg;
GLuint default_fg, url_style, inverted, main_cursor_is_special, extra_cursor_fg, extra_cursor_bg;
GLuint columns, lines, sprites_xnum, sprites_ynum, cursor_shape, cell_width, cell_height;
GLuint cursor_x1, cursor_x2, cursor_y1, cursor_y2;
GLfloat cursor_opacity, inactive_text_alpha, fg_override_threshold, row_offset, dim_opacity, blink_opacity;
GLuint bg_colors0, bg_colors1, bg_colors2, bg_colors3, bg_colors4, bg_colors5, bg_colors6, bg_colors7;
GLfloat bg_opacities0, bg_opacities1, bg_opacities2, bg_opacities3, bg_opacities4, bg_opacities5, bg_opacities6, bg_opacities7;
};
// Send the uniform data
ColorProfile *cp = screen->paused_rendering.expires_at ? &screen->paused_rendering.color_profile : screen->color_profile;
if (cp->dirty || screen->reload_all_gpu_data) {
const ArrayInformation ai = program_uniform_array(CELL_PROGRAM, "color_table");
uint8_t *base = (uint8_t *)map_vao_buffer_for_write_only(vao_idx, color_table_buf, 0, program_uniform_block(CELL_PROGRAM, "ColorTable").size);
GLuint *ct_buf = (GLuint *)(base + ai.offset);
copy_color_table_to_buffer(cp, ct_buf, 0, ai.stride / sizeof(GLuint));
unmap_vao_buffer(vao_idx, color_table_buf);
}
// This struct is a hand written mirror of the std140 layout of
// CellRenderData in background.slang, a mismatch silently corrupts every
// color, so check it once.
const UniformBlock crd_block = program_uniform_block(CELL_PROGRAM, "CellRenderData");
static bool crd_block_size_checked = false;
if (!crd_block_size_checked) {
crd_block_size_checked = true;
if ((size_t)crd_block.size < sizeof(struct GPUCellRenderData))
fatal("The CellRenderData uniform block is %d bytes but the C struct mirroring it is %zu bytes", crd_block.size, sizeof(struct GPUCellRenderData));
}
struct GPUCellRenderData *rd = (struct GPUCellRenderData *)map_vao_buffer_for_write_only(vao_idx, uniform_buffer, 0, crd_block.size);
#define COLOR(name) colorprofile_to_color(cp, cp->overridden.name, cp->configured.name).rgb
// The shaders need these only in linear space, they are converted and
// written into rd just before the buffer is unmapped. The cursor colors
// are left as is when there is no cursor to render.
const color_type highlight_fg = COLOR(highlight_fg), highlight_bg = COLOR(highlight_bg);
color_type main_cursor_fg = 0, main_cursor_bg = 0;
rd->default_fg = COLOR(default_fg);
rd->extra_cursor_fg = screen->extra_cursors.color.text.val;
rd->extra_cursor_bg = screen->extra_cursors.color.cursor.val;
rd->main_cursor_is_special = IS_SPECIAL_COLOR(cursor_color);
rd->bg_colors0 = COLOR(default_bg);
rd->bg_opacities0 = bg_alpha;
rd->fg_override_threshold = OPT(text_fg_override_threshold);
rd->row_offset = row_offset_for_screen(screen);
#define SETBG(which) \
{ colorprofile_to_transparent_color(cp, which - 1, &rd->bg_colors##which, &rd->bg_opacities##which); }
SETBG(1);
SETBG(2);
SETBG(3);
SETBG(4);
SETBG(5);
SETBG(6);
SETBG(7);
#undef SETBG
// selection
if (IS_SPECIAL_COLOR(highlight_fg)) {
if (IS_SPECIAL_COLOR(highlight_bg)) {
rd->use_cell_bg_for_selection_fg = 1.f;
rd->use_cell_fg_for_selection_color = 0.f;
} else {
rd->use_cell_bg_for_selection_fg = 0.f;
rd->use_cell_fg_for_selection_color = 1.f;
}
} else {
rd->use_cell_bg_for_selection_fg = 0.f;
rd->use_cell_fg_for_selection_color = 0.f;
}
rd->use_cell_for_selection_bg = IS_SPECIAL_COLOR(highlight_bg) ? 1. : 0.;
// Cursor position
Line *line_for_cursor = NULL, overlay_line_for_cursor;
rd->cursor_opacity = MAX(0, MIN(cursor->cursor_opacity, 1));
rd->blink_opacity = MAX(0, MIN(cursor->text_blink_opacity, 1));
// The extra cursors inherit the colors of the main cursor and, optionally,
// its shape, so the main cursor render data must be computed even when the
// main cursor itself is hidden by DECTCEM.
if (rd->cursor_opacity != 0 && (cursor->is_visible || cursor->multicursor_count)) {
rd->cursor_x1 = cursor->x, rd->cursor_y1 = cursor->y;
rd->cursor_x2 = cursor->x, rd->cursor_y2 = cursor->y;
if (pixel_scroll_enabled(screen)) {
rd->cursor_y1 += 1;
rd->cursor_y2 += 1;
}
CursorShape cs = (cursor->is_focused || OPT(cursor_shape_unfocused) == NO_CURSOR_SHAPE) ? cursor->shape : OPT(cursor_shape_unfocused);
rd->cursor_shape = cs;
color_type cell_fg = rd->default_fg, cell_bg = rd->bg_colors0;
index_type cell_color_x = cursor->x;
bool reversed = false;
if (cursor->x < screen->columns && cursor->y < screen->lines) {
if (screen->paused_rendering.expires_at) {
linebuf_init_line(screen->paused_rendering.linebuf, cursor->y);
line_for_cursor = screen->paused_rendering.linebuf->line;
} else {
linebuf_init_line(screen->linebuf, cursor->y);
line_for_cursor = screen->linebuf->line;
if (screen_is_overlay_active(screen) && cursor->y == screen->overlay_line.ynum) {
// The linebuf holds the original line, the pre-edit text is
// drawn over it only in the overlay cells, so use those to size
// and color the cursor, for example when it is over a wide character.
overlay_line_for_cursor = *line_for_cursor;
overlay_line_for_cursor.cpu_cells = screen->overlay_line.cpu_cells;
overlay_line_for_cursor.gpu_cells = screen->overlay_line.gpu_cells;
line_for_cursor = &overlay_line_for_cursor;
}
}
}
if (line_for_cursor) {
colors_for_cell(line_for_cursor, cp, &cell_color_x, &cell_fg, &cell_bg, &reversed);
const CPUCell *cursor_cell;
const bool large_cursor = ((cursor_cell = &line_for_cursor->cpu_cells[cursor->x])->is_multicell) && cursor_cell->x == 0 && cursor_cell->y == 0;
if (large_cursor) {
switch (cs) {
case CURSOR_BEAM: rd->cursor_y2 += cursor_cell->scale - 1; break;
case CURSOR_UNDERLINE:
rd->cursor_y1 += cursor_cell->scale - 1;
rd->cursor_y2 = rd->cursor_y1;
rd->cursor_x2 += mcd_x_limit(cursor_cell) - 1;
break;
case CURSOR_BLOCK:
rd->cursor_y2 += cursor_cell->scale - 1;
rd->cursor_x2 += mcd_x_limit(cursor_cell) - 1;
break;
case CURSOR_HOLLOW:
case NUM_OF_CURSOR_SHAPES:
case NO_CURSOR_SHAPE: break;
};
}
}
// If you change the following algorithm remember to change it in the cell shader for extra cursors too
if (rd->main_cursor_is_special) {
if (line_for_cursor && line_for_cursor->gpu_cells[cell_color_x].attrs.dim) {
cell_fg = dimmed_cursor_color(cell_fg, cell_bg, OPT(dim_opacity));
pick_cursor_color(cell_fg, cell_bg, &main_cursor_fg, &main_cursor_bg, rd->default_fg, rd->bg_colors0);
} else if (cell_bg == cell_fg) {
main_cursor_fg = rd->bg_colors0;
main_cursor_bg = rd->default_fg;
} else {
main_cursor_fg = cell_bg;
main_cursor_bg = cell_fg;
}
} else {
main_cursor_bg = COLOR(cursor_color);
if (IS_SPECIAL_COLOR(cursor_text_color)) main_cursor_fg = cell_bg;
else main_cursor_fg = COLOR(cursor_text_color);
}
// store last rendered cursor color for trail rendering
screen->last_rendered.cursor_bg = main_cursor_bg;
if (!cursor->is_visible) {
// Move the main cursor off screen so that only it is hidden, the
// extra cursors keep their shape and colors.
rd->cursor_x1 = screen->columns + 1;
rd->cursor_x2 = screen->columns;
rd->cursor_y1 = screen->lines + 1;
rd->cursor_y2 = screen->lines;
}
} else {
rd->cursor_shape = 0;
rd->cursor_x1 = screen->columns + 1;
rd->cursor_x2 = screen->columns;
rd->cursor_y1 = screen->lines + 1;
rd->cursor_y2 = screen->lines;
}
rd->columns = screen->columns;
rd->lines = screen->lines;
unsigned int x, y, z;
sprite_tracker_current_layout(os_window->fonts_data, &x, &y, &z);
rd->sprites_xnum = x;
rd->sprites_ynum = y;
rd->inverted = screen_invert_colors(screen) ? 1 : 0;
rd->cell_width = os_window->fonts_data->fcm.cell_width;
rd->cell_height = os_window->fonts_data->fcm.cell_height;
rd->inactive_text_alpha = inactive_text_alpha;
rd->dim_opacity = OPT(dim_opacity);
#undef COLOR
rd->url_style = OPT(url_style);
linear_color_vec4(rd->default_fg_lin, rd->default_fg);
linear_color_vec4(rd->default_bg_lin, rd->bg_colors0);
linear_color_vec4(rd->highlight_fg_lin, highlight_fg);
linear_color_vec4(rd->highlight_bg_lin, highlight_bg);
linear_color_vec4(rd->main_cursor_fg_lin, main_cursor_fg);
linear_color_vec4(rd->main_cursor_bg_lin, main_cursor_bg);
linear_color_vec4(rd->url_color_lin, OPT(url_color));
linear_color_vec4(rd->extra_cursor_fg_lin, dynamic_color_for_gpu(cp, rd->extra_cursor_fg));
linear_color_vec4(rd->extra_cursor_bg_lin, dynamic_color_for_gpu(cp, rd->extra_cursor_bg));
color_type default_bg = rd->bg_colors0;
unmap_vao_buffer(vao_idx, uniform_buffer);
rd = NULL;
return default_bg;
}
static bool
cell_prepare_to_render(ssize_t vao_idx, Screen *screen, FONTS_DATA_HANDLE fonts_data) {
size_t sz;
CELL_BUFFERS;
void *address;
bool changed = false;
ensure_sprite_map(fonts_data);
const Cursor *cursor = screen->paused_rendering.expires_at ? &screen->paused_rendering.cursor : screen->cursor;
bool cursor_pos_changed = cursor->x != screen->last_rendered.cursor.x || cursor->y != screen->last_rendered.cursor.y;
bool disable_ligatures = screen->disable_ligatures == DISABLE_LIGATURES_CURSOR;
bool screen_resized = screen->last_rendered.columns != screen->columns || screen->last_rendered.lines != screen->lines;
#define update_cell_data \
{ \
const unsigned int render_lines = render_lines_for_screen(screen); \
sz = sizeof(GPUCell) * render_lines * screen->columns; \
address = alloc_and_map_vao_buffer_for_full_rewrite(vao_idx, sz, cell_data_buffer); \
screen_update_cell_data(screen, address, fonts_data, disable_ligatures && cursor_pos_changed); \
unmap_vao_buffer(vao_idx, cell_data_buffer); \
address = NULL; \
changed = true; \
}
if (screen->paused_rendering.expires_at) {
if (!screen->paused_rendering.cell_data_updated) update_cell_data;
} else if (screen->reload_all_gpu_data || screen->scroll_changed || screen->is_dirty || screen_resized || (disable_ligatures && cursor_pos_changed))
update_cell_data;
#define update_selection_data \
{ \
const unsigned int render_lines = render_lines_for_screen(screen); \
sz = (size_t)render_lines * screen->columns; \
address = alloc_and_map_vao_buffer_for_full_rewrite(vao_idx, sz, selection_buffer); \
screen_apply_selection(screen, address, sz); \
unmap_vao_buffer(vao_idx, selection_buffer); \
address = NULL; \
changed = true; \
}
if (screen->paused_rendering.expires_at) {
if (!screen->paused_rendering.cell_data_updated) {
update_selection_data;
screen_update_graphics_layers(screen);
}
screen->paused_rendering.cell_data_updated = true;
} else {
if (screen->reload_all_gpu_data || screen_resized || screen_is_selection_dirty(screen)) update_selection_data;
if (screen_update_graphics_layers(screen)) changed = true;
}
screen->last_rendered.scrolled_by = rendered_scrolled_by(screen);
#undef update_selection_data
#undef update_cell_data
screen->last_rendered.columns = screen->columns;
screen->last_rendered.lines = screen->lines;
screen->last_rendered.cursor = screen->cursor_render_info;
return changed;
}
static void
draw_graphics(int program, ImageRenderData *data, GLuint start, GLuint count, float extra_alpha) {
bind_program(program);
if (program != GRAPHICS_ALPHA_MASK_PROGRAM) glUniform1f(program_uniform_location(program, "extra_alpha"), extra_alpha);
glActiveTexture(GL_TEXTURE0 + GRAPHICS_UNIT);
GLint src_rect_loc = program_uniform_location(program, "src_rect");
GLint dest_rect_loc = program_uniform_location(program, "dest_rect");
const bool is_indexed = program == GRAPHICS_INDEXED_PROGRAM || program == GRAPHICS_INDEXED_UNSCALED_PROGRAM;
const GLint has_alpha_loc = is_indexed ? program_uniform_location(program, "has_alpha") : -1;
for (GLuint i = 0; i < count;) {
ImageRenderData *group = data + start + i;
glBindTexture(GL_TEXTURE_2D, group->texture_id);
if (has_alpha_loc > -1) glUniform1ui(has_alpha_loc, group->indexed_px_sz == 2);
if (group->group_count == 0) {
i++;
continue;
}
for (GLuint k = 0; k < group->group_count; k++, i++) {
ImageRenderData *rd = data + start + i;
glUniform4f(src_rect_loc, rd->src_rect.left, rd->src_rect.top, rd->src_rect.right, rd->src_rect.bottom);
glUniform4f(dest_rect_loc, rd->dest_rect.left, rd->dest_rect.top, rd->dest_rect.right, rd->dest_rect.bottom);
draw_quad(true, 0);
}
}
}
static ImageRenderData *
load_alpha_mask_texture(size_t width, size_t height, uint8_t *canvas) {
static ImageRenderData data = {.group_count = 1};
if (!data.texture_id) { glGenTextures(1, &data.texture_id); }
glBindTexture(GL_TEXTURE_2D, data.texture_id);
glPixelStorei(GL_UNPACK_ALIGNMENT, 1);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RED, width, height, 0, GL_RED, GL_UNSIGNED_BYTE, canvas);
return &data;
}
static void
setup_texture_as_render_target(unsigned width, unsigned height, GLuint *texture_id, GLuint *framebuffer_id) {
glGenTextures(1, texture_id);
glGenFramebuffers(1, framebuffer_id);
glBindTexture(GL_TEXTURE_2D, *texture_id);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
// We use GL_RGBA16 to avoid incorrect colors due to quantization loss when
// blending, see https://github.com/kovidgoyal/kitty/issues/8953
static struct {
bool ok;
int fmt;
} status = {false, GL_RGBA16};
glTexImage2D(GL_TEXTURE_2D, 0, status.fmt, width, height, 0, GL_RGBA, GL_UNSIGNED_BYTE, NULL);
bind_framebuffer_for_output(*framebuffer_id);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_2D, *texture_id, 0);
if (!status.ok) {
if (check_framebuffer_status() == NULL) {
status.ok = true;
} else {
if (status.fmt == GL_RGBA16) {
// Driver does not support 16 bit FBO so let it choose the
// format. It will probably end up choosing 8 bit but
// inaccurate colors are better than completely broken rendering.
// See https://github.com/kovidgoyal/kitty/issues/9068
status.fmt = GL_RGBA;
free_framebuffer(framebuffer_id);
free_texture(texture_id);
setup_texture_as_render_target(width, height, texture_id, framebuffer_id);
log_error("WARNING: Your GPU driver does not support 16bit textures as framebuffer targets, some colors may be slightly inaccurate.");
} else {
fatal("Your GPU driver does not support indirect rendering to a GL_RGBA texture via a framebuffer");
}
}
}
}
static void
set_cell_uniforms(bool force) {
static bool constants_set = false;
if (!constants_set || force) {
float text_contrast = 1.0f + OPT(text_contrast) * 0.01f;
float text_gamma_adjustment = OPT(text_gamma_adjustment) < 0.01f ? 1.0f : 1.0f / OPT(text_gamma_adjustment);
for (int i = GRAPHICS_PROGRAM; i <= GRAPHICS_INDEXED_UNSCALED_PROGRAM; i++) {
bind_program(i);
glUniform1i(program_uniform_location(i, "image"), GRAPHICS_UNIT);
}
// These are set for every cell program, but the background program draws no
// text and so does not have them.
#define set_optional(setter, name, value) \
{ \
GLint loc = try_program_uniform_location(i, name); \
if (loc >= 0) setter(loc, value); \
}
for (int i = CELL_PROGRAM; i < CELL_PROGRAM_SENTINEL; i++) {
bind_program(i);
set_optional(glUniform1i, "sprites", SPRITE_MAP_UNIT);
set_optional(glUniform1i, "sprite_decorations_map", SPRITE_DECORATIONS_MAP_UNIT);
set_optional(glUniform1f, "text_contrast", text_contrast);
set_optional(glUniform1f, "text_gamma_adjustment", text_gamma_adjustment);
}
#undef set_optional
bind_program(BLIT_PROGRAM);
glUniform1i(program_uniform_location(BLIT_PROGRAM, "image"), GRAPHICS_UNIT);
bind_program(SCREENSHOT_PROGRAM);
glUniform1i(program_uniform_location(SCREENSHOT_PROGRAM, "image"), GRAPHICS_UNIT);
constants_set = true;
}
}
// UI Layer {{{
static Animation *default_visual_bell_animation = NULL;
static float
get_visual_bell_intensity(Screen *screen) {
if (screen->start_visual_bell_at > 0) {
if (!default_visual_bell_animation) {
default_visual_bell_animation = alloc_animation();
if (!default_visual_bell_animation) fatal("Out of memory");
add_cubic_bezier_animation(default_visual_bell_animation, 0, 1, EASE_IN_OUT);
add_cubic_bezier_animation(default_visual_bell_animation, 1, 0, EASE_IN_OUT);
}
const monotonic_t progress = monotonic() - screen->start_visual_bell_at;
const monotonic_t duration = OPT(visual_bell_duration) / 2;
if (progress <= duration) {
Animation *a = animation_is_valid(OPT(animation.visual_bell)) ? OPT(animation.visual_bell) : default_visual_bell_animation;
return (float)apply_easing_curve(a, progress / (double)duration, duration);
}
screen->start_visual_bell_at = 0;
}
return 0.0f;
}
static void
draw_visual_bell_flash(GLfloat intensity, const color_type flash) {
bind_program(TINT_PROGRAM);
GLfloat attenuation = 0.4f;
#define C(shift) srgb_color((flash >> shift) & 0xFF)
const GLfloat r = C(16), g = C(8), b = C(0);
const GLfloat max_channel = r > g ? (r > b ? r : b) : (g > b ? g : b);
#undef C
#define C(x) (x * intensity * attenuation)
if (max_channel > 0.45) attenuation = 0.6f; // light color
glUniform4f(program_uniform_location(TINT_PROGRAM, "tint_color"), C(r), C(g), C(b), C(1));
#undef C
glUniform4f(program_uniform_location(TINT_PROGRAM, "edges"), -1, 1, 1, -1);
draw_quad(true, 0);
}
static color_type
get_flash_color(const Screen *screen) {
#define COLOR(name, fallback) \
colorprofile_to_color_with_fallback( \
screen->color_profile, \
screen->color_profile->overridden.name, \
screen->color_profile->configured.name, \
screen->color_profile->overridden.fallback, \
screen->color_profile->configured.fallback)
return !IS_SPECIAL_COLOR(highlight_bg) ? COLOR(visual_bell_color, highlight_bg) : COLOR(visual_bell_color, default_fg);
#undef COLOR
}
static void
draw_visual_bell(const UIRenderData *ui) {
if (!ui->screen->start_visual_bell_at) return;
Screen *screen = ui->screen;
float intensity = get_visual_bell_intensity(screen);
if (intensity <= 0) return;
draw_visual_bell_flash(intensity, get_flash_color(screen));
}
static void
draw_drag_preview_overlay(const UIRenderData *ui) {
const Screen *screen = ui->screen;
if (!screen->start_drag_overlay_at || !screen->drag_overlay_quadrant) return;
const monotonic_t elapsed = monotonic() - screen->start_drag_overlay_at;
const monotonic_t fade_ms = ms_to_monotonic_t(150ll);
float intensity = elapsed >= fade_ms ? 1.0f : (float)elapsed / (float)fade_ms;
GLfloat left = -1.f, top = 1.f, right = 1.f, bottom = -1.f;
switch (screen->drag_overlay_quadrant) {
case 1: right = 0.f; break; // left half
case 2: left = 0.f; break; // right half
case 3: bottom = 0.f; break; // top half
case 4: top = 0.f; break; // bottom half
case 5: break; // full window + title bar highlight (title bar hover)
case 6: break; // full window, no title bar highlight (body hover)
default: return;
}
bind_program(TINT_PROGRAM);
float a = intensity * 0.25f;
color_type hint = get_flash_color(screen);
#define C(shift) (srgb_color((hint >> shift) & 0xFF) * a)
glUniform4f(program_uniform_location(TINT_PROGRAM, "tint_color"), C(16), C(8), C(0), a);
#undef C
glUniform4f(program_uniform_location(TINT_PROGRAM, "edges"), left, top, right, bottom);
draw_quad(true, 0);
}
static bool
has_scrollbar(Window *w, Screen *screen) {
return screen_has_scrollbar(screen, w->scrollbar.is_hovering);
}
static unsigned
bar_border_width(const UIRenderData *ui) {
return MAX(1u, (unsigned)ceil(thickness_as_float(ui->os_window, 1)));
}
// The rect, including borders, of a full width bar along the top of the window
static Viewport
window_bar_rect(const UIRenderData *ui) {
return (Viewport){.left = ui->screen_left, .top = ui->screen_top, .width = ui->screen_width, .height = ui->cell_height + 2 + 2 * bar_border_width(ui)};
}
static void
bar_colors(const UIRenderData *ui, color_type *fg, color_type *bg) {
#define RGBCOL(which, fallback) \
(0xff000000 | colorprofile_to_color_with_fallback( \
ui->screen->color_profile, \
ui->screen->color_profile->overridden.which, \
ui->screen->color_profile->configured.which, \
ui->screen->color_profile->overridden.fallback, \
ui->screen->color_profile->configured.fallback))
*fg = RGBCOL(default_fg, default_fg);
*bg = RGBCOL(default_bg, default_bg);
#undef RGBCOL
}
// Render title into bar->buf, if needed. The rendered text is max_width x
// height pixels, unless fit_to_text is true, in which case its width is reduced
// to the width of the text. The actual width is stored in bar->width.
static bool
render_bar_text(
const UIRenderData *ui, WindowBarData *bar, PyObject *title, unsigned max_width, unsigned height, bool fit_to_text, color_type fg, color_type bg) {
if (!title) return false;
if (!bar->buf || bar->requested_width != max_width || bar->height != height || bar->fit_to_text != fit_to_text) {
free(bar->buf);
bar->buf = malloc((size_t)4 * max_width * height);
if (!bar->buf) return false;
bar->requested_width = max_width;
bar->width = max_width;
bar->height = height;
bar->fit_to_text = fit_to_text;
bar->needs_render = true;
}
if (bar->last_drawn_title_object_id != title || bar->needs_render) {
static char titlebuf[2048] = {0};
snprintf(titlebuf, arraysz(titlebuf), fit_to_text ? " %s " : " %s", PyUnicode_AsUTF8(title));
size_t actual_width = max_width;
if (!draw_window_title(
ui->os_window->fonts_data->font_sz_in_pts,
ui->os_window->fonts_data->logical_dpi_y,
titlebuf,
fg,
bg,
bar->buf,
max_width,
height,
fit_to_text ? &actual_width : NULL))
return false;
bar->width = MAX(1u, (unsigned)actual_width);
Py_CLEAR(bar->last_drawn_title_object_id);
bar->last_drawn_title_object_id = Py_NewRef(title);
}
return true;
}
// Draw the text previously rendered by render_bar_text() surrounded by a
// border. border_rect must be the size of the rendered text plus the borders.
static void
draw_bar(const UIRenderData *ui, const WindowBarData *bar, Viewport border_rect, color_type fg, color_type bg) {
const double thickness = thickness_as_float(ui->os_window, 1);
const unsigned border_width = bar_border_width(ui);
// The rounded rect shader anti-aliases each edge over a band extending
// one pixel on either side of it, so keep the drawn line at least half a
// pixel inside the border_width pixels reserved for it, otherwise it would
// bleed onto the rendered title.
const float drawn_thickness = (float)MIN(thickness, border_width - 0.5);
static ImageRenderData data = {.group_count = 1};
gpu_data_for_image(&data, -1, 1, 1, -1);
glGenTextures(1, &data.texture_id);
glBindTexture(GL_TEXTURE_2D, data.texture_id);
glPixelStorei(GL_UNPACK_ALIGNMENT, 1);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
glTexImage2D(GL_TEXTURE_2D, 0, GL_SRGB_ALPHA, bar->width, bar->height, 0, GL_RGBA, GL_UNSIGNED_BYTE, bar->buf);
bind_program(GRAPHICS_PROGRAM);
const unsigned sh = ui->full_framebuffer_height;
// first blank the area to be drawn to background
enable_scissor_using_top_left_origin(border_rect, sh);
blank_canvas(ui->bg_alpha, bg, false);
disable_scissor();
// then draw the rendered text
save_viewport_using_top_left_origin(border_rect.left + border_width, border_rect.top + border_width, bar->width, bar->height, sh);
draw_graphics(GRAPHICS_PROGRAM, &data, 0, 1, 1.f);
restore_viewport();
free_texture(&data.texture_id);
// finally draw border with transparent bg
draw_rounded_rect(border_rect, sh, drawn_thickness, ui->cell_width, fg, bg, 0.f, 0, 0.f);
}
static unsigned
render_a_bar(const UIRenderData *ui, WindowBarData *bar, PyObject *title, Viewport border_rect) {
const unsigned border_width = bar_border_width(ui);
if (border_rect.width <= 2 * border_width || border_rect.height <= 2 * border_width) return 0;
color_type fg, bg;
bar_colors(ui, &fg, &bg);
if (!render_bar_text(ui, bar, title, border_rect.width - 2 * border_width, border_rect.height - 2 * border_width, false, fg, bg)) return 0;
draw_bar(ui, bar, border_rect, fg, bg);
return border_rect.height;
}
static bool
show_hyperlink_targets_with_modifiers(int mods) {
switch (OPT(show_hyperlink_targets)) {
case SHOW_HYPERLINK_TARGETS_ALWAYS: return true;
case SHOW_HYPERLINK_TARGETS_CTRL: return (mods & GLFW_MOD_CONTROL) != 0;
case SHOW_HYPERLINK_TARGETS_SHIFT: return (mods & GLFW_MOD_SHIFT) != 0;
case SHOW_HYPERLINK_TARGETS_SUPER: return (mods & GLFW_MOD_SUPER) != 0;
case SHOW_HYPERLINK_TARGETS_ALT: return (mods & GLFW_MOD_ALT) != 0;
default: return false;
}
}
static bool
has_hyperlink_target(OSWindow *os_window, Window *w, Screen *screen) {
return show_hyperlink_targets_with_modifiers(global_state.mods_at_last_key_or_button_event) && screen->current_hyperlink_under_mouse.id && w &&
!is_mouse_hidden(os_window) && global_state.mouse_hover_in_window == w->id;
}
// Return the offset from the top of the screen at which to draw a bar of the
// specified height so that it does not obscure the hovered line
static unsigned
hyperlink_target_bar_top(const UIRenderData *ui, unsigned bar_height) {
const double max_top = ui->screen_height - bar_height;
const double row_top = (double)ui->screen->current_hyperlink_under_mouse.y * ui->cell_height + ui->screen->pixel_scroll_offset_y;
const double row_bottom = row_top + ui->cell_height;
double top = 0;
switch (OPT(hyperlink_target_position)) {
case HYPERLINK_TARGET_TOP: top = row_top < bar_height ? max_top : 0; break;
case HYPERLINK_TARGET_BOTTOM: top = row_bottom > max_top ? 0 : max_top; break;
case HYPERLINK_TARGET_BELOW:
case HYPERLINK_TARGET_TOOLTIP: top = row_bottom > max_top ? row_top - bar_height : row_bottom; break;
}
return (unsigned)MAX(0., MIN(top, max_top));
}
// Draw the URL in a box sized to fit it, starting at the hovered cell, moved
// left as needed to fit in the window. rect is the full width bar it replaces.
static void
draw_hyperlink_target_tooltip(const UIRenderData *ui, Viewport rect, PyObject *url) {
const unsigned border_width = bar_border_width(ui);
if (rect.width <= 2 * border_width || rect.height <= 2 * border_width) return;
WindowBarData *bar = &ui->window->title_bar_data;
color_type fg, bg;
bar_colors(ui, &fg, &bg);
if (!render_bar_text(ui, bar, url, rect.width - 2 * border_width, rect.height - 2 * border_width, true, fg, bg)) return;
const unsigned width = bar->width + 2 * border_width;
const unsigned left = ui->screen->current_hyperlink_under_mouse.x * ui->cell_width;
rect.left += MIN(left, rect.width - width);
rect.width = width;
draw_bar(ui, bar, rect, fg, bg);
}
static void
draw_hyperlink_target(const UIRenderData *ui) {
if (!has_hyperlink_target(ui->os_window, ui->window, ui->screen)) return;
Screen *screen = ui->screen;
Window *window = ui->window;
Viewport rect = window_bar_rect(ui);
if (rect.height > ui->screen_height) return;
WindowBarData *bd = &window->url_target_bar_data;
hyperlink_id_type hid = screen->current_hyperlink_under_mouse.id;
if (bd->hyperlink_id_for_title_object != hid) {
bd->hyperlink_id_for_title_object = hid;
Py_CLEAR(bd->last_drawn_title_object_id);
const char *url = get_hyperlink_for_id(screen->hyperlink_pool, bd->hyperlink_id_for_title_object, true);
if (url == NULL) url = "";
bd->last_drawn_title_object_id = PyObject_CallMethod(global_state.boss, "sanitize_url_for_display_to_user", "s", url);
if (bd->last_drawn_title_object_id == NULL) {
PyErr_Print();
return;
}
bd->needs_render = true;
}
if (bd->last_drawn_title_object_id == NULL) return;
PyObject *ref = Py_NewRef(bd->last_drawn_title_object_id); // render_bar_text() clears bd->last_drawn_title_object_id
rect.top += hyperlink_target_bar_top(ui, rect.height);
if (OPT(hyperlink_target_position) == HYPERLINK_TARGET_TOOLTIP) draw_hyperlink_target_tooltip(ui, rect, bd->last_drawn_title_object_id);
else render_a_bar(ui, &window->title_bar_data, bd->last_drawn_title_object_id, rect);
Py_DECREF(ref);
}
static bool
has_window_number(Window *w, Screen *screen) {
return w != NULL && screen->display_window_char != 0;
}
static void
draw_window_number(const UIRenderData *ui) {
if (!has_window_number(ui->window, ui->screen)) return;
unsigned title_bar_height = 0, requested_height = ui->screen_height;
if (ui->window->title && PyUnicode_Check(ui->window->title) && (requested_height > (ui->cell_height + 1) * 2)) {
title_bar_height = render_a_bar(ui, &ui->window->title_bar_data, ui->window->title, window_bar_rect(ui));
}
unsigned height_for_letter = ui->screen_height - title_bar_height - ui->cell_height;
unsigned width_for_letter = ui->screen_width - ui->cell_width;
requested_height = MIN(12 * ui->cell_height, MIN(height_for_letter, width_for_letter));
if (requested_height < 4) return;
#define lr ui->screen->last_rendered_window_char
if (!lr.canvas || lr.ch != ui->screen->display_window_char || lr.requested_height != requested_height) {
free(lr.canvas);
lr.canvas = NULL;
lr.requested_height = requested_height;
lr.height_px = requested_height;
lr.ch = 0;
lr.canvas = draw_single_ascii_char(ui->screen->display_window_char, &lr.width_px, &lr.height_px);
if (lr.height_px < 4 || lr.width_px < 4 || !lr.canvas) return;
lr.ch = ui->screen->display_window_char;
}
unsigned letter_x = 0, letter_y = title_bar_height;
if (lr.width_px < ui->screen_width) letter_x = (ui->screen_width - lr.width_px) / 2;
if (lr.height_px + title_bar_height < ui->screen_height) letter_y += (ui->screen_height - lr.height_px - title_bar_height) / 2;
bind_program(GRAPHICS_ALPHA_MASK_PROGRAM);
ImageRenderData *ird = load_alpha_mask_texture(lr.width_px, lr.height_px, lr.canvas);
gpu_data_for_image(ird, -1, 1, 1, -1);
glUniform1i(program_uniform_location(GRAPHICS_ALPHA_MASK_PROGRAM, "image"), GRAPHICS_UNIT);
color_type digit_color = colorprofile_to_color_with_fallback(
ui->screen->color_profile,
ui->screen->color_profile->overridden.highlight_bg,
ui->screen->color_profile->configured.highlight_bg,
ui->screen->color_profile->overridden.default_fg,
ui->screen->color_profile->configured.default_fg);
color_vec3(program_uniform_location(GRAPHICS_ALPHA_MASK_PROGRAM, "amask_fg"), digit_color);
glUniform4f(program_uniform_location(GRAPHICS_ALPHA_MASK_PROGRAM, "amask_bg_premult"), 0.f, 0.f, 0.f, 0.f);
save_viewport_using_top_left_origin(ui->screen_left + letter_x, ui->screen_top + letter_y, lr.width_px, lr.height_px, ui->full_framebuffer_height);
draw_graphics(GRAPHICS_ALPHA_MASK_PROGRAM, ird, 0, 1, 1.f);
restore_viewport();
#undef lr
}
// Helper function to extract and apply opacity to color components
static void
set_color_uniform_with_opacity(color_type color, float opacity) {
float r = srgb_color((color >> 16) & 0xFF) * opacity;
float g = srgb_color((color >> 8) & 0xFF) * opacity;
float b = srgb_color(color & 0xFF) * opacity;
glUniform4f(program_uniform_location(TINT_PROGRAM, "tint_color"), r, g, b, opacity);
}
static color_type
scrollbar_color(Screen *screen, unsigned val) {
switch (val & 0xff) {
#define C(which) colorprofile_to_color(screen->color_profile, screen->color_profile->overridden.which, screen->color_profile->configured.which).rgb
case 0: return C(default_fg);
case 1: return C(highlight_bg);
#undef C
default: return val >> 8;
}
}
static void
draw_scrollbar(const UIRenderData *ui) {
Screen *screen = ui->screen;
Window *window = ui->window;
if (!window || !screen || !has_scrollbar(window, screen)) return;
color_type bar_color = scrollbar_color(screen, OPT(scrollbar_handle_color)), track_color = scrollbar_color(screen, OPT(scrollbar_track_color));
float bar_frac, visible_fraction;
screen_scrollbar_geometry(screen, &bar_frac, &visible_fraction);
float opacity = OPT(scrollbar_handle_opacity);
float track_opacity = window->scrollbar.is_hovering ? OPT(scrollbar_track_hover_opacity) : OPT(scrollbar_track_opacity);
GLsizei scrollbar_width_px = (GLsizei)(OPT(scrollbar_width) * ui->cell_width);
if (window->scrollbar.is_hovering) scrollbar_width_px = (GLsizei)(OPT(scrollbar_hover_width) * ui->cell_width);
GLsizei scrollbar_gap_px = (GLsizei)(OPT(scrollbar_gap) * ui->cell_width);
unsigned scrollbar_radius = (unsigned)(OPT(scrollbar_radius) * ui->cell_width);
// Calculate window boundaries including padding
GLsizei window_right_edge = ui->screen_left + ui->screen_width + window->render_data.geometry.spaces.right;
GLsizei window_top_edge = ui->screen_top - window->render_data.geometry.spaces.top;
GLsizei window_height = ui->screen_height + window->render_data.geometry.spaces.top + window->render_data.geometry.spaces.bottom;
// Position scrollbar on right side with gap
GLsizei scrollbar_left = window_right_edge - scrollbar_width_px - scrollbar_gap_px;
GLsizei scrollbar_top = window_top_edge + scrollbar_gap_px;
GLsizei scrollbar_height = window_height - 2 * scrollbar_gap_px;
// Calculate thumb size and position
float min_thumb_height_fraction = (OPT(scrollbar_min_handle_height) * ui->cell_height) / (float)window_height;
float thumb_height_fraction = MAX(min_thumb_height_fraction, visible_fraction);
// Convert to OpenGL coordinates (range -1.0 to 1.0, total span = 2.0)
const float GL_COORD_SPAN = 2.0f;
float thumb_height_gl = thumb_height_fraction * GL_COORD_SPAN;
float available_space = GL_COORD_SPAN - thumb_height_gl;
float thumb_bottom_gl = -1.0f + available_space * bar_frac;
float thumb_top_gl = thumb_bottom_gl + thumb_height_gl;
// Store thumb position for mouse interaction (normalized window coordinates)
float scrollbar_top_in_window = (float)(window_top_edge + scrollbar_gap_px) / (float)ui->full_framebuffer_height;
float scrollbar_height_in_window = (float)(window_height - 2 * scrollbar_gap_px) / (float)ui->full_framebuffer_height;
float thumb_top_fraction = (1.0f - thumb_top_gl) / 2.0f;
float thumb_bottom_fraction = (1.0f - thumb_bottom_gl) / 2.0f;
window->scrollbar.thumb_top = scrollbar_top_in_window + thumb_top_fraction * scrollbar_height_in_window;
window->scrollbar.thumb_bottom = scrollbar_top_in_window + thumb_bottom_fraction * scrollbar_height_in_window;
// Set viewport for scrollbar area
save_viewport_using_top_left_origin(scrollbar_left, scrollbar_top, scrollbar_width_px, scrollbar_height, ui->full_framebuffer_height);
// Draw scrollbar track (background)
if (track_opacity > 0) {
bind_program(TINT_PROGRAM);
set_color_uniform_with_opacity(track_color, track_opacity);
glUniform4f(program_uniform_location(TINT_PROGRAM, "edges"), -1.f, 1.f, 1.f, -1.f);
draw_quad(true, 0);
}
// Draw scrollbar thumb (handle)
if (scrollbar_radius > 0) {
// Rounded thumb - use separate viewport and rounded rect program
GLsizei thumb_height_px = (GLsizei)(thumb_height_fraction * scrollbar_height);
GLsizei thumb_top_px = scrollbar_top + (GLsizei)(thumb_top_fraction * scrollbar_height);
restore_viewport();
bind_program(ROUNDED_RECT_PROGRAM);
color_vec4(program_uniform_location(ROUNDED_RECT_PROGRAM, "color"), bar_color, opacity);
color_vec4(program_uniform_location(ROUNDED_RECT_PROGRAM, "background_color"), 0, 0.0f);
color_vec4(program_uniform_location(ROUNDED_RECT_PROGRAM, "outer_color"), 0, 0.0f);
float y = (float)ui->full_framebuffer_height - (float)(thumb_top_px + thumb_height_px);
glUniform4f(program_uniform_location(ROUNDED_RECT_PROGRAM, "rect"), (float)scrollbar_left, y, (float)scrollbar_width_px, (float)thumb_height_px);
float thickness = (float)MAX(scrollbar_width_px, thumb_height_px);
glUniform2f(program_uniform_location(ROUNDED_RECT_PROGRAM, "params"), thickness, (float)scrollbar_radius);
save_viewport_using_top_left_origin(scrollbar_left, thumb_top_px, scrollbar_width_px, thumb_height_px, ui->full_framebuffer_height);
draw_quad(true, 0);
restore_viewport();
} else {
set_color_uniform_with_opacity(bar_color, opacity);
glUniform4f(program_uniform_location(TINT_PROGRAM, "edges"), -1.f, thumb_top_gl, 1.f, thumb_bottom_gl);
draw_quad(true, 0);
restore_viewport();
}
}
static bool
has_progress_bar(Screen *screen) {
if (OPT(progress_bar) == PROGRESS_BAR_HIDDEN) return false;
return screen && screen->progress_state != PROGRESS_STATE_UNSET;
}
static void
draw_progress_handle(
const UIRenderData *ui,
color_type bar_color,
float opacity,
unsigned bar_radius,
GLsizei track_left,
GLsizei track_top,
GLsizei track_width,
GLsizei track_height,
float handle_start,
float handle_size,
bool is_horizontal) {
// handle_start and handle_size are fractions of the track length (0..1)
// For horizontal: handle moves left-to-right; For vertical: handle moves top-to-bottom
// Use lroundf to avoid sub-pixel jitter at the leading edge
GLsizei handle_left, handle_top;
GLsizei handle_w, handle_h;
if (is_horizontal) {
GLsizei handle_start_px = (GLsizei)lroundf(handle_start * track_width);
GLsizei handle_end_px = (GLsizei)lroundf((handle_start + handle_size) * track_width);
GLsizei handle_w_px = handle_end_px - handle_start_px;
if (handle_w_px < 1) handle_w_px = 1;
handle_left = track_left + handle_start_px;
handle_top = track_top;
handle_w = handle_w_px;
handle_h = track_height;
} else {
GLsizei handle_start_px = (GLsizei)lroundf(handle_start * track_height);
GLsizei handle_end_px = (GLsizei)lroundf((handle_start + handle_size) * track_height);
GLsizei handle_h_px = handle_end_px - handle_start_px;
if (handle_h_px < 1) handle_h_px = 1;
handle_left = track_left;
handle_top = track_top + handle_start_px;
handle_w = track_width;
handle_h = handle_h_px;
}
if (bar_radius > 0) {
bind_program(ROUNDED_RECT_PROGRAM);
color_vec4(program_uniform_location(ROUNDED_RECT_PROGRAM, "color"), bar_color, opacity);
color_vec4(program_uniform_location(ROUNDED_RECT_PROGRAM, "background_color"), 0, 0.0f);
color_vec4(program_uniform_location(ROUNDED_RECT_PROGRAM, "outer_color"), 0, 0.0f);
float y = (float)ui->full_framebuffer_height - (float)(handle_top + handle_h);
glUniform4f(program_uniform_location(ROUNDED_RECT_PROGRAM, "rect"), (float)handle_left, y, (float)handle_w, (float)handle_h);
float thickness = (float)(is_horizontal ? handle_h : handle_w);
glUniform2f(program_uniform_location(ROUNDED_RECT_PROGRAM, "params"), thickness, (float)bar_radius);
save_viewport_using_top_left_origin(handle_left, handle_top, handle_w, handle_h, ui->full_framebuffer_height);
draw_quad(true, 0);
restore_viewport();
} else {
// Use GL coordinates within the track viewport, snapped to pixel boundaries
GLsizei track_len = is_horizontal ? track_width : track_height;
float start_snapped = (float)lroundf(handle_start * track_len) / (float)track_len;
float end_snapped = (float)lroundf((handle_start + handle_size) * track_len) / (float)track_len;
float start_gl = -1.0f + 2.0f * start_snapped;
float end_gl = -1.0f + 2.0f * end_snapped;
bind_program(TINT_PROGRAM);
set_color_uniform_with_opacity(bar_color, opacity);
if (is_horizontal) {
// edges: left, top, right, bottom
save_viewport_using_top_left_origin(track_left, track_top, track_width, track_height, ui->full_framebuffer_height);
glUniform4f(program_uniform_location(TINT_PROGRAM, "edges"), start_gl, 1.f, end_gl, -1.f);
} else {
save_viewport_using_top_left_origin(track_left, track_top, track_width, track_height, ui->full_framebuffer_height);
// For vertical: start_gl is top in GL coords (inverted y), so bottom_gl = -end_gl, top_gl = -start_gl
glUniform4f(program_uniform_location(TINT_PROGRAM, "edges"), -1.f, -start_gl, 1.f, -end_gl);
}
draw_quad(true, 0);
restore_viewport();
}
}
static void
draw_progress_bar(const UIRenderData *ui) {
Screen *screen = ui->screen;
Window *window = ui->window;
if (!window || !has_progress_bar(screen)) return;
const ProgressBarPosition pos = OPT(progress_bar);
const bool is_horizontal = (pos == PROGRESS_BAR_TOP || pos == PROGRESS_BAR_BOTTOM);
color_type bar_color = scrollbar_color(screen, OPT(scrollbar_handle_color));
color_type track_color = scrollbar_color(screen, OPT(scrollbar_track_color));
float opacity = OPT(scrollbar_handle_opacity);
float track_opacity = OPT(scrollbar_track_hover_opacity);
GLsizei bar_thickness_px = (GLsizei)(OPT(scrollbar_width) * ui->cell_width);
GLsizei gap_px = (GLsizei)(OPT(scrollbar_gap) * ui->cell_width);
unsigned bar_radius = (unsigned)(OPT(scrollbar_radius) * ui->cell_width);
if (bar_thickness_px < 1) return;
// Calculate window boundaries including padding
GLsizei window_left_edge = ui->screen_left - (GLsizei)window->render_data.geometry.spaces.left;
GLsizei window_top_edge = ui->screen_top - (GLsizei)window->render_data.geometry.spaces.top;
GLsizei window_width = ui->screen_width + (GLsizei)(window->render_data.geometry.spaces.left + window->render_data.geometry.spaces.right);
GLsizei window_height = ui->screen_height + (GLsizei)(window->render_data.geometry.spaces.top + window->render_data.geometry.spaces.bottom);
// Position the track depending on the chosen edge
GLsizei track_left, track_top, track_width, track_height;
switch (pos) {
case PROGRESS_BAR_BOTTOM:
track_left = window_left_edge + gap_px;
track_width = window_width - 2 * gap_px;
track_height = bar_thickness_px;
track_top = window_top_edge + window_height - bar_thickness_px - gap_px;
break;
case PROGRESS_BAR_TOP:
track_left = window_left_edge + gap_px;
track_width = window_width - 2 * gap_px;
track_height = bar_thickness_px;
track_top = window_top_edge + gap_px;
break;
case PROGRESS_BAR_LEFT:
track_left = window_left_edge + gap_px;
track_width = bar_thickness_px;
track_top = window_top_edge + gap_px;
track_height = window_height - 2 * gap_px;
break;
case PROGRESS_BAR_RIGHT:
track_left = window_left_edge + window_width - bar_thickness_px - gap_px;
track_width = bar_thickness_px;
track_top = window_top_edge + gap_px;
track_height = window_height - 2 * gap_px;
break;
default: return;
}
if (track_width <= 0 || track_height <= 0) return;
// Calculate fill fraction and indeterminate animation
float fill_fraction = 0.0f;
bool is_indeterminate = false;
switch (screen->progress_state) {
case PROGRESS_STATE_SET:
case PROGRESS_STATE_PAUSED: fill_fraction = screen->progress_percent / 100.0f; break;
case PROGRESS_STATE_ERROR: fill_fraction = 1.0f; break;
case PROGRESS_STATE_INDETERMINATE: is_indeterminate = true; break;
default: return;
}
// Draw track (background)
save_viewport_using_top_left_origin(track_left, track_top, track_width, track_height, ui->full_framebuffer_height);
if (track_opacity > 0) {
bind_program(TINT_PROGRAM);
set_color_uniform_with_opacity(track_color, track_opacity);
glUniform4f(program_uniform_location(TINT_PROGRAM, "edges"), -1.f, 1.f, 1.f, -1.f);
draw_quad(true, 0);
}
restore_viewport();
// Draw fill or indeterminate handle
// For vertical bars, progress grows from bottom to top, so invert the fraction
if (is_indeterminate) {
// Animate a handle bouncing back and forth like a scrollbar thumb
const float handle_size = 0.15f; // 15% of track length
const monotonic_t cycle = s_double_to_monotonic_t(2.0); // 2 second full cycle (forth and back)
monotonic_t elapsed = screen->progress_indeterminate_anim_at > 0 ? monotonic() - screen->progress_indeterminate_anim_at : 0;
double t = (double)(elapsed % cycle) / (double)cycle; // 0..1 over one cycle
// Triangle wave: goes 0→1→0 over one cycle
float pos_frac = (float)(t < 0.5 ? t * 2.0 : 2.0 - t * 2.0);
if (!is_horizontal) pos_frac = 1.0f - pos_frac; // vertical: bounce bottom-to-top first
float handle_start = pos_frac * (1.0f - handle_size);
if (opacity > 0.0f) {
draw_progress_handle(
ui, bar_color, opacity, bar_radius, track_left, track_top, track_width, track_height, handle_start, handle_size, is_horizontal);
}
} else if (fill_fraction > 0.0f && opacity > 0.0f) {
float handle_start = is_horizontal ? 0.0f : 1.0f - fill_fraction;
draw_progress_handle(ui, bar_color, opacity, bar_radius, track_left, track_top, track_width, track_height, handle_start, fill_fraction, is_horizontal);
}
}
static void
draw_window_logo(const UIRenderData *ui) {
struct {
unsigned width, height;
int left, top;
} w;
WindowLogoRenderData *wl = ui->window_logo;
w.height = wl->instance->height;
w.width = wl->instance->width;
if (OPT(window_logo_scale.width) > 0 || OPT(window_logo_scale.height) > 0) {
unsigned scaled_wl_width = ui->screen_width, scaled_wl_height = ui->screen_height;
// [sx] Scales logo to sx % of the viewports shortest dimension, preserving aspect ratio
if (OPT(window_logo_scale.height) < 0) {
if (ui->screen_height < ui->screen_width) {
scaled_wl_height = (int)(ui->screen_height * OPT(window_logo_scale.width) / 100);
scaled_wl_width = wl->instance->width * scaled_wl_height / wl->instance->height;
} else {
scaled_wl_width = (int)(ui->screen_width * OPT(window_logo_scale.width) / 100);
scaled_wl_height = wl->instance->height * scaled_wl_width / wl->instance->width;
}
}
// [0 sy] Scales logo's y dimension to sy % of viewporty keeping original x dimension
else if (OPT(window_logo_scale.width) == 0.0) {
scaled_wl_height = (int)(scaled_wl_height * OPT(window_logo_scale.height) / 100);
scaled_wl_width = wl->instance->width;
}
// [sx 0] Scales logo's x dimension to sx % of viewportx keeping original y dimension
else if (OPT(window_logo_scale.height) == 0.0) {
scaled_wl_width = (int)(scaled_wl_width * OPT(window_logo_scale.width) / 100);
scaled_wl_height = wl->instance->height;
}
// [sx sy] Scales logo's x and y dimension to sx and sy % of viewportx and viewporty respectively
else {
scaled_wl_height = (int)(scaled_wl_height * OPT(window_logo_scale.height) / 100);
scaled_wl_width = (int)(scaled_wl_width * OPT(window_logo_scale.width) / 100);
}
w.width = scaled_wl_width;
w.height = scaled_wl_height;
}
w.left = (int)(ui->screen_width * wl->position.canvas_x - w.width * wl->position.image_x);
w.top = (int)(ui->screen_height * wl->position.canvas_y - w.height * wl->position.image_y);
float left = gl_pos_x(w.left, ui->screen_width), top = gl_pos_y(w.top, ui->screen_height);
ImageRenderData d = {.texture_id = wl->instance->texture_id};
gpu_data_for_image(&d, left, top, left + gl_size(w.width, ui->screen_width), top - gl_size(w.height, ui->screen_height));
draw_graphics(GRAPHICS_PROGRAM, &d, 0, 1, ui->inactive_text_alpha * OPT(window_logo_alpha));
}
bool
screen_needs_rendering_in_layers(OSWindow *os_window, Window *w, Screen *screen) {
const bool has_ui = w && ((screen->start_visual_bell_at | screen->start_drag_overlay_at) || has_scrollbar(w, screen) || has_progress_bar(screen) ||
has_hyperlink_target(os_window, w, screen) || has_window_number(w, screen) || w->window_logo.id);
GraphicsManager *grman = screen->paused_rendering.expires_at && screen->paused_rendering.grman ? screen->paused_rendering.grman : screen->grman;
return has_ui || grman_has_images(grman);
}
bool
current_framebuffer_is_ok(void) {
return check_framebuffer_status() == NULL;
}
// }}}
enum { DRAW_NEITHER_BG = 0, DRAW_DEFAULT_BG = 1, DRAW_NON_DEFAULT_BG = 2, DRAW_BOTH_BG = 3 };
static void
call_cell_program(int program, const UIRenderData *ui, ssize_t vao_idx, bool for_final_output, unsigned draw_bg_bitfield) {
bind_program(program);
CELL_BUFFERS;
bind_vao_uniform_buffer(vao_idx, uniform_buffer, CELL_RENDER_DATA_BINDING_POINT);
bind_vao_uniform_buffer(vao_idx, color_table_buffer, COLOR_TABLE_BINDING_POINT);
glUniform1ui(program_uniform_location(program, "draw_bg_bitfield"), draw_bg_bitfield);
if (for_final_output) glEnable(GL_FRAMEBUFFER_SRGB);
draw_quad(!for_final_output, render_lines_for_screen(ui->screen) * ui->screen->columns);
if (for_final_output) glDisable(GL_FRAMEBUFFER_SRGB);
}
static void
draw_cells_without_layers(const UIRenderData *ui, ssize_t vao_idx) {
call_cell_program(CELL_PROGRAM, ui, vao_idx, true, DRAW_BOTH_BG);
}
static void
configure_cell_vao_attributes(ssize_t vao_idx, unsigned int base_cell, unsigned int cell_step) {
// (Re)point the instanced cell attributes so that instance i corresponds to
// the cell at (base_cell + i*cell_step). Used to restore the canonical
// layout after padding draws (base_cell=0, cell_step=1). The VAO must be
// bound before calling this.
CELL_BUFFERS;
const GLsizei cell_stride = (GLsizei)(cell_step * sizeof(GPUCell));
const uintptr_t cell_base = (uintptr_t)base_cell * sizeof(GPUCell);
set_vao_attribute(
vao_idx,
cell_data_buffer,
program_attribute_location(CELL_PROGRAM, "sprite_idx"),
2,
GL_UNSIGNED_INT,
cell_stride,
(void *)(cell_base + offsetof(GPUCell, sprite_idx)),
1);
set_vao_attribute(
vao_idx,
cell_data_buffer,
program_attribute_location(CELL_PROGRAM, "colors"),
3,
GL_UNSIGNED_INT,
cell_stride,
(void *)(cell_base + offsetof(GPUCell, fg)),
1);
set_vao_attribute(
vao_idx,
selection_buffer,
program_attribute_location(CELL_PROGRAM, "is_selected"),
1,
GL_UNSIGNED_BYTE,
(GLsizei)(cell_step * sizeof(GLubyte)),
(void *)((uintptr_t)base_cell * sizeof(GLubyte)),
1);
}
static void
configure_padding_vao_attributes(ssize_t vao_idx) {
// Point the instanced cell attributes at the dedicated padding buffers
// (base=0, stride=sizeof(GPUCell)), which are pre-packed before each
// combined draw. The VAO must be bound before calling this.
CELL_BUFFERS;
set_vao_attribute(
vao_idx,
padding_cell_data_buffer,
program_attribute_location(CELL_PROGRAM, "sprite_idx"),
2,
GL_UNSIGNED_INT,
sizeof(GPUCell),
(void *)offsetof(GPUCell, sprite_idx),
1);
set_vao_attribute(
vao_idx,
padding_cell_data_buffer,
program_attribute_location(CELL_PROGRAM, "colors"),
3,
GL_UNSIGNED_INT,
sizeof(GPUCell),
(void *)offsetof(GPUCell, fg),
1);
set_vao_attribute(
vao_idx, padding_selection_buffer, program_attribute_location(CELL_PROGRAM, "is_selected"), 1, GL_UNSIGNED_BYTE, sizeof(GLubyte), NULL, 1);
}
static void
draw_window_padding(const UIRenderData *ui, unsigned int cl, unsigned int ct, unsigned int cr, unsigned int cb, ssize_t vao_idx, bool for_final_output) {
// Color the compensatory padding strips (the innermost slice of the window
// padding, arising from the window size not being an exact multiple of the
// cell size) to match their neighboring cell. The strips lie outside the
// per-window cell viewport, so this runs with the full framebuffer viewport.
//
// Two instanced draws are issued: one covers the horizontal pair (top+bottom)
// and one the vertical pair (left+right). Each draw packs its strip cells into
// a dedicated buffer so the VAO needs no per-strip reconfiguration. The shader
// selects per-strip geometry and cell indices branch-free via lerp.
if (OPT(padding_fill_strategy) != PADDING_FILL_NEIGHBORING_CELL || !(cl | ct | cr | cb)) return;
Screen *screen = ui->screen;
const unsigned int columns = screen->columns, lines = screen->lines;
if (!columns || !lines) return;
const unsigned int render_offset = pixel_scroll_enabled(screen) ? 1u : 0u;
const unsigned int top_row = render_offset, bottom_row = render_offset + lines - 1u;
const float fbw = (float)ui->full_framebuffer_width, fbh = (float)ui->full_framebuffer_height;
const float cw = (float)ui->cell_width, ch = (float)ui->cell_height;
const float fL = (float)ui->screen_left, T = (float)ui->screen_top;
const float R = fL + (float)ui->screen_width, B = T + (float)ui->screen_height;
#define NX(px) (2.f * (px) / fbw - 1.f)
#define NY(px) (1.f - 2.f * (px) / fbh)
const float dx = 2.f * cw / fbw, dy = 2.f * ch / fbh;
bind_program(PADDING_PROGRAM);
bind_vertex_array(vao_idx);
CELL_BUFFERS;
bind_vao_uniform_buffer(vao_idx, uniform_buffer, CELL_RENDER_DATA_BINDING_POINT);
bind_vao_uniform_buffer(vao_idx, color_table_buffer, COLOR_TABLE_BINDING_POINT);
if (for_final_output) glEnable(GL_FRAMEBUFFER_SRGB);
// Point instanced attributes at the padding-specific buffers once for both draws.
configure_padding_vao_attributes(vao_idx);
#define PL(x) program_uniform_location(PADDING_PROGRAM, #x)
// Horizontal combined draw: top strip (strip 0) + bottom strip (strip 1).
// Cell data is packed contiguously via GPU-side copies (no CPU round-trip).
if (ct || cb) {
const unsigned int nH = (ct ? 1u : 0u) + (cb ? 1u : 0u);
alloc_vao_buffer(vao_idx, (GLsizeiptr)(sizeof(GPUCell) * nH * columns), padding_cell_data_buffer, GL_DYNAMIC_DRAW);
alloc_vao_buffer(vao_idx, (GLsizeiptr)(sizeof(GLubyte) * nH * columns), padding_selection_buffer, GL_DYNAMIC_DRAW);
unsigned int sidx = 0u;
if (ct) {
copy_vao_buffer_region(
vao_idx,
cell_data_buffer,
(GLintptr)(sizeof(GPUCell) * top_row * columns),
padding_cell_data_buffer,
(GLintptr)(sizeof(GPUCell) * sidx * columns),
(GLsizeiptr)(columns * sizeof(GPUCell)));
copy_vao_buffer_region(
vao_idx,
selection_buffer,
(GLintptr)(sizeof(GLubyte) * top_row * columns),
padding_selection_buffer,
(GLintptr)(sizeof(GLubyte) * sidx * columns),
(GLsizeiptr)(columns * sizeof(GLubyte)));
sidx++;
}
if (cb) {
copy_vao_buffer_region(
vao_idx,
cell_data_buffer,
(GLintptr)(sizeof(GPUCell) * bottom_row * columns),
padding_cell_data_buffer,
(GLintptr)(sizeof(GPUCell) * sidx * columns),
(GLsizeiptr)(sizeof(GPUCell) * columns));
copy_vao_buffer_region(
vao_idx,
selection_buffer,
(GLintptr)(sizeof(GLubyte) * bottom_row * columns),
padding_selection_buffer,
(GLintptr)(sizeof(GLubyte) * sidx * columns),
(GLsizeiptr)(columns * sizeof(GLubyte)));
}
// Strip-0 is top (or bottom when only bottom exists); strip-1 is bottom.
// For a single-strip draw base_instance2/across2 equal strip-0 values so
// the shader lerp is a no-op for all instances (strip_f is always 0.0).
const unsigned int base0 = (ct ? top_row : bottom_row) * columns;
const unsigned int base1 = bottom_row * columns;
glUniform1ui(PL(is_horizontal), 1u);
glUniform1ui(PL(along_count), columns);
glUniform1ui(PL(instance_step), 1u);
glUniform1ui(PL(base_instance), base0);
glUniform1ui(PL(base_instance2), base1);
glUniform2f(PL(across), ct ? NY(T - ct) : NY(B), ct ? NY(T) : NY(B + cb));
glUniform2f(PL(across2), NY(B), NY(B + cb));
glUniform1f(PL(along_start), NX(fL));
glUniform1f(PL(along_step), dx);
glUniform2f(PL(along_clamp), NX(fL), NX(R));
draw_quad(!for_final_output, nH * columns);
}
// Vertical combined draw: left strip (strip 0) + right strip (strip 1).
// Column cells are non-contiguous in the cell buffer so we gather them
// CPU-side by mapping the buffer for reading, then upload in one shot.
if (cl || cr) {
const unsigned int nV = (cl ? 1u : 0u) + (cr ? 1u : 0u);
// VLA sizes: lines is bounded by the window height / cell height (~<500).
GPUCell gathered_cells[2 * lines];
GLubyte gathered_sel[2 * lines];
const unsigned int strip0_col = cl ? 0u : (columns - 1u);
const unsigned int strip1_col = columns - 1u;
const GPUCell *cells = (const GPUCell *)map_vao_buffer_for_reading(vao_idx, cell_data_buffer);
for (unsigned int i = 0; i < lines; i++) {
gathered_cells[i] = cells[(top_row + i) * columns + strip0_col];
if (nV == 2u) gathered_cells[lines + i] = cells[(top_row + i) * columns + strip1_col];
}
unmap_vao_buffer(vao_idx, cell_data_buffer);
const GLubyte *sel = (const GLubyte *)map_vao_buffer_for_reading(vao_idx, selection_buffer);
for (unsigned int i = 0; i < lines; i++) {
gathered_sel[i] = sel[(top_row + i) * columns + strip0_col];
if (nV == 2u) gathered_sel[lines + i] = sel[(top_row + i) * columns + strip1_col];
}
unmap_vao_buffer(vao_idx, selection_buffer);
void *dst = alloc_and_map_vao_buffer(vao_idx, (GLsizeiptr)(sizeof(GPUCell) * nV * lines), padding_cell_data_buffer);
memcpy(dst, gathered_cells, sizeof(GPUCell) * nV * lines);
unmap_vao_buffer(vao_idx, padding_cell_data_buffer);
dst = alloc_and_map_vao_buffer(vao_idx, (GLsizeiptr)(sizeof(GLubyte) * nV * lines), padding_selection_buffer);
memcpy(dst, gathered_sel, sizeof(GLubyte) * nV * lines);
unmap_vao_buffer(vao_idx, padding_selection_buffer);
const unsigned int base0 = top_row * columns + (cl ? 0u : (columns - 1u));
const unsigned int base1 = top_row * columns + (columns - 1u);
glUniform1ui(PL(is_horizontal), 0u);
glUniform1ui(PL(along_count), lines);
glUniform1ui(PL(instance_step), columns);
glUniform1ui(PL(base_instance), base0);
glUniform1ui(PL(base_instance2), base1);
glUniform2f(PL(across), cl ? NX(fL - cl) : NX(R), cl ? NX(fL) : NX(R + cr));
glUniform2f(PL(across2), NX(R), NX(R + cr));
glUniform1f(PL(along_start), NY(T));
glUniform1f(PL(along_step), -dy);
glUniform2f(PL(along_clamp), NY(T - ct), NY(B + cb));
draw_quad(!for_final_output, nV * lines);
}
#undef PL
if (for_final_output) glDisable(GL_FRAMEBUFFER_SRGB);
// Restore the canonical cell attribute layout so subsequent cell rendering is unaffected.
configure_cell_vao_attributes(vao_idx, 0u, 1u);
unbind_program();
#undef NX
#undef NY
}
static void
draw_tint(const UIRenderData *ui) {
bind_program(TINT_PROGRAM);
color_vec4_premult(program_uniform_location(TINT_PROGRAM, "tint_color"), ui->background_color, OPT(background_tint));
glUniform4f(program_uniform_location(TINT_PROGRAM, "edges"), -1, 1, 1, -1);
draw_quad(true, 0);
}
// Draw a range of the image render data. Indexed color images need a
// different program, so when any are present, switch programs at the
// boundaries between runs of images that need different programs, preserving
// draw order.
static int
program_for_image(const ImageRenderData *rd) {
if (!rd->indexed_px_sz) return GRAPHICS_PROGRAM;
return rd->is_unscaled ? GRAPHICS_INDEXED_UNSCALED_PROGRAM : GRAPHICS_INDEXED_PROGRAM;
}
static void
draw_image_layer(const UIRenderData *ui, ssize_t vao_idx, GLuint start, GLuint count) {
if (!ui->grd.num_of_indexed_refs) {
draw_graphics(GRAPHICS_PROGRAM, ui->grd.images, start, count, ui->inactive_text_alpha);
return;
}
CELL_BUFFERS;
// The color table may not be bound yet, for example, when the window has a background image
bind_vao_uniform_buffer(vao_idx, color_table_buffer, COLOR_TABLE_BINDING_POINT);
const ImageRenderData *data = ui->grd.images;
for (GLuint run_start = start, end = start + count; run_start < end;) {
const int program = program_for_image(data + run_start);
GLuint run_end = run_start + 1;
while (run_end < end && program_for_image(data + run_end) == program) run_end++;
draw_graphics(program, ui->grd.images, run_start, run_end - run_start, ui->inactive_text_alpha);
run_start = run_end;
}
}
static void
draw_cells_with_layers(const UIRenderData *ui, ssize_t vao_idx) {
if (ui->has_background_image && OPT(background_tint) > 0) draw_tint(ui);
const bool has_content_between_background_and_foreground = ui->window_logo != NULL || ui->grd.num_of_below_refs > 0 || ui->grd.num_of_negative_refs > 0;
if (has_content_between_background_and_foreground) {
if (!ui->has_background_image) call_cell_program(CELL_BG_PROGRAM, ui, vao_idx, false, DRAW_DEFAULT_BG);
if (ui->window_logo != NULL) draw_window_logo(ui);
if (ui->grd.num_of_below_refs > 0) draw_image_layer(ui, vao_idx, 0, ui->grd.num_of_below_refs);
call_cell_program(CELL_BG_PROGRAM, ui, vao_idx, false, DRAW_NON_DEFAULT_BG);
if (ui->grd.num_of_negative_refs) draw_image_layer(ui, vao_idx, ui->grd.num_of_below_refs, ui->grd.num_of_negative_refs);
call_cell_program(CELL_FG_PROGRAM, ui, vao_idx, false, DRAW_NEITHER_BG);
} else call_cell_program(CELL_PROGRAM, ui, vao_idx, false, ui->has_background_image ? DRAW_NON_DEFAULT_BG : DRAW_BOTH_BG);
if (ui->grd.num_of_positive_refs > 0) draw_image_layer(ui, vao_idx, ui->grd.num_of_below_refs + ui->grd.num_of_negative_refs, ui->grd.num_of_positive_refs);
draw_visual_bell(ui);
draw_drag_preview_overlay(ui);
draw_progress_bar(ui);
draw_scrollbar(ui);
draw_hyperlink_target(ui);
draw_window_number(ui);
}
void
blank_canvas(float background_opacity, color_type color_in_srgb, bool for_final_output) {
if (for_final_output) {
// we need to write pre-multiplied sRGB color to framebuffer
#define C(shift) ((((color_in_srgb >> shift) & 0xFF) / 255.f) * background_opacity)
glClearColor(C(16), C(8), C(0), background_opacity);
#undef C
} else {
// we need to write pre-multiplied linear color to framebuffer
#define C(shift) srgb_color((color_in_srgb >> shift) & 0xFF) * background_opacity
glClearColor(C(16), C(8), C(0), background_opacity);
#undef C
}
glClear(GL_COLOR_BUFFER_BIT);
}
bool
send_cell_data_to_gpu(ssize_t vao_idx, Screen *screen, OSWindow *os_window) {
bool changed = false;
if (os_window->fonts_data) {
if (cell_prepare_to_render(vao_idx, screen, os_window->fonts_data)) changed = true;
}
return changed;
}
void
draw_cells(
const WindowRenderData *srd,
OSWindow *os_window,
bool is_active_window,
bool is_tab_bar,
bool is_single_window,
Window *window,
const Window *title_bar_of,
monotonic_t now) {
Screen *screen = srd->screen;
CELL_BUFFERS;
bind_vertex_array(srd->vao_idx);
// When inactive_text_alpha is negative, its absolute value is used as the
// opacity and fading is based only on whether the current window is active.
const float inactive_text_alpha = fabsf(OPT(inactive_text_alpha));
const bool use_active_window_only = OPT(inactive_text_alpha) < 0.f;
float current_inactive_text_alpha =
use_active_window_only
? (is_tab_bar || is_active_window ? 1.0f : inactive_text_alpha)
: (is_tab_bar || (!is_single_window && is_active_window) || (is_single_window && screen->cursor_render_info.is_focused) ? 1.0f
: inactive_text_alpha);
float bg_alpha = effective_os_window_alpha(os_window);
color_type default_bg = cell_update_uniform_block(
srd->vao_idx, screen, uniform_buffer, color_table_buffer, &screen->cursor_render_info, os_window, current_inactive_text_alpha, bg_alpha);
set_cell_uniforms(screen->reload_all_gpu_data);
WindowLogoRenderData *wl;
if (window && (wl = &window->window_logo) && wl->id && (wl->instance = find_window_logo(global_state.all_window_logos, wl->id)) &&
wl->instance->load_from_disk_ok) {
if (!window->window_logo.instance->texture_id) { set_on_gpu_state(window->window_logo.instance, true); }
} else wl = NULL;
GraphicsManager *grman = screen->paused_rendering.expires_at && screen->paused_rendering.grman ? screen->paused_rendering.grman : screen->grman;
UIRenderData ui = {
.screen_width = srd->geometry.right - srd->geometry.left,
.screen_height = srd->geometry.bottom - srd->geometry.top,
.cell_width = os_window->fonts_data->fcm.cell_width,
.cell_height = os_window->fonts_data->fcm.cell_height,
.screen_left = srd->geometry.left,
.screen_top = srd->geometry.top,
.full_framebuffer_width = os_window->viewport_width,
.full_framebuffer_height = os_window->viewport_height,
.window = window,
.screen = screen,
.os_window = os_window,
.grd = grman_render_data(grman),
.window_logo = wl,
.inactive_text_alpha = current_inactive_text_alpha,
.has_background_image = has_bgimage(os_window),
.background_color = default_bg,
.bg_alpha = effective_os_window_alpha(os_window),
.now = now,
};
screen->reload_all_gpu_data = false;
save_viewport_using_top_left_origin(ui.screen_left, ui.screen_top, ui.screen_width, ui.screen_height, ui.full_framebuffer_height);
if (ui.os_window->needs_layers) draw_cells_with_layers(&ui, srd->vao_idx);
else draw_cells_without_layers(&ui, srd->vao_idx);
restore_viewport();
// The compensatory padding lies outside the per-window cell viewport, so it
// is drawn after restoring the full framebuffer viewport.
if (window) {
draw_window_padding(
&ui,
window->size_mismatch_padding.left,
window->size_mismatch_padding.top,
window->size_mismatch_padding.right,
window->size_mismatch_padding.bottom,
srd->vao_idx,
!ui.os_window->needs_layers);
} else if (title_bar_of) {
// The left and right compensatory padding of a window extends alongside
// its title bar. The layout ensures there is no compensatory padding
// between the title bar and the window border.
draw_window_padding(
&ui, title_bar_of->size_mismatch_padding.left, 0, title_bar_of->size_mismatch_padding.right, 0, srd->vao_idx, !ui.os_window->needs_layers);
}
}
// }}}
// Borders {{{
static void
populate_border_colors(GLuint colors[9], color_type active_window_bg, unsigned int num_visible_windows, bool all_windows_have_same_bg, const OSWindow *w) {
colors[0] = (num_visible_windows > 1 && !all_windows_have_same_bg) ? OPT(background) : active_window_bg;
colors[1] = OPT(active_border_color);
colors[2] = OPT(inactive_border_color);
colors[3] = 0;
colors[4] = OPT(bell_border_color);
colors[5] = OPT(tab_bar_background);
colors[6] = OPT(tab_bar_margin_color);
colors[7] = w->tab_bar_edge_color.left;
colors[8] = w->tab_bar_edge_color.right;
}
ssize_t
create_border_vao(void) {
ssize_t vao_idx = create_vao();
add_buffer_to_vao(vao_idx, GL_ARRAY_BUFFER);
add_attribute_to_vao(
vao_idx,
program_attribute_location(BORDERS_PROGRAM, "rect"),
/*size=*/4,
/*dtype=*/GL_FLOAT,
/*stride=*/sizeof(BorderRect),
/*offset=*/(void *)offsetof(BorderRect, left),
/*divisor=*/1);
add_attribute_to_vao(
vao_idx,
program_attribute_location(BORDERS_PROGRAM, "rect_color"),
/*size=*/1,
/*dtype=*/GL_UNSIGNED_INT,
/*stride=*/sizeof(BorderRect),
/*offset=*/(void *)(offsetof(BorderRect, color)),
/*divisor=*/1);
return vao_idx;
}
void
draw_borders(
ssize_t vao_idx,
unsigned int num_border_rects,
BorderRect *rect_buf,
bool rect_data_is_dirty,
color_type active_window_bg,
unsigned int num_visible_windows,
bool all_windows_have_same_bg,
OSWindow *w) {
float background_opacity = effective_os_window_alpha(w);
if (!num_border_rects) return;
bind_program(BORDERS_PROGRAM);
bind_vertex_array(vao_idx);
const bool has_background_image = has_bgimage(w);
if (has_background_image) background_opacity = OPT(background_tint) * OPT(background_tint_gaps);
if (rect_data_is_dirty) {
const size_t sz = sizeof(BorderRect) * num_border_rects;
void *borders_buf_address = alloc_and_map_vao_buffer(vao_idx, sz, 0);
if (borders_buf_address) memcpy(borders_buf_address, rect_buf, sz);
unmap_vao_buffer(vao_idx, 0);
}
GLuint colors[9];
populate_border_colors(colors, active_window_bg, num_visible_windows, all_windows_have_same_bg, w);
void *colors_buf =
map_vao_buffer_for_write_only(shader_globals_vao_idx, BORDER_COLORS_GLOBAL_BUFFER, 0, program_uniform_block(BORDERS_PROGRAM, "Colors").size);
const ArrayInformation border_colors_array = program_uniform_array(BORDERS_PROGRAM, "colors");
write_uint_array_to_ubo(colors_buf, colors, arraysz(colors), &border_colors_array);
unmap_vao_buffer(shader_globals_vao_idx, BORDER_COLORS_GLOBAL_BUFFER);
glUniform1f(program_uniform_location(BORDERS_PROGRAM, "background_opacity"), background_opacity);
if (!w->needs_layers) glEnable(GL_FRAMEBUFFER_SRGB);
draw_quad(has_background_image, num_border_rects);
if (!w->needs_layers) glDisable(GL_FRAMEBUFFER_SRGB);
unbind_program();
unbind_vertex_array();
}
void
draw_rounded_borders(BorderRects *br, color_type active_window_bg, unsigned int num_visible_windows, bool all_windows_have_same_bg, OSWindow *w) {
if (!br->num_rounded_rects) return;
GLuint colors[9];
populate_border_colors(colors, active_window_bg, num_visible_windows, all_windows_have_same_bg, w);
// The corner regions between the rounded outline and its bounding rect
// must be painted every frame: when not rendering in layers the
// framebuffer is never cleared, kitty relies on full pixel coverage.
// Fill them with the default background at the same opacity draw_borders()
// uses for non-border rects, so corners match the surrounding gaps.
float corner_bg_alpha = has_bgimage(w) ? OPT(background_tint) * OPT(background_tint_gaps) : effective_os_window_alpha(w);
bind_vertex_array(br->vao_idx);
if (!w->needs_layers) glEnable(GL_FRAMEBUFFER_SRGB);
for (unsigned i = 0; i < br->num_border_rects; i++) {
const BorderRect *r = br->rect_buf + i;
if (!r->radius || !r->thickness) continue;
Viewport rect = {.left = r->px.left, .top = r->px.top, .width = r->px.right - r->px.left, .height = r->px.bottom - r->px.top};
unsigned radius = MIN(r->radius, MIN(rect.width, rect.height) / 2);
unsigned color_index = r->color & 0xff;
color_type color = color_index == 3 ? r->color >> 8 : colors[MIN(color_index, 8u)];
draw_rounded_rect(rect, w->viewport_height, (float)r->thickness, radius, color, 0, 0.f, colors[0], corner_bg_alpha);
}
if (!w->needs_layers) glDisable(GL_FRAMEBUFFER_SRGB);
unbind_program();
// Deliberately leave the VAO bound: this runs after all cell drawing and
// subsequent draw_quad() calls (e.g. the fullscreen blit in
// stop_os_window_rendering()) need some VAO bound in a core profile context.
}
// }}}
// Cursor Trail {{{
void
draw_cursor_trail(CursorTrail *trail, Window *active_window) {
bind_program(TRAIL_PROGRAM);
glUniform4fv(program_uniform_location(TRAIL_PROGRAM, "x_coords"), 1, trail->corner_x);
glUniform4fv(program_uniform_location(TRAIL_PROGRAM, "y_coords"), 1, trail->corner_y);
glUniform2fv(program_uniform_location(TRAIL_PROGRAM, "cursor_edge_x"), 1, trail->cursor_edge_x);
glUniform2fv(program_uniform_location(TRAIL_PROGRAM, "cursor_edge_y"), 1, trail->cursor_edge_y);
color_type trail_color = OPT(cursor_trail_color);
if (trail_color == 0) { // 0 means "none" was specified
trail_color = active_window ? active_window->render_data.screen->last_rendered.cursor_bg : OPT(foreground);
}
color_vec3(program_uniform_location(TRAIL_PROGRAM, "trail_color"), trail_color);
glUniform1f(program_uniform_location(TRAIL_PROGRAM, "trail_opacity"), trail->opacity);
draw_quad(true, 0);
unbind_program();
}
// }}}
// OSWindow {{{
static void
draw_bg_image(OSWindow *os_window, Tab *tab) {
BackgroundImage *bg = background_image_for_os_window(os_window);
if (!bg) return;
BackgroundImageLayout layout = os_window->background_image.has_layout ? os_window->background_image.layout : OPT(background_image_layout);
BackgroundImageRenderSettings s = {
.os_window.width = os_window->viewport_width,
.os_window.height = os_window->viewport_height,
.instance_id = bg->id,
.layout = layout,
.linear = OPT(background_image_linear),
.bgcolor = OPT(background),
.opacity = effective_os_window_alpha(os_window),
};
GLfloat iwidth = bg->width, iheight = bg->height;
GLfloat vwidth = s.os_window.width, vheight = s.os_window.height;
if (CENTER_SCALED == layout) {
GLfloat ifrac = iwidth / iheight;
if (ifrac > (vwidth / vheight)) {
iheight = vheight;
iwidth = iheight * ifrac;
} else {
iwidth = vwidth;
iheight = iwidth / ifrac;
}
}
GLfloat tiled = 0.f;
;
GLfloat left = -1.0, top = 1.0, right = 1.0, bottom = -1.0;
switch (layout) {
case TILING:
case MIRRORED:
case CLAMPED: tiled = 1.f; break;
case SCALED: break;
case CENTER_CLAMPED:
case CENTER_SCALED: {
GLfloat wfrac = (vwidth - iwidth) / vwidth;
GLfloat hfrac = (vheight - iheight) / vheight;
left += wfrac;
right -= wfrac;
top -= hfrac;
bottom += hfrac;
} break;
}
bind_program(BGIMAGE_PROGRAM);
// altough we dont use this VO we need to ensure *some* VAO is bound at this point.
bind_vertex_array(tab->border_rects.vao_idx);
glUniform4f(program_uniform_location(BGIMAGE_PROGRAM, "sizes"), vwidth, vheight, iwidth, iheight);
glUniform1f(program_uniform_location(BGIMAGE_PROGRAM, "tiled"), tiled);
glUniform4f(program_uniform_location(BGIMAGE_PROGRAM, "positions"), left, top, right, bottom);
glUniform1i(program_uniform_location(BGIMAGE_PROGRAM, "image"), GRAPHICS_UNIT);
color_vec4(program_uniform_location(BGIMAGE_PROGRAM, "background"), s.bgcolor, s.opacity);
glActiveTexture(GL_TEXTURE0 + GRAPHICS_UNIT);
glBindTexture(GL_TEXTURE_2D, bg->texture_id);
draw_quad(false, 0);
unbind_program();
}
static void
gpu_data_for_centered_image(ImageRenderData *ans, unsigned int screen_width_px, unsigned int screen_height_px, unsigned int width, unsigned int height) {
float width_frac = 2 * MIN(1, width / (float)screen_width_px), height_frac = 2 * MIN(1, height / (float)screen_height_px);
float hmargin = (2 - width_frac) / 2;
float vmargin = (2 - height_frac) / 2;
gpu_data_for_image(ans, -1 + hmargin, 1 - vmargin, -1 + hmargin + width_frac, 1 - vmargin - height_frac);
}
static void
draw_centered_alpha_mask(size_t screen_width, size_t screen_height, size_t width, size_t height, uint8_t *canvas, float background_opacity) {
ImageRenderData *data = load_alpha_mask_texture(width, height, canvas);
gpu_data_for_centered_image(data, screen_width, screen_height, width, height);
bind_program(GRAPHICS_ALPHA_MASK_PROGRAM);
glUniform1i(program_uniform_location(GRAPHICS_ALPHA_MASK_PROGRAM, "image"), GRAPHICS_UNIT);
color_vec3(program_uniform_location(GRAPHICS_ALPHA_MASK_PROGRAM, "amask_fg"), OPT(foreground));
color_vec4_premult(program_uniform_location(GRAPHICS_ALPHA_MASK_PROGRAM, "amask_bg_premult"), OPT(background), background_opacity);
draw_graphics(GRAPHICS_ALPHA_MASK_PROGRAM, data, 0, 1, 1.0);
}
static void
draw_resizing_text(OSWindow *w) {
if (monotonic() - w->created_at > ms_to_monotonic_t(1000) && w->live_resize.num_of_resize_events > 1) {
char text[32] = {0};
unsigned int width = w->live_resize.width, height = w->live_resize.height;
snprintf(text, sizeof(text), "%u by %u cells", width / w->fonts_data->fcm.cell_width, height / w->fonts_data->fcm.cell_height);
StringCanvas rendered = render_simple_text(w->fonts_data, text);
if (rendered.canvas) {
draw_centered_alpha_mask(width, height, rendered.width, rendered.height, rendered.canvas, MAX(0.2f, MIN(OPT(background_opacity), 0.8f)));
free(rendered.canvas);
}
}
}
void
blank_os_window(OSWindow *osw) {
color_type color = OPT(background);
if (osw->num_tabs > 0) {
Tab *t = osw->tabs + osw->active_tab;
if (t->num_windows == 1) {
Window *w = t->windows + t->active_window;
Screen *s = w->render_data.screen;
if (s) { color = colorprofile_to_color(s->color_profile, s->color_profile->overridden.default_bg, s->color_profile->configured.default_bg).rgb; }
}
}
blank_canvas(effective_os_window_alpha(osw), color, true);
}
static void
start_os_window_rendering(OSWindow *os_window, Tab *tab) {
// note that during live resize rendering is done in layers
if (os_window->live_resize.in_progress) blank_os_window(os_window);
if (os_window->needs_layers) {
// Ensure the global shared texture is large enough for this window
if (global_state.layers_render_texture.width < os_window->viewport_width || global_state.layers_render_texture.height < os_window->viewport_height) {
if (global_state.layers_render_texture.texture_id) free_texture(&global_state.layers_render_texture.texture_id);
if (global_state.layers_render_texture.framebuffer_id) free_framebuffer(&global_state.layers_render_texture.framebuffer_id);
if (global_state.layers_render_texture.extra_texture_id) free_texture(&global_state.layers_render_texture.extra_texture_id);
if (global_state.layers_render_texture.extra_texture_setup_fbo_id) free_framebuffer(&global_state.layers_render_texture.extra_texture_setup_fbo_id);
if (global_state.layers_render_texture.texture_a_id) free_texture(&global_state.layers_render_texture.texture_a_id);
if (global_state.layers_render_texture.texture_a_fbo_id) free_framebuffer(&global_state.layers_render_texture.texture_a_fbo_id);
if (global_state.layers_render_texture.texture_b_id) free_texture(&global_state.layers_render_texture.texture_b_id);
if (global_state.layers_render_texture.texture_b_fbo_id) free_framebuffer(&global_state.layers_render_texture.texture_b_fbo_id);
unsigned new_w = (unsigned)MAX(global_state.layers_render_texture.width, os_window->viewport_width);
unsigned new_h = (unsigned)MAX(global_state.layers_render_texture.height, os_window->viewport_height);
setup_texture_as_render_target(new_w, new_h, &global_state.layers_render_texture.texture_id, &global_state.layers_render_texture.framebuffer_id);
global_state.layers_render_texture.width = (int)new_w;
global_state.layers_render_texture.height = (int)new_h;
global_state.layers_render_texture.texture_generation++;
}
// Create per-window framebuffer if needed and attach the global texture to it
if (!os_window->indirect_output.framebuffer_id) glGenFramebuffers(1, &os_window->indirect_output.framebuffer_id);
if (os_window->indirect_output.attached_texture_generation != global_state.layers_render_texture.texture_generation) {
bind_framebuffer_for_output(os_window->indirect_output.framebuffer_id);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_2D, global_state.layers_render_texture.texture_id, 0);
os_window->indirect_output.attached_texture_generation = global_state.layers_render_texture.texture_generation;
}
if (custom_shaders.end.num_groups > 1) {
// Lazily create the global shared extra texture (ping-pong target for multi-group end shaders)
if (!global_state.layers_render_texture.extra_texture_id) {
setup_texture_as_render_target(
(unsigned)global_state.layers_render_texture.width,
(unsigned)global_state.layers_render_texture.height,
&global_state.layers_render_texture.extra_texture_id,
&global_state.layers_render_texture.extra_texture_setup_fbo_id);
}
// Create/re-attach per-window extra FBO when generation changes
if (os_window->indirect_output.extra_fbo_generation != global_state.layers_render_texture.texture_generation) {
if (os_window->indirect_output.extra_fbo_id) free_framebuffer(&os_window->indirect_output.extra_fbo_id);
glGenFramebuffers(1, &os_window->indirect_output.extra_fbo_id);
bind_framebuffer_for_output(os_window->indirect_output.extra_fbo_id);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_2D, global_state.layers_render_texture.extra_texture_id, 0);
os_window->indirect_output.extra_fbo_generation = global_state.layers_render_texture.texture_generation;
}
}
// Named texture A: global shared, per-window FBO
if (custom_shaders.end.textures & (1u << CUSTOM_SHADER_TEXTURE_A)) {
if (!global_state.layers_render_texture.texture_a_id) {
setup_texture_as_render_target(
(unsigned)global_state.layers_render_texture.width,
(unsigned)global_state.layers_render_texture.height,
&global_state.layers_render_texture.texture_a_id,
&global_state.layers_render_texture.texture_a_fbo_id);
}
if (os_window->indirect_output.fbo_a_generation != global_state.layers_render_texture.texture_generation) {
if (os_window->indirect_output.fbo_a_id) free_framebuffer(&os_window->indirect_output.fbo_a_id);
glGenFramebuffers(1, &os_window->indirect_output.fbo_a_id);
bind_framebuffer_for_output(os_window->indirect_output.fbo_a_id);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_2D, global_state.layers_render_texture.texture_a_id, 0);
os_window->indirect_output.fbo_a_generation = global_state.layers_render_texture.texture_generation;
}
} else {
if (os_window->indirect_output.fbo_a_id) free_framebuffer(&os_window->indirect_output.fbo_a_id);
}
// Named texture B: global shared, per-window FBO
if (custom_shaders.end.textures & (1u << CUSTOM_SHADER_TEXTURE_B)) {
if (!global_state.layers_render_texture.texture_b_id) {
setup_texture_as_render_target(
(unsigned)global_state.layers_render_texture.width,
(unsigned)global_state.layers_render_texture.height,
&global_state.layers_render_texture.texture_b_id,
&global_state.layers_render_texture.texture_b_fbo_id);
}
if (os_window->indirect_output.fbo_b_generation != global_state.layers_render_texture.texture_generation) {
if (os_window->indirect_output.fbo_b_id) free_framebuffer(&os_window->indirect_output.fbo_b_id);
glGenFramebuffers(1, &os_window->indirect_output.fbo_b_id);
bind_framebuffer_for_output(os_window->indirect_output.fbo_b_id);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_2D, global_state.layers_render_texture.texture_b_id, 0);
os_window->indirect_output.fbo_b_generation = global_state.layers_render_texture.texture_generation;
}
} else {
if (os_window->indirect_output.fbo_b_id) free_framebuffer(&os_window->indirect_output.fbo_b_id);
}
// Named texture persist: per-window, recreated when global texture dimensions change
if (custom_shaders.end.textures & (1u << CUSTOM_SHADER_TEXTURE_PERSIST)) {
if (os_window->persist_texture_generation != global_state.layers_render_texture.texture_generation) {
if (os_window->persist_texture_id) free_texture(&os_window->persist_texture_id);
if (os_window->persist_fbo_id) free_framebuffer(&os_window->persist_fbo_id);
setup_texture_as_render_target(
(unsigned)global_state.layers_render_texture.width,
(unsigned)global_state.layers_render_texture.height,
&os_window->persist_texture_id,
&os_window->persist_fbo_id);
os_window->persist_texture_generation = global_state.layers_render_texture.texture_generation;
}
} else {
if (os_window->persist_texture_id) {
free_texture(&os_window->persist_texture_id);
os_window->persist_texture_generation = 0;
}
if (os_window->persist_fbo_id) free_framebuffer(&os_window->persist_fbo_id);
}
set_framebuffer_to_use_for_output(os_window->indirect_output.framebuffer_id);
bind_framebuffer_for_output(0);
save_viewport_using_bottom_left_origin(0, 0, os_window->viewport_width, os_window->viewport_height);
clear_current_framebuffer();
draw_bg_image(os_window, tab);
}
}
// Bilinear lookup of the position u, v (viewport UV, clamped to [0, 1]) in a
// pointer map, replacing it with the position of the content shown there
static void
sample_pointer_map(const float *data, double *u, double *v) {
const double last = POINTER_MAP_SIZE - 1;
const double fx = MAX(0., MIN(*u, 1.)) * last, fy = MAX(0., MIN(*v, 1.)) * last;
const unsigned x0 = MIN((unsigned)fx, POINTER_MAP_SIZE - 2), y0 = MIN((unsigned)fy, POINTER_MAP_SIZE - 2);
const double tx = fx - x0, ty = fy - y0;
#define P(x, y, c) ((double)data[2 * ((y) * POINTER_MAP_SIZE + (x)) + (c)])
#define L(c) ((P(x0, y0, c) * (1 - tx) + P(x0 + 1, y0, c) * tx) * (1 - ty) + (P(x0, y0 + 1, c) * (1 - tx) + P(x0 + 1, y0 + 1, c) * tx) * ty)
*u = L(0);
*v = L(1);
#undef L
#undef P
}
// The groups whose pointer_map() functions currently apply: those that are
// always active and those turned on by events that are active right now
static uint32_t
active_pointer_map_groups(const OSWindow *os_window) {
uint32_t ans = 0;
for (size_t g = 0; g < custom_shaders.end.num_groups; g++) {
if (!(custom_shaders.end.pointer_map_groups & (1u << g))) continue;
if (!group_is_animated(custom_shaders.end.groups + g) || os_window->shader_group_anim[g].active) ans |= 1u << g;
}
return ans;
}
// Moves a pointer position, in viewport pixels with the origin at the top left,
// to the position of the content that custom shaders such as crt actually show
// there, so that clicks and selections land on what the user sees.
void
map_pointer_position(const OSWindow *os_window, double *x, double *y) {
const float *data = os_window->pointer_map.data;
if (!data || !custom_shaders.end.has_pointer_map || os_window->pointer_map.pipeline_generation != custom_shaders.generation) return;
// The map is updated only when the custom shaders are drawn, which they
// are not once every group has been turned off, so check that it is
// still for the groups that are active
const uint32_t groups = active_pointer_map_groups(os_window);
if (!groups || groups != os_window->pointer_map.groups) return;
const double width = os_window->viewport_width, height = os_window->viewport_height;
if (width <= 0 || height <= 0) return;
const double u = *x / width, v = 1. - *y / height;
const double cu = MAX(0., MIN(u, 1.)), cv = MAX(0., MIN(v, 1.));
double mu = cu, mv = cv;
sample_pointer_map(data, &mu, &mv);
// Positions outside the viewport, such as when drag selecting past its
// edge, move along with the nearest edge
*x = (mu + u - cu) * width;
*y = (1. - (mv + v - cv)) * height;
}
// Renders the pointer_map() functions of the pipeline into a small texture
// and reads it back so that pointer events can be mapped without the GPU. A
// pointer map can depend only on the pipeline, the viewport size and which
// groups are active, so this is done only when those change.
static void
update_pointer_map(OSWindow *os_window, GLint group_loc) {
const uint32_t groups = active_pointer_map_groups(os_window);
if (os_window->pointer_map.pipeline_generation == custom_shaders.generation && os_window->pointer_map.groups == groups &&
(os_window->pointer_map.data || !groups)) {
if (os_window->live_resize.in_progress) return; // avoid a GPU sync on every frame of a live resize
if (os_window->pointer_map.viewport_width == os_window->viewport_width && os_window->pointer_map.viewport_height == os_window->viewport_height) return;
}
os_window->pointer_map.viewport_width = os_window->viewport_width;
os_window->pointer_map.viewport_height = os_window->viewport_height;
os_window->pointer_map.pipeline_generation = custom_shaders.generation;
os_window->pointer_map.groups = 0;
// No pointer_map() applies, such as when content moving groups are toggled off, so pointer events are not moved
if (!groups) return;
if (!os_window->pointer_map.data && !(os_window->pointer_map.data = malloc(sizeof(float) * 2 * POINTER_MAP_SIZE * POINTER_MAP_SIZE))) return;
if (!pointer_map_target.framebuffer_id)
setup_texture_as_render_target(POINTER_MAP_SIZE, POINTER_MAP_SIZE, &pointer_map_target.texture_id, &pointer_map_target.framebuffer_id);
// The pointer map does not sample any textures, but every sampler must be bound to a valid one
static const int units[] = {GRAPHICS_UNIT, CUSTOM_END_TEXTURE_A_UNIT, CUSTOM_END_TEXTURE_B_UNIT, CUSTOM_END_TEXTURE_PERSIST_UNIT};
for (size_t i = 0; i < arraysz(units); i++) {
glActiveTexture(GL_TEXTURE0 + units[i]);
glBindTexture(GL_TEXTURE_2D, global_state.layers_render_texture.texture_id);
}
bind_framebuffer_for_output(pointer_map_target.framebuffer_id);
glViewport(0, 0, POINTER_MAP_SIZE, POINTER_MAP_SIZE);
glUniform1i(group_loc, -1);
// Absent when no group with a pointer_map() is turned on and off by events
GLint active_groups_loc = try_program_uniform_location(CUSTOM_END_PROGRAM, "active_groups");
if (active_groups_loc >= 0) glUniform1ui(active_groups_loc, groups);
draw_quad(false, 0);
static uint8_t pixels[POINTER_MAP_SIZE * POINTER_MAP_SIZE * 4];
glReadPixels(0, 0, POINTER_MAP_SIZE, POINTER_MAP_SIZE, GL_RGBA, GL_UNSIGNED_BYTE, pixels);
// See encode_pointer_map() in pipeline.slang
for (size_t i = 0; i < POINTER_MAP_SIZE * POINTER_MAP_SIZE; i++) {
const uint8_t *p = pixels + 4 * i;
os_window->pointer_map.data[2 * i] = (float)(p[0] * 256 + p[2]) / 65535.f;
os_window->pointer_map.data[2 * i + 1] = (float)(p[1] * 256 + p[3]) / 65535.f;
}
os_window->pointer_map.groups = groups;
}
static void
free_pointer_map_target(void) {
if (pointer_map_target.texture_id) free_texture(&pointer_map_target.texture_id);
if (pointer_map_target.framebuffer_id) free_framebuffer(&pointer_map_target.framebuffer_id);
}
static void
run_custom_end_shader(OSWindow *os_window, float sx, float sy, monotonic_t now) {
bind_program(CUSTOM_END_PROGRAM);
struct GPUCustomEndData {
float src_rect[4];
float dest_rect[4];
float background[4];
float foreground[4];
float active_window_background[4];
float mouse_pos[4];
float mouse_button_pressed[4];
float active_window_geometry[4];
float active_window_padding[4];
float bell_window_geometry[4];
float central_area[4];
float cursor_trail_corners_x[4];
float cursor_trail_corners_y[4];
float cursor_trail_prev_corners_x[4];
float cursor_trail_prev_corners_y[4];
float cursor_trail_edge[4];
float cursor_trail_prev_edge[4];
float cursor_color[4];
float cursor_trail_color[4];
float cursor_trail_history_from[CURSOR_TRAIL_HISTORY_SIZE][4];
float cursor_trail_history_to[CURSOR_TRAIL_HISTORY_SIZE][4];
float cursor_trail_history_age[CURSOR_TRAIL_HISTORY_SIZE / 4][4];
uint32_t viewport_size_pixels[2];
float mouse_pointer_hidden;
float cursor_trail_state;
float timestamp, last_rendered_at;
uint32_t frame_counter;
float cursor_trail_change_time;
uint32_t timestamp_seconds;
float timestamp_fraction;
float time_since_last_render;
float cursor_trail_age;
float active_window_alternate_screen;
float active_window_command_running;
float active_window_alternate_screen_age;
float active_window_command_state_age;
};
// This struct is a hand written mirror of the std140 layout of
// KittyCustomShaderData in types.slang, so check that they match once.
const UniformBlock data_block = program_uniform_block(CUSTOM_END_PROGRAM, "KittyCustomShaderData");
static bool data_block_size_checked = false;
if (!data_block_size_checked) {
data_block_size_checked = true;
if ((size_t)data_block.size < sizeof(struct GPUCustomEndData))
fatal(
"The KittyCustomShaderData uniform block is %d bytes but the C struct mirroring it is %zu bytes",
data_block.size,
sizeof(struct GPUCustomEndData));
}
struct GPUCustomEndData *d = (struct GPUCustomEndData *)map_vao_buffer_for_write_only(custom_end_vao_idx, 0, 0, data_block.size);
memset(d, 0, sizeof(struct GPUCustomEndData));
d->src_rect[1] = sy;
d->src_rect[2] = sx;
d->dest_rect[0] = -1;
d->dest_rect[1] = 1;
d->dest_rect[2] = 1;
d->dest_rect[3] = -1;
#define FILL_COLOR(dst, c) \
do { \
(dst)[0] = srgb_color(((c) >> 16) & 0xFF); \
(dst)[1] = srgb_color(((c) >> 8) & 0xFF); \
(dst)[2] = srgb_color((c) & 0xFF); \
(dst)[3] = 1.f; \
} while (0)
FILL_COLOR(d->background, OPT(background));
FILL_COLOR(d->foreground, OPT(foreground));
color_type active_bg = OPT(background);
color_type cursor_color = OPT(foreground);
float cursor_opacity = 0.f;
WindowGeometry active_win_geom = {0};
d->active_window_alternate_screen_age = -1.f;
d->active_window_command_state_age = -1.f;
if (os_window->num_tabs > 0) {
Tab *t = os_window->tabs + os_window->active_tab;
if (t->num_windows) {
Window *w = t->windows + t->active_window;
active_win_geom = w->render_data.geometry;
if (os_window->viewport_width && os_window->viewport_height) {
const float width = (float)os_window->viewport_width, height = (float)os_window->viewport_height;
// spaces is all the space the layout reserved around the window:
// margin, border (only if the layout reserves it), padding and
// cell-size compensation.
const WindowGeometry *sg = &w->render_data.geometry;
d->active_window_padding[0] = (float)sg->spaces.left / width;
d->active_window_padding[1] = (float)sg->spaces.bottom / height;
d->active_window_padding[2] = (float)sg->spaces.right / width;
d->active_window_padding[3] = (float)sg->spaces.top / height;
// The content geometry excludes the title bar. Include it before
// extending to the outer padding, so the effect boundary cannot
// cut through a title bar at either edge of the window.
const WindowRenderData *trd = &w->window_title_render_data;
if (trd->screen && trd->geometry.right > trd->geometry.left && trd->geometry.bottom > trd->geometry.top) {
if (trd->geometry.top < active_win_geom.top) d->active_window_padding[3] += (float)(active_win_geom.top - trd->geometry.top) / height;
if (trd->geometry.bottom > active_win_geom.bottom)
d->active_window_padding[1] += (float)(trd->geometry.bottom - active_win_geom.bottom) / height;
}
}
Screen *s = w->render_data.screen;
if (s) {
active_bg = colorprofile_to_color(s->color_profile, s->color_profile->overridden.default_bg, s->color_profile->configured.default_bg).rgb;
cursor_color = s->last_rendered.cursor_bg;
cursor_opacity = MAX(0.f, MIN(s->cursor_render_info.cursor_opacity, 1.f));
d->active_window_alternate_screen = s->linebuf == s->alt_linebuf ? 1.f : 0.f;
d->active_window_command_running = s->command_state.running ? 1.f : 0.f;
const monotonic_t alt_changed_at = MAX(s->alternate_screen.entered_at, s->alternate_screen.exited_at);
if (alt_changed_at) d->active_window_alternate_screen_age = (float)monotonic_t_to_s_double(now - alt_changed_at);
const monotonic_t cmd_changed_at = MAX(s->command_state.started_at, s->command_state.finished_at);
if (cmd_changed_at) d->active_window_command_state_age = (float)monotonic_t_to_s_double(now - cmd_changed_at);
}
}
}
FILL_COLOR(d->active_window_background, active_bg);
#undef FILL_COLOR
#define NDC_TO_UV(v) (((v) + 1.0f) * 0.5f)
// Convert to seconds via double, as float32 cannot represent times since
// kitty started precisely once kitty has been running for a few hours.
#define SECONDS(t) ((float)monotonic_t_to_s_double(t))
{
CursorTrail *ct = NULL;
d->cursor_trail_age = -1.f;
for (int i = 0; i < CURSOR_TRAIL_HISTORY_SIZE; i++) d->cursor_trail_history_age[i / 4][i % 4] = -1.f;
if (OPT(cursor_trail) && os_window->num_tabs > 0) {
CursorTrail *candidate = &os_window->tabs[os_window->active_tab].cursor_trail;
if (candidate->needs_render) ct = candidate;
// The history outlives the trail so that effects started by a move can finish
for (int i = 0; i < CURSOR_TRAIL_HISTORY_SIZE; i++) {
const CursorMove *m = candidate->history + i;
if (!m->at) break;
for (int k = 0; k < 2; k++) {
d->cursor_trail_history_from[i][k] = NDC_TO_UV(m->from_x[k]);
d->cursor_trail_history_from[i][k + 2] = NDC_TO_UV(m->from_y[k]);
d->cursor_trail_history_to[i][k] = NDC_TO_UV(m->to_x[k]);
d->cursor_trail_history_to[i][k + 2] = NDC_TO_UV(m->to_y[k]);
}
d->cursor_trail_history_age[i / 4][i % 4] = SECONDS(now - m->at);
}
}
if (ct) {
for (int i = 0; i < 4; i++) {
d->cursor_trail_corners_x[i] = NDC_TO_UV(ct->corner_x[i]);
d->cursor_trail_corners_y[i] = NDC_TO_UV(ct->corner_y[i]);
d->cursor_trail_prev_corners_x[i] = NDC_TO_UV(ct->previous_corner_x[i]);
d->cursor_trail_prev_corners_y[i] = NDC_TO_UV(ct->previous_corner_y[i]);
}
d->cursor_trail_edge[0] = NDC_TO_UV(ct->cursor_edge_x[0]); // left
d->cursor_trail_edge[1] = NDC_TO_UV(ct->cursor_edge_x[1]); // right
d->cursor_trail_edge[2] = NDC_TO_UV(ct->cursor_edge_y[0]); // top
d->cursor_trail_edge[3] = NDC_TO_UV(ct->cursor_edge_y[1]); // bottom
if (ct->prev_edge_valid) {
d->cursor_trail_prev_edge[0] = NDC_TO_UV(ct->prev_cursor_edge_x[0]); // left
d->cursor_trail_prev_edge[1] = NDC_TO_UV(ct->prev_cursor_edge_x[1]); // right
d->cursor_trail_prev_edge[2] = NDC_TO_UV(ct->prev_cursor_edge_y[0]); // top
d->cursor_trail_prev_edge[3] = NDC_TO_UV(ct->prev_cursor_edge_y[1]); // bottom
}
d->cursor_trail_color[3] = ct->opacity;
d->cursor_trail_state = 1.0f;
if (ct->cursor_changed_at) {
d->cursor_trail_change_time = SECONDS(ct->cursor_changed_at);
d->cursor_trail_age = SECONDS(now - ct->cursor_changed_at);
}
}
}
#undef NDC_TO_UV
#define FILL_COLOR3(dst, c) \
do { \
(dst)[0] = srgb_color(((c) >> 16) & 0xFF); \
(dst)[1] = srgb_color(((c) >> 8) & 0xFF); \
(dst)[2] = srgb_color((c) & 0xFF); \
} while (0)
FILL_COLOR3(d->cursor_color, cursor_color);
d->cursor_color[3] = cursor_opacity;
{
color_type trail_color = OPT(cursor_trail_color);
if (trail_color == 0) trail_color = cursor_color;
FILL_COLOR3(d->cursor_trail_color, trail_color);
}
#undef FILL_COLOR3
d->viewport_size_pixels[0] = (uint32_t)os_window->viewport_width;
d->viewport_size_pixels[1] = (uint32_t)os_window->viewport_height;
const float vpw = (float)os_window->viewport_width, vph = (float)os_window->viewport_height;
if (vpw > 0 && vph > 0) {
d->mouse_pos[0] = (float)os_window->mouse_x / vpw;
d->mouse_pos[1] = 1.f - (float)os_window->mouse_y / vph;
d->mouse_pos[2] = (float)os_window->mouse_left_press_x / vpw;
d->mouse_pos[3] = 1.f - (float)os_window->mouse_left_press_y / vph;
}
d->mouse_button_pressed[0] = os_window->mouse_button_pressed[GLFW_MOUSE_BUTTON_LEFT];
d->mouse_button_pressed[1] = os_window->mouse_button_pressed[GLFW_MOUSE_BUTTON_RIGHT];
d->mouse_button_pressed[2] = os_window->mouse_button_pressed[GLFW_MOUSE_BUTTON_MIDDLE];
d->mouse_button_pressed[3] = os_window->mouse_button_pressed[3];
d->mouse_pointer_hidden = is_mouse_hidden(os_window) ? 1.f : 0.f;
d->timestamp = SECONDS(now);
d->last_rendered_at = SECONDS(os_window->last_rendered_at);
d->timestamp_seconds = (uint32_t)(now / MONOTONIC_T_1e9);
d->timestamp_fraction = (float)monotonic_t_to_s_double(now % MONOTONIC_T_1e9);
if (os_window->last_rendered_at) d->time_since_last_render = SECONDS(now - os_window->last_rendered_at);
#undef SECONDS
d->frame_counter = os_window->frame_counter;
if (vpw > 0 && vph > 0 && active_win_geom.right > active_win_geom.left) {
d->active_window_geometry[0] = (float)active_win_geom.left / vpw;
d->active_window_geometry[1] = 1.f - (float)active_win_geom.bottom / vph;
d->active_window_geometry[2] = (float)(active_win_geom.right - active_win_geom.left) / vpw;
d->active_window_geometry[3] = (float)(active_win_geom.bottom - active_win_geom.top) / vph;
}
if (vpw > 0 && vph > 0) {
Region central_rgn = {0}, tab_bar_rgn = {0};
os_window_regions(os_window, &central_rgn, &tab_bar_rgn);
d->central_area[0] = (float)central_rgn.left / vpw;
d->central_area[1] = 1.f - (float)central_rgn.bottom / vph;
d->central_area[2] = (float)(central_rgn.right - central_rgn.left) / vpw;
d->central_area[3] = (float)(central_rgn.bottom - central_rgn.top) / vph;
} else {
d->central_area[2] = 1.f;
d->central_area[3] = 1.f;
}
if (os_window->last_bell_window_id && os_window->num_tabs > 0 && vpw > 0 && vph > 0) {
WindowGeometry bell_win_geom = {0};
bool bell_win_found = false;
Tab *bt = os_window->tabs + os_window->active_tab;
for (unsigned i = 0; i < bt->num_windows; i++) {
Window *bw = bt->windows + i;
if (bw->id == os_window->last_bell_window_id && bw->visible) {
bell_win_geom = bw->render_data.geometry;
bell_win_found = true;
break;
}
}
if (bell_win_found && bell_win_geom.right > bell_win_geom.left) {
d->bell_window_geometry[0] = (float)bell_win_geom.left / vpw;
d->bell_window_geometry[1] = 1.f - (float)bell_win_geom.bottom / vph;
d->bell_window_geometry[2] = (float)(bell_win_geom.right - bell_win_geom.left) / vpw;
d->bell_window_geometry[3] = (float)(bell_win_geom.bottom - bell_win_geom.top) / vph;
} else {
memcpy(d->bell_window_geometry, d->central_area, sizeof(d->bell_window_geometry));
}
}
unmap_vao_buffer(custom_end_vao_idx, 0);
bind_vao_uniform_buffer(custom_end_vao_idx, 0, CUSTOM_END_DATA_BINDING_POINT);
// A custom shader need not reference any of these, in which case the driver
// drops them and glUniform*() on the resulting -1 location is a no-op.
GLint group_loc = try_program_uniform_location(CUSTOM_END_PROGRAM, "group");
GLint viewport_loc = try_program_uniform_location(CUSTOM_END_PROGRAM, "viewport");
GLint anim_progress_loc = try_program_uniform_location(CUSTOM_END_PROGRAM, "animation_progress");
GLint convert_to_srgb_loc = try_program_uniform_location(CUSTOM_END_PROGRAM, "convert_to_srgb");
if (custom_shaders.end.has_pointer_map && group_loc >= 0) update_pointer_map(os_window, group_loc);
const unsigned num_groups = (unsigned)custom_shaders.end.num_groups;
const unsigned textures_mask = custom_shaders.end.textures;
const int vw = os_window->viewport_width, vh = os_window->viewport_height;
// Named texture IDs and their render-target FBOs, indexed by NamedTexture enum value
const GLuint named_tex_ids[4] = {
0, // DEFAULT — unused
global_state.layers_render_texture.texture_a_id,
global_state.layers_render_texture.texture_b_id,
os_window->persist_texture_id,
};
const GLuint named_tex_fbos[4] = {
0, // DEFAULT — unused
os_window->indirect_output.fbo_a_id,
os_window->indirect_output.fbo_b_id,
os_window->persist_fbo_id,
};
static const int named_tex_units[4] = {
GRAPHICS_UNIT, // DEFAULT maps to backbuffer slot
CUSTOM_END_TEXTURE_A_UNIT,
CUSTOM_END_TEXTURE_B_UNIT,
CUSTOM_END_TEXTURE_PERSIST_UNIT,
};
// Pre-scan to find the last group that will actually run this frame.
// That group gets convert_to_srgb=true so its draw call replaces the
// blit pass; all other groups write premultiplied linear RGB and skip it.
int last_active_g = -1;
for (unsigned g = 0; g < num_groups; g++) {
const CustomShaderGroup *cg = &custom_shaders.end.groups[g];
// Non-attached always-active groups always run; attached and animated groups run only when active.
if (!group_is_animated(cg) || os_window->shader_group_anim[g].active) last_active_g = (int)g;
}
// Ping-pong state: when true, backbuffer = texture_id and the ping-pong
// target is extra_texture_id; toggled each time a group outputs to DEFAULT.
bool backbuffer_is_main = true;
float anim_progress[MAX_CUSTOM_SHADER_GROUPS];
custom_shader_group_progress(os_window, now, anim_progress);
for (unsigned g = 0; g < num_groups; g++) {
const CustomShaderGroup *cg = &custom_shaders.end.groups[g];
const bool is_last = (last_active_g >= 0 && g == (unsigned)last_active_g);
// Skip inactive groups. Non-attached, non-animated groups always run; all others require
// shader_group_anim[g].active (set by update_custom_shader_animations, including attach logic).
if (group_is_animated(cg) && !os_window->shader_group_anim[g].active) continue;
const GLuint backbuffer = backbuffer_is_main ? global_state.layers_render_texture.texture_id : global_state.layers_render_texture.extra_texture_id;
// Bind backbuffer to the backbuffer texture unit
glActiveTexture(GL_TEXTURE0 + GRAPHICS_UNIT);
glBindTexture(GL_TEXTURE_2D, backbuffer);
// Bind each named texture to its own unit.
// If the texture is the output for this group, or is not declared in the
// pipeline, fall back to backbuffer so the sampler is always valid.
for (int t = CUSTOM_SHADER_TEXTURE_A; t <= CUSTOM_SHADER_TEXTURE_PERSIST; t++) {
GLuint tex;
if (!is_last && cg->output_texture == (NamedTexture)t) {
tex = backbuffer; // can't read from the render target we're writing to
} else if (textures_mask & (1u << t)) {
tex = named_tex_ids[t];
} else {
tex = backbuffer; // not present — fall back to backbuffer
}
glActiveTexture(GL_TEXTURE0 + named_tex_units[t]);
glBindTexture(GL_TEXTURE_2D, tex);
}
// Set the output framebuffer
if (is_last) {
set_framebuffer_to_use_for_output(0);
bind_framebuffer_for_output(0);
} else if (cg->output_texture != CUSTOM_SHADER_TEXTURE_DEFAULT) {
// Output to named texture; backbuffer is unchanged for the next group
bind_framebuffer_for_output(named_tex_fbos[cg->output_texture]);
} else {
// Ping-pong: output to the other texture, which becomes the new backbuffer
bind_framebuffer_for_output(backbuffer_is_main ? os_window->indirect_output.extra_fbo_id : os_window->indirect_output.framebuffer_id);
backbuffer_is_main = !backbuffer_is_main;
}
// Set the viewport
float vp_x, vp_y, vp_w, vp_h;
if (is_last) {
// Last group always covers the full screen region
if (os_window->live_resize.in_progress) {
glViewport(0, os_window->live_resize.height - vh, vw, vh);
} else {
glViewport(0, 0, vw, vh);
}
vp_x = 0.f;
vp_y = 0.f;
vp_w = 1.f;
vp_h = 1.f;
} else {
float px = cg->viewport_pos.x, py = cg->viewport_pos.y;
float sw = cg->viewport_size.w, sh = cg->viewport_size.h;
// viewport_pos/size are standard UV (origin at bottom-left, y increasing
// upward), which matches GL pixel coords directly — no y-flip needed.
GLint x0 = MAX(0, (GLint)(px * vw));
GLint y0 = MAX(0, (GLint)(py * vh));
GLint x1 = MIN(vw, (GLint)((px + sw) * vw));
GLint y1 = MIN(vh, (GLint)((py + sh) * vh));
glViewport(x0, y0, MAX(0, x1 - x0), MAX(0, y1 - y0));
vp_x = px;
vp_y = py;
vp_w = sw;
vp_h = sh;
}
glUniform1i(group_loc, (GLint)g);
glUniform4f(viewport_loc, vp_x, vp_y, vp_w, vp_h);
glUniform1f(anim_progress_loc, anim_progress[g]);
glUniform1i(convert_to_srgb_loc, is_last ? 1 : 0);
// debug_rendering("Custom shader draw group %u: anim_progress=%.4f viewport=(%.2f,%.2f,%.2f,%.2f)\n", g, anim_progress[g], vp_x, vp_y, vp_w, vp_h);
draw_quad(false, 0);
}
}
static void
stop_os_window_rendering(OSWindow *os_window, Tab *tab, Window *active_window, monotonic_t now) {
if (OPT(cursor_trail) && tab->cursor_trail.needs_render && !(custom_shaders.end.active && custom_shaders.end.bypass_builtin_trail))
draw_cursor_trail(&tab->cursor_trail, active_window);
if (os_window->needs_layers) {
float sx = global_state.layers_render_texture.width > 0 ? (float)os_window->viewport_width / (float)global_state.layers_render_texture.width : 1.f;
float sy = global_state.layers_render_texture.height > 0 ? (float)os_window->viewport_height / (float)global_state.layers_render_texture.height : 1.f;
if (custom_shaders.end.active && os_window->shader_anim.has_active_shaders) {
// debug_rendering("Custom end shader running (groups=%zu)\n", custom_shaders.end.num_groups);
restore_viewport();
if (os_window->live_resize.in_progress)
save_viewport_using_top_left_origin(0, 0, os_window->viewport_width, os_window->viewport_height, os_window->live_resize.height);
run_custom_end_shader(os_window, sx, sy, now);
} else {
// debug_rendering("Custom shaders inactive, falling back to blit\n");
set_framebuffer_to_use_for_output(0);
bind_framebuffer_for_output(0);
bind_program(BLIT_PROGRAM);
glActiveTexture(GL_TEXTURE0 + GRAPHICS_UNIT);
glBindTexture(GL_TEXTURE_2D, global_state.layers_render_texture.texture_id);
glUniform4f(program_uniform_location(BLIT_PROGRAM, "src_rect"), 0, sy, sx, 0);
glUniform4f(program_uniform_location(BLIT_PROGRAM, "dest_rect"), -1, 1, 1, -1);
restore_viewport();
if (os_window->live_resize.in_progress)
save_viewport_using_top_left_origin(0, 0, os_window->viewport_width, os_window->viewport_height, os_window->live_resize.height);
draw_quad(false, 0);
}
if (os_window->live_resize.in_progress) {
restore_viewport();
draw_resizing_text(os_window);
}
}
os_window->last_rendered_at = now;
os_window->frame_counter++;
}
void
setup_os_window_for_rendering(OSWindow *os_window, Tab *tab, Window *active_window, bool start, monotonic_t now) {
if (start) start_os_window_rendering(os_window, tab);
else stop_os_window_rendering(os_window, tab, active_window, now);
}
// Take a screenshot of the OS Window, must be called immediately after
// the OSWindow is rendered into the back buffer and before the buffers
// are swapped. If thumb_w or thumb_h are zero the are set to the corresponding
// dimension of the source region (viewport or central region without tab bar).
// Takes a screenshot of a rectangular region of the OSWindow's framebuffer.
// The region parameter specifies which part of the framebuffer to capture.
// Scaling is performed on the GPU using the SCREENSHOT_PROGRAM shader for better performance.
// The shader properly handles sRGB color space conversion and downscaling.
// Setting the thumbnail dimensions to zero disables scaling.
// If no_scaling is true, thumb_w/thumb_h are ignored and the region is read
// back directly from the framebuffer with no GPU resampling, for a pixel
// perfect capture (the output is in GL's native bottom-up row order, unlike
// the scaled path below which flips to top-down).
void
take_screenshot_of_rectangular_region(OSWindow *os_window, Region region, unsigned char *dst_buf, unsigned *thumb_w, unsigned *thumb_h, bool no_scaling) {
unsigned vh = os_window->viewport_height;
// Calculate the source region dimensions
unsigned src_height = region.bottom - region.top;
unsigned src_width = region.right - region.left;
if (no_scaling) {
*thumb_w = src_width;
*thumb_h = src_height;
glPixelStorei(GL_PACK_ALIGNMENT, 1);
glReadPixels((GLint)region.left, (GLint)(vh - region.bottom), (GLsizei)src_width, (GLsizei)src_height, GL_RGBA, GL_UNSIGNED_BYTE, dst_buf);
glPixelStorei(GL_PACK_ALIGNMENT, 4);
return;
}
unsigned vw = os_window->viewport_width;
if (!*thumb_w) *thumb_w = src_width;
if (!*thumb_h) *thumb_h = src_height;
*thumb_w = MIN(src_width, *thumb_w);
*thumb_h = MIN(src_height, *thumb_h);
// Create a texture to hold the current framebuffer content
GLuint src_texture = 0;
glGenTextures(1, &src_texture);
glBindTexture(GL_TEXTURE_2D, src_texture);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
// Copy the current framebuffer to the texture
glCopyTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, 0, 0, vw, vh, 0);
// Create a temporary framebuffer for GPU-based scaling
GLuint temp_texture = 0, temp_framebuffer = 0;
setup_texture_as_render_target(*thumb_w, *thumb_h, &temp_texture, &temp_framebuffer);
// Save current state
GLint current_framebuffer;
glGetIntegerv(GL_DRAW_FRAMEBUFFER_BINDING, &current_framebuffer);
// Bind our temporary framebuffer for rendering
bind_framebuffer_for_output(temp_framebuffer);
save_viewport_using_bottom_left_origin(0, 0, *thumb_w, *thumb_h);
// Use the screenshot program to render the scaled framebuffer with proper color space handling
bind_program(SCREENSHOT_PROGRAM);
// Set source rectangle (normalized coordinates: 0 to 1)
// Note: OpenGL texture origin is bottom-left, but Region uses top-left origin
// Convert from screen coordinates (top-left origin) to OpenGL texture coordinates (bottom-left origin)
float src_left_norm = (float)region.left / (float)vw;
float src_right_norm = (float)region.right / (float)vw;
float src_bottom_norm = (float)(vh - region.bottom) / (float)vh;
float src_top_norm = (float)(vh - region.top) / (float)vh;
glUniform4f(program_uniform_location(SCREENSHOT_PROGRAM, "src_rect"), src_left_norm, src_top_norm, src_right_norm, src_bottom_norm);
// Set destination rectangle (NDC coordinates: -1 to 1)
glUniform4f(program_uniform_location(SCREENSHOT_PROGRAM, "dest_rect"), -1.0f, -1.0f, 1.0f, 1.0f);
// Set the source texture size for proper downscaling
glUniform2f(program_uniform_location(SCREENSHOT_PROGRAM, "src_size"), (float)vw, (float)vh);
// Bind the source texture
glActiveTexture(GL_TEXTURE0 + GRAPHICS_UNIT);
glBindTexture(GL_TEXTURE_2D, src_texture);
// Draw the scaled quad
draw_quad(false, 0);
// Read the scaled result
glPixelStorei(GL_PACK_ALIGNMENT, 1);
glReadPixels(0, 0, *thumb_w, *thumb_h, GL_RGBA, GL_UNSIGNED_BYTE, dst_buf);
glPixelStorei(GL_PACK_ALIGNMENT, 4);
// Restore previous state
bind_framebuffer_for_output(current_framebuffer);
restore_viewport();
// Clean up temporary resources
free_texture(&src_texture);
free_texture(&temp_texture);
free_framebuffer(&temp_framebuffer);
}
// }}}
// Python API {{{
static PyObject *
pygpu_driver_version_string(PyObject *self UNUSED, PyObject *args UNUSED) {
return PyUnicode_FromString(global_state.gl_version ? gl_version_string() : "");
}
static void
free_custom_shader_pipeline(CustomShaderPipeline *p) {
for (size_t i = 0; i < p->num_groups; i++) {
if (p->groups[i].animation_curve_is_shared) p->groups[i].animation_curve = NULL;
else p->groups[i].animation_curve = free_animation(p->groups[i].animation_curve);
}
}
static NamedTexture
named_texture_from_str(const char *s) {
if (strcmp(s, "a") == 0) return CUSTOM_SHADER_TEXTURE_A;
if (strcmp(s, "b") == 0) return CUSTOM_SHADER_TEXTURE_B;
if (strcmp(s, "persist") == 0) return CUSTOM_SHADER_TEXTURE_PERSIST;
return CUSTOM_SHADER_TEXTURE_DEFAULT;
}
static ShaderAnimEventMask
shader_anim_event_bit(const char *s, CustomShaderPipeline *p) {
if (strcmp(s, "pointer-left-button-press") == 0) return 1u << SHADER_ANIM_EVENT_POINTER_LEFT_BUTTON_PRESS;
if (strcmp(s, "os-window-focus-in") == 0) return 1u << SHADER_ANIM_EVENT_OS_WINDOW_FOCUS_IN;
if (strcmp(s, "os-window-focus-out") == 0) return 1u << SHADER_ANIM_EVENT_OS_WINDOW_FOCUS_OUT;
if (strcmp(s, "window-focus-in") == 0) return 1u << SHADER_ANIM_EVENT_WINDOW_FOCUS_IN;
if (strcmp(s, "window-focus-out") == 0) return 1u << SHADER_ANIM_EVENT_WINDOW_FOCUS_OUT;
if (strcmp(s, "tab-change") == 0) return 1u << SHADER_ANIM_EVENT_TAB_CHANGE;
if (strcmp(s, "bell-in-window") == 0) return 1u << SHADER_ANIM_EVENT_BELL_IN_WINDOW;
if (strcmp(s, "user-activity") == 0) return 1u << SHADER_ANIM_EVENT_USER_ACTIVITY;
if (strcmp(s, "user-idle") == 0) return 1u << SHADER_ANIM_EVENT_USER_IDLE;
if (strcmp(s, "cursor-trail-move") == 0) return 1u << SHADER_ANIM_EVENT_CURSOR_TRAIL_MOVE;
if (strcmp(s, "cursor-trail-stop") == 0) return 1u << SHADER_ANIM_EVENT_CURSOR_TRAIL_STOP;
if (strcmp(s, "command-start") == 0) return 1u << SHADER_ANIM_EVENT_COMMAND_START;
if (strcmp(s, "command-finish") == 0) return 1u << SHADER_ANIM_EVENT_COMMAND_FINISH;
if (strcmp(s, "alternate-screen-enter") == 0) return 1u << SHADER_ANIM_EVENT_ALTERNATE_SCREEN_ENTER;
if (strcmp(s, "alternate-screen-exit") == 0) return 1u << SHADER_ANIM_EVENT_ALTERNATE_SCREEN_EXIT;
static const char user_prefix[] = "user:";
if (strncmp(s, user_prefix, sizeof(user_prefix) - 1) == 0) {
const char *name = s + sizeof(user_prefix) - 1;
if (!name[0] || strlen(name) > MAX_SHADER_ANIM_USER_EVENT_NAME_LEN) return 0;
size_t i = 0;
while (i < p->num_user_events && strcmp(p->user_events[i], name) != 0) i++;
if (i == p->num_user_events) {
if (i >= MAX_SHADER_ANIM_USER_EVENTS) return 0;
memcpy(p->user_events[p->num_user_events++], name, strlen(name) + 1);
}
return 1ull << (SHADER_ANIM_FIRST_USER_EVENT + i);
}
return 0;
}
static ShaderAnimEventMask
parse_anim_event_sequence(PyObject *seq, CustomShaderPipeline *p) {
ShaderAnimEventMask mask = 0;
for (Py_ssize_t i = 0; i < PySequence_Fast_GET_SIZE(seq); i++) {
PyObject *item = PySequence_Fast_GET_ITEM(seq, i);
if (PyUnicode_Check(item)) mask |= shader_anim_event_bit(PyUnicode_AsUTF8(item), p);
}
return mask;
}
static bool
transfer_pipeline_to_struct(PyObject *pg, CustomShaderPipeline *p) {
#define is_seq(x) (x && (PyList_Check(x) || PyTuple_Check(x)))
PyObject *textures = PyDict_GetItemString(pg, "textures");
if (is_seq(textures)) {
for (Py_ssize_t i = 0; i < PySequence_Fast_GET_SIZE(textures); i++) {
PyObject *t = PySequence_Fast_GET_ITEM(textures, i);
if (!PyUnicode_Check(t)) {
PyErr_SetString(PyExc_TypeError, "pipeline textures must be strings");
return false;
}
p->textures |= (1u << named_texture_from_str(PyUnicode_AsUTF8(t)));
}
}
PyObject *groups = PyDict_GetItemString(pg, "groups");
if (!is_seq(groups)) {
PyErr_SetString(PyExc_TypeError, "pipeline groups must be a sequence");
return false;
}
Py_ssize_t num_groups = PySequence_Fast_GET_SIZE(groups);
if ((size_t)num_groups > arraysz(p->groups)) {
PyErr_SetString(PyExc_ValueError, "too many pipeline groups");
return false;
}
p->num_groups = (size_t)num_groups;
for (Py_ssize_t i = 0; i < num_groups; i++) {
PyObject *g = PySequence_Fast_GET_ITEM(groups, i);
if (!PyDict_Check(g)) {
PyErr_SetString(PyExc_TypeError, "pipeline group must be a dict");
return false;
}
CustomShaderGroup *cg = &p->groups[i];
PyObject *vp = PyDict_GetItemString(g, "viewport_pos");
if (is_seq(vp) && PySequence_Fast_GET_SIZE(vp) == 2) {
cg->viewport_pos.x = (float)PyFloat_AsDouble(PySequence_Fast_GET_ITEM(vp, 0));
cg->viewport_pos.y = (float)PyFloat_AsDouble(PySequence_Fast_GET_ITEM(vp, 1));
}
PyObject *vs = PyDict_GetItemString(g, "viewport_size");
if (is_seq(vs) && PySequence_Fast_GET_SIZE(vs) == 2) {
cg->viewport_size.w = (float)PyFloat_AsDouble(PySequence_Fast_GET_ITEM(vs, 0));
cg->viewport_size.h = (float)PyFloat_AsDouble(PySequence_Fast_GET_ITEM(vs, 1));
}
PyObject *ot = PyDict_GetItemString(g, "output_texture");
if (ot && PyUnicode_Check(ot)) { cg->output_texture = named_texture_from_str(PyUnicode_AsUTF8(ot)); }
PyObject *astart = PyDict_GetItemString(g, "animation_start");
if (is_seq(astart)) cg->animation_start_events = parse_anim_event_sequence(astart, p);
PyObject *atoggle = PyDict_GetItemString(g, "animation_toggle");
if (is_seq(atoggle)) cg->animation_toggle_events = parse_anim_event_sequence(atoggle, p);
PyObject *acurve = PyDict_GetItemString(g, "animation_curve");
if (acurve && PyObject_IsTrue(acurve)) {
if ((cg->animation_curve = alloc_animation()) != NULL) add_easing_function(cg->animation_curve, acurve, 0.0, 1.0);
for (Py_ssize_t j = 0; j < i; j++) {
if (animations_equal(p->groups[j].animation_curve, cg->animation_curve)) {
free_animation(cg->animation_curve);
cg->animation_curve = p->groups[j].animation_curve;
cg->animation_curve_is_shared = true;
break;
}
}
}
PyObject *astep = PyDict_GetItemString(g, "animation_step");
cg->animation_step = (astep && PyLong_Check(astep)) ? (monotonic_t)PyLong_AsLong(astep) : ANIMATION_SAMPLE_WAIT;
cg->animation_step = cg->animation_step == 0 ? MONOTONIC_T_MAX : MAX(cg->animation_step, OPT(repaint_delay));
PyObject *ae_events = PyDict_GetItemString(g, "animation_end_events");
if (is_seq(ae_events)) cg->animation_end_events = parse_anim_event_sequence(ae_events, p);
PyObject *ae_dur = PyDict_GetItemString(g, "animation_end_duration");
if (ae_dur && PyLong_Check(ae_dur)) cg->animation_end_duration = (monotonic_t)PyLong_AsLong(ae_dur);
PyObject *a_dur = PyDict_GetItemString(g, "animation_duration");
cg->animation_duration = (a_dur && PyLong_Check(a_dur)) ? (monotonic_t)PyLong_AsLong(a_dur) : -1;
PyObject *a_rev = PyDict_GetItemString(g, "animation_reverse");
cg->animation_reverse = (a_rev && PyLong_Check(a_rev)) ? (monotonic_t)PyLong_AsLong(a_rev) : 0;
PyObject *a_loops = PyDict_GetItemString(g, "animation_loops");
cg->animation_loops = (a_loops && PyLong_Check(a_loops)) ? (unsigned)PyLong_AsUnsignedLong(a_loops) : 1;
PyObject *attached = PyDict_GetItemString(g, "attached");
cg->attached = attached && PyObject_IsTrue(attached);
}
p->all_animation_start_events = 0;
for (size_t i = 0; i < p->num_groups; i++) p->all_animation_start_events |= p->groups[i].animation_start_events | p->groups[i].animation_toggle_events;
p->bypass_builtin_trail = (p->all_animation_start_events & (1u << SHADER_ANIM_EVENT_CURSOR_TRAIL_MOVE)) != 0;
#undef is_seq
return true;
}
static bool
attach_shaders(PyObject *sources, GLuint program_id, GLenum shader_type) {
RAII_ALLOC(const GLchar *, c_sources, calloc(PyTuple_GET_SIZE(sources), sizeof(GLchar *)));
for (Py_ssize_t i = 0; i < PyTuple_GET_SIZE(sources); i++) {
PyObject *temp = PyTuple_GET_ITEM(sources, i);
if (!PyUnicode_Check(temp)) {
PyErr_SetString(PyExc_TypeError, "shaders must be strings");
return false;
}
c_sources[i] = PyUnicode_AsUTF8(temp);
}
GLuint shader_id = compile_shaders(shader_type, PyTuple_GET_SIZE(sources), c_sources);
if (shader_id == 0) return false;
glAttachShader(program_id, shader_id);
glDeleteShader(shader_id);
return true;
}
static PyObject *
compile_program(PyObject UNUSED *self, PyObject *args) {
PyObject *vertex_shaders, *fragment_shaders, *metadata;
int which, allow_recompile = 0;
if (!PyArg_ParseTuple(
args, "iO!O!O!|p", &which, &PyTuple_Type, &vertex_shaders, &PyTuple_Type, &fragment_shaders, &PyDict_Type, &metadata, &allow_recompile))
return NULL;
if (which == -1) {
init_cell_program();
Py_RETURN_NONE;
}
if (which == -2) {
init_custom_programs();
Py_RETURN_NONE;
}
if (which < 0 || which >= NUM_PROGRAMS) {
PyErr_Format(PyExc_ValueError, "Unknown program: %d", which);
return NULL;
}
Program *program = program_ptr(which);
if (!PyTuple_GET_SIZE(vertex_shaders)) {
if (which == CUSTOM_END_PROGRAM) {
free_custom_shader_pipeline(&custom_shaders.end);
zero_at_ptr(&custom_shaders.end);
custom_shaders.generation++;
}
if (program->id != 0) {
glDeleteProgram(program->id);
program->id = 0;
}
Py_RETURN_NONE;
}
if (program->id != 0) {
if (allow_recompile) {
glDeleteProgram(program->id);
program->id = 0;
} else {
PyErr_SetString(PyExc_ValueError, "program already compiled");
return NULL;
}
}
#define fail_compile() \
{ \
glDeleteProgram(program->id); \
return NULL; \
}
program->id = glCreateProgram();
if (!attach_shaders(vertex_shaders, program->id, GL_VERTEX_SHADER)) fail_compile();
if (!attach_shaders(fragment_shaders, program->id, GL_FRAGMENT_SHADER)) fail_compile();
glLinkProgram(program->id);
GLint ret = GL_FALSE;
glGetProgramiv(program->id, GL_LINK_STATUS, &ret);
if (ret != GL_TRUE) {
GLsizei len;
static char glbuf[4096];
glGetProgramInfoLog(program->id, sizeof(glbuf), &len, glbuf);
PyErr_Format(PyExc_ValueError, "Failed to link GLSL shaders:\n%s", glbuf);
fail_compile();
}
#undef fail_compile
if (which == CUSTOM_END_PROGRAM) {
free_custom_shader_pipeline(&custom_shaders.end);
zero_at_ptr(&custom_shaders.end);
PyObject *pg = PyDict_GetItemString(metadata, "pipeline");
if (pg && !transfer_pipeline_to_struct(pg, &custom_shaders.end)) {
glDeleteProgram(program->id);
return NULL;
}
custom_shaders.end.active = true;
PyObject *has_pointer_map = PyDict_GetItemString(metadata, "has_pointer_map");
custom_shaders.end.has_pointer_map = has_pointer_map && PyObject_IsTrue(has_pointer_map) == 1;
PyObject *pointer_map_groups = PyDict_GetItemString(metadata, "pointer_map_groups");
if (pointer_map_groups && (PyList_Check(pointer_map_groups) || PyTuple_Check(pointer_map_groups))) {
for (Py_ssize_t i = 0; i < PySequence_Fast_GET_SIZE(pointer_map_groups); i++) {
long g = PyLong_AsLong(PySequence_Fast_GET_ITEM(pointer_map_groups, i));
if (g >= 0 && (size_t)g < custom_shaders.end.num_groups) custom_shaders.end.pointer_map_groups |= 1u << g;
}
if (PyErr_Occurred()) {
glDeleteProgram(program->id);
return NULL;
}
}
custom_shaders.generation++;
}
init_uniforms(which);
set_program_layout(which, metadata);
return Py_BuildValue("I", program->id);
}
#define PYWRAP0(name) static PyObject *py##name(PYNOARG)
#define PYWRAP1(name) static PyObject *py##name(PyObject UNUSED *self, PyObject *args)
#define PA(fmt, ...) \
if (!PyArg_ParseTuple(args, fmt, __VA_ARGS__)) return NULL;
#define ONE_INT(name) \
PYWRAP1(name) { \
name(PyLong_AsSsize_t(args)); \
Py_RETURN_NONE; \
}
#define TWO_INT(name) \
PYWRAP1(name) { \
int a, b; \
PA("ii", &a, &b); \
name(a, b); \
Py_RETURN_NONE; \
}
#define NO_ARG(name) \
PYWRAP0(name) { \
name(); \
Py_RETURN_NONE; \
}
#define NO_ARG_INT(name) \
PYWRAP0(name) { return PyLong_FromSsize_t(name()); }
ONE_INT(bind_program)
NO_ARG(unbind_program)
PYWRAP0(create_vao) {
int ans = create_vao();
if (ans < 0) return NULL;
return Py_BuildValue("i", ans);
}
ONE_INT(bind_vertex_array)
NO_ARG(unbind_vertex_array)
TWO_INT(unmap_vao_buffer)
static PyObject *
sprite_map_set_limits(PyObject UNUSED *self, PyObject *args) {
unsigned int w, h;
if (!PyArg_ParseTuple(args, "II", &w, &h)) return NULL;
sprite_tracker_set_limits(w, h);
max_texture_size = w;
max_array_texture_layers = h;
Py_RETURN_NONE;
}
// test only wrapper for decorations_map_new_size
static PyObject *
pydecorations_map_new_size(PyObject *self UNUSED, PyObject *args) {
unsigned long count, current_capacity, max_size;
if (!PyArg_ParseTuple(args, "kkk", &count, &current_capacity, &max_size)) return NULL;
GLint w, h;
if (!decorations_map_new_size(count, current_capacity, max_size, &w, &h)) Py_RETURN_NONE;
return Py_BuildValue("ii", (int)w, (int)h);
}
// Test only. Wraps custom_shader_needs_render() so the redraw decision can be
// exercised from the Python test suite without a GPU context. Each state is a
// (has_active_shaders, min_step, next_end_at) tuple.
static bool
parse_shader_anim_state(PyObject *obj, ShaderAnimState *ans) {
int has_active;
long long min_step, next_end_at;
if (!PyArg_ParseTuple(obj, "pLL", &has_active, &min_step, &next_end_at)) return false;
*ans = (ShaderAnimState){.has_active_shaders = has_active, .min_step = min_step, .next_end_at = next_end_at};
return true;
}
static PyObject *
pycustom_shader_needs_render(PyObject *self UNUSED, PyObject *args) {
PyObject *b, *a;
unsigned long long event_mask;
long long now;
ShaderAnimState before, after;
if (!PyArg_ParseTuple(args, "O!O!KL", &PyTuple_Type, &b, &PyTuple_Type, &a, &event_mask, &now)) return NULL;
if (!parse_shader_anim_state(b, &before) || !parse_shader_anim_state(a, &after)) return NULL;
if (custom_shader_needs_render(&before, &after, event_mask, now)) Py_RETURN_TRUE;
Py_RETURN_FALSE;
}
// Test only. Runs the renderer's per-frame custom shader bookkeeping (the same
// sequence as prepare_to_render_os_window) against a freshly allocated, zero
// initialized OSWindow and returns whether each tick asked for a redraw. Used
// to assert that an idle OS window with no shader work pending stays idle.
static PyObject *
pysimulate_custom_shader_render_ticks(PyObject *self UNUSED, PyObject *args) {
unsigned long num_ticks;
unsigned long long event_mask = 0;
int initialize = 0;
if (!PyArg_ParseTuple(args, "k|Kp", &num_ticks, &event_mask, &initialize)) return NULL;
RAII_ALLOC(OSWindow, osw, calloc(1, sizeof(OSWindow)));
if (!osw) return PyErr_NoMemory();
if (initialize) init_shader_animation_state(osw);
RAII_PyObject(ans, PyList_New(0));
if (!ans) return NULL;
monotonic_t now = monotonic();
for (unsigned long i = 0; i < num_ticks; i++, now += ms_to_monotonic_t(10ll)) {
const ShaderAnimState before = osw->shader_anim;
update_custom_shader_animations(event_mask, now, osw);
PyObject *v = custom_shader_needs_render(&before, &osw->shader_anim, event_mask, now) ? Py_True : Py_False;
if (PyList_Append(ans, v) != 0) return NULL;
}
return Py_NewRef(ans);
}
static PyObject *
pyfire_custom_shader_event(PyObject *self UNUSED, PyObject *args) {
const char *name;
unsigned long long os_window_id = 0;
if (!PyArg_ParseTuple(args, "s|K", &name, &os_window_id)) return NULL;
if (fire_custom_shader_user_event(name, os_window_id)) Py_RETURN_TRUE;
Py_RETURN_FALSE;
}
// Test only. Loads pipeline, a dict as sent by compile_program(), in place of
// the current one and runs the animation state machine over an OS window, one
// tick per item of ticks, each a sequence of event names, 10ms apart. Returns
// the active state of every group after each tick, or if with_progress is
// true, the animation_progress of active groups and None for inactive ones.
static PyObject *
pysimulate_custom_shader_animation(PyObject *self UNUSED, PyObject *args) {
PyObject *pipeline, *ticks;
int with_progress = 0;
if (!PyArg_ParseTuple(args, "O!O!|p", &PyDict_Type, &pipeline, &PyList_Type, &ticks, &with_progress)) return NULL;
RAII_ALLOC(OSWindow, osw, calloc(1, sizeof(OSWindow)));
if (!osw) return PyErr_NoMemory();
const CustomShaderPipeline saved = custom_shaders.end;
const size_t saved_count = custom_shaders.count;
zero_at_ptr(&custom_shaders.end);
PyObject *ans = NULL;
if (transfer_pipeline_to_struct(pipeline, &custom_shaders.end)) {
custom_shaders.count = 1;
init_shader_animation_state(osw);
ans = PyList_New(0);
monotonic_t now = monotonic();
for (Py_ssize_t i = 0; ans && i < PyList_GET_SIZE(ticks); i++, now += ms_to_monotonic_t(10ll)) {
PyObject *names = PyList_GET_ITEM(ticks, i);
if (!PyTuple_Check(names) && !PyList_Check(names)) {
PyErr_SetString(PyExc_TypeError, "each tick must be a sequence of event names");
Py_CLEAR(ans);
break;
}
// focus:in and focus:out change OS window focus, as the GLFW focus
// callback does, before the tick's events are consumed, the way
// prepare_to_render_os_window() does
for (Py_ssize_t n = 0; n < PySequence_Fast_GET_SIZE(names); n++) {
PyObject *item = PySequence_Fast_GET_ITEM(names, n);
if (!PyUnicode_Check(item)) continue;
const char *name = PyUnicode_AsUTF8(item);
if (!name) {
Py_CLEAR(ans);
break;
}
const bool focus_in = strcmp(name, "focus:in") == 0;
if (focus_in || strcmp(name, "focus:out") == 0) {
const bool focused = focus_in;
if (osw->is_focused != focused) record_os_window_focus_change_for_shaders(osw, focused);
osw->is_focused = focused;
}
}
if (!ans) break;
update_custom_shader_animations(parse_anim_event_sequence(names, &custom_shaders.end) | osw->shader_anim_event_registry, now, osw);
osw->shader_anim_event_registry = 0;
RAII_PyObject(states, PyTuple_New(custom_shaders.end.num_groups));
if (!states) {
Py_CLEAR(ans);
break;
}
float progress[MAX_CUSTOM_SHADER_GROUPS];
custom_shader_group_progress(osw, now, progress);
for (size_t g = 0; g < custom_shaders.end.num_groups; g++) {
const CustomShaderGroup *cg = custom_shaders.end.groups + g;
const bool active = !group_is_animated(cg) || osw->shader_group_anim[g].active;
PyObject *state =
with_progress && active ? PyFloat_FromDouble(progress[g]) : Py_NewRef(with_progress ? Py_None : (active ? Py_True : Py_False));
if (!state) break;
PyTuple_SET_ITEM(states, g, state);
}
if (PyErr_Occurred()) {
Py_CLEAR(ans);
break;
}
if (PyList_Append(ans, states) != 0) Py_CLEAR(ans);
}
}
free_custom_shader_pipeline(&custom_shaders.end);
custom_shaders.end = saved;
custom_shaders.count = saved_count;
return ans;
}
// Test only. Looks up a position in a pointer map given as the bytes of
// POINTER_MAP_SIZE x POINTER_MAP_SIZE (x, y) float pairs.
static PyObject *
pysample_pointer_map(PyObject *self UNUSED, PyObject *args) {
Py_buffer data;
double u, v;
if (!PyArg_ParseTuple(args, "y*dd", &data, &u, &v)) return NULL;
if ((size_t)data.len != sizeof(float) * 2 * POINTER_MAP_SIZE * POINTER_MAP_SIZE) {
PyBuffer_Release(&data);
PyErr_SetString(PyExc_ValueError, "pointer map has the wrong size");
return NULL;
}
sample_pointer_map(data.buf, &u, &v);
PyBuffer_Release(&data);
return Py_BuildValue("dd", u, v);
}
#define M(name, arg_type) {#name, (PyCFunction)name, arg_type, NULL}
#define MW(name, arg_type) {#name, (PyCFunction)py##name, arg_type, NULL}
static PyMethodDef module_methods[] = {
M(compile_program, METH_VARARGS),
M(sprite_map_set_limits, METH_VARARGS),
MW(create_vao, METH_NOARGS),
MW(gpu_driver_version_string, METH_NOARGS),
MW(bind_vertex_array, METH_O),
MW(unbind_vertex_array, METH_NOARGS),
MW(unmap_vao_buffer, METH_VARARGS),
MW(bind_program, METH_O),
MW(unbind_program, METH_NOARGS),
MW(custom_shader_needs_render, METH_VARARGS),
MW(decorations_map_new_size, METH_VARARGS),
MW(simulate_custom_shader_render_ticks, METH_VARARGS),
MW(sample_pointer_map, METH_VARARGS),
MW(fire_custom_shader_event, METH_VARARGS),
MW(simulate_custom_shader_animation, METH_VARARGS),
{NULL, NULL, 0, NULL} /* Sentinel */
};
void
initialize_gpu(void) {
gl_init();
query_gpu_texture_limits();
}
void
free_vao(ssize_t vao_idx) {
remove_vao(vao_idx);
}
void
cleanup_shader_resources_on_terminate(void) {
free_pointer_map_target();
if (shader_globals_vao_idx != -1) {
remove_vao(shader_globals_vao_idx);
shader_globals_vao_idx = -1;
}
}
static void
finalize(void) {
default_visual_bell_animation = free_animation(default_visual_bell_animation);
free_custom_shader_pipeline(&custom_shaders.end);
free_program_layouts();
}
bool
init_shaders(PyObject *module) {
#define C(x) \
if (PyModule_AddIntConstant(module, #x, x) != 0) { \
PyErr_NoMemory(); \
return false; \
}
C(CELL_PROGRAM);
C(CELL_FG_PROGRAM);
C(CELL_BG_PROGRAM);
C(BORDERS_PROGRAM);
C(GRAPHICS_PROGRAM);
C(GRAPHICS_PREMULT_PROGRAM);
C(GRAPHICS_ALPHA_MASK_PROGRAM);
C(GRAPHICS_INDEXED_PROGRAM);
C(GRAPHICS_INDEXED_UNSCALED_PROGRAM);
C(BGIMAGE_PROGRAM);
C(TINT_PROGRAM);
C(TRAIL_PROGRAM);
C(BLIT_PROGRAM);
C(SCREENSHOT_PROGRAM);
C(ROUNDED_RECT_PROGRAM);
C(PADDING_PROGRAM);
C(CUSTOM_END_PROGRAM);
C(MAX_CUSTOM_SHADER_GROUPS);
C(MAX_SHADER_ANIM_USER_EVENTS);
C(MAX_SHADER_ANIM_USER_EVENT_NAME_LEN);
C(POINTER_MAP_SIZE);
C(GLSL_VERSION);
C(GL_VERSION);
C(GL_VENDOR);
C(GL_SHADING_LANGUAGE_VERSION);
C(GL_RENDERER);
C(GL_TRIANGLE_FAN);
C(GL_TRIANGLE_STRIP);
C(GL_TRIANGLES);
C(GL_LINE_LOOP);
C(GL_COLOR_BUFFER_BIT);
C(GL_VERTEX_SHADER);
C(GL_FRAGMENT_SHADER);
C(GL_TRUE);
C(GL_FALSE);
C(GL_COMPILE_STATUS);
C(GL_LINK_STATUS);
C(GL_MAX_ARRAY_TEXTURE_LAYERS);
C(GL_TEXTURE_BINDING_BUFFER);
C(GL_MAX_TEXTURE_BUFFER_SIZE);
C(GL_MAX_TEXTURE_SIZE);
C(GL_TEXTURE_2D_ARRAY);
C(GL_LINEAR);
C(GL_CLAMP_TO_EDGE);
C(GL_NEAREST);
C(GL_TEXTURE_MIN_FILTER);
C(GL_TEXTURE_MAG_FILTER);
C(GL_TEXTURE_WRAP_S);
C(GL_TEXTURE_WRAP_T);
C(GL_UNPACK_ALIGNMENT);
C(GL_R8);
C(GL_RED);
C(GL_UNSIGNED_BYTE);
C(GL_UNSIGNED_SHORT);
C(GL_R32UI);
C(GL_RGB32UI);
C(GL_RGBA);
C(GL_TEXTURE_BUFFER);
C(GL_STATIC_DRAW);
C(GL_STREAM_DRAW);
C(GL_DYNAMIC_DRAW);
C(GL_SRC_ALPHA);
C(GL_ONE_MINUS_SRC_ALPHA);
C(GL_WRITE_ONLY);
C(GL_READ_ONLY);
C(GL_READ_WRITE);
C(GL_BLEND);
C(GL_FLOAT);
C(GL_UNSIGNED_INT);
C(GL_ARRAY_BUFFER);
C(GL_UNIFORM_BUFFER);
C(ANIMATION_SAMPLE_WAIT);
#undef C
if (PyModule_AddFunctions(module, module_methods) != 0) return false;
register_at_exit_cleanup_func(SHADERS_CLEANUP_FUNC, finalize);
return true;
}
// }}}