caiyu/solvespace
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
55baaf310f
solvespace/src/render/render.cpp

456 lines
15 KiB
C++
Raw Normal View History

Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
//-----------------------------------------------------------------------------
// Backend-agnostic rendering interface, and various backends we use.
//
// Copyright 2016 whitequark
//-----------------------------------------------------------------------------
#include "solvespace.h"
namespace SolveSpace {
//-----------------------------------------------------------------------------
// Camera transformations.
//-----------------------------------------------------------------------------
Point2d Camera::ProjectPoint(Vector p) const {
Vector p3 = ProjectPoint3(p);
Point2d p2 = { p3.x, p3.y };
return p2;
}
Vector Camera::ProjectPoint3(Vector p) const {
double w;
Vector r = ProjectPoint4(p, &w);
return r.ScaledBy(scale/w);
}
Vector Camera::ProjectPoint4(Vector p, double *w) const {
p = p.Plus(offset);
Vector r;
r.x = p.Dot(projRight);
r.y = p.Dot(projUp);
r.z = p.Dot(projUp.Cross(projRight));
*w = 1 + r.z*tangent*scale;
return r;
}
Vector Camera::UnProjectPoint(Point2d p) const {
Vector orig = offset.ScaledBy(-1);
// Note that we're ignoring the effects of perspective. Since our returned
// point has the same component normal to the screen as the offset, it
// will have z = 0 after the rotation is applied, thus w = 1. So this is
// correct.
orig = orig.Plus(projRight.ScaledBy(p.x / scale)).Plus(
projUp. ScaledBy(p.y / scale));
return orig;
}
Vector Camera::UnProjectPoint3(Vector p) const {
p.z = p.z / (scale - p.z * tangent * scale);
double w = 1 + p.z * tangent * scale;
p.x *= w / scale;
p.y *= w / scale;
Vector orig = offset.ScaledBy(-1);
orig = orig.Plus(projRight.ScaledBy(p.x)).Plus(
projUp. ScaledBy(p.y).Plus(
Unbreak Camera::UnProjectPoint3. The handedness was wrong, and this resulted in weird glitches while drawing anything that depends on AlignToPixelGrid, like normals.
2016-11-19 14:50:38 +08:00
projUp.Cross(projRight). ScaledBy(p.z)));
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
return orig;
}
Vector Camera::VectorFromProjs(Vector rightUpForward) const {
Vector n = projRight.Cross(projUp);
Vector r = (projRight.ScaledBy(rightUpForward.x));
r = r.Plus(projUp.ScaledBy(rightUpForward.y));
r = r.Plus(n.ScaledBy(rightUpForward.z));
return r;
}
Vector Camera::AlignToPixelGrid(Vector v) const {
if(!hasPixels) return v;
v = ProjectPoint3(v);
v.x = floor(v.x) + 0.5;
v.y = floor(v.y) + 0.5;
return UnProjectPoint3(v);
}
Add a new renderer that prepares geometry for 2d backends. SurfaceRenderer is a new renderer implementing the Canvas interface running entirely on the CPU; it projects strokes and triangles in the exact same way as OpenGL would, and it can be used for rendering into raster or vector 2d surfaces.
2016-07-22 04:27:53 +08:00
SBezier Camera::ProjectBezier(SBezier b) const {
Quaternion q = Quaternion::From(projRight, projUp);
q = q.Inverse();
// we want Q*(p - o) = Q*p - Q*o
b = b.TransformedBy(q.Rotate(offset).ScaledBy(scale), q, scale);
for(int i = 0; i <= b.deg; i++) {
Vector4 ct = Vector4::From(b.weight[i], b.ctrl[i]);
// so the desired curve, before perspective, is
// (x/w, y/w, z/w)
// and after perspective is
// ((x/w)/(1 - (z/w)*tangent, ...
// = (x/(w - z*tangent), ...
// so we want to let w' = w - z*tangent
ct.w = ct.w - ct.z*tangent;
b.ctrl[i] = ct.PerspectiveProject();
b.weight[i] = ct.w;
}
return b;
}
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
void Camera::LoadIdentity() {
offset = { 0.0, 0.0, 0.0 };
projRight = { 1.0, 0.0, 0.0 };
projUp = { 0.0, 1.0, 0.0 };
scale = 1.0;
tangent = 0.0;
}
void Camera::NormalizeProjectionVectors() {
if(projRight.Magnitude() < LENGTH_EPS) {
projRight = Vector::From(1, 0, 0);
}
Vector norm = projRight.Cross(projUp);
// If projRight and projUp somehow ended up parallel, then pick an
// arbitrary projUp normal to projRight.
if(norm.Magnitude() < LENGTH_EPS) {
norm = projRight.Normal(0);
}
projUp = norm.Cross(projRight);
projUp = projUp.WithMagnitude(1);
projRight = projRight.WithMagnitude(1);
}
//-----------------------------------------------------------------------------
// Stroke and fill caching.
//-----------------------------------------------------------------------------
bool Canvas::Stroke::Equals(const Stroke &other) const {
return (layer == other.layer &&
zIndex == other.zIndex &&
color.Equals(other.color) &&
width == other.width &&
Fix many rendering bugs introduced in df83ee4 and 9f97e9a.
2016-10-13 21:23:22 +08:00
unit == other.unit &&
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
stipplePattern == other.stipplePattern &&
stippleScale == other.stippleScale);
}
Allow selecting unit (px/mm) in Canvas::Stroke. By directly specifying the desired end result to the renderer, we can avoid regeneration of geometry.
2016-06-30 23:54:35 +08:00
double Canvas::Stroke::WidthMm(const Camera &camera) const {
switch(unit) {
case Canvas::Unit::MM:
return width;
case Canvas::Unit::PX:
return width / camera.scale;
default:
ssassert(false, "Unexpected unit");
}
}
double Canvas::Stroke::WidthPx(const Camera &camera) const {
switch(unit) {
case Canvas::Unit::MM:
return width * camera.scale;
case Canvas::Unit::PX:
return width;
default:
ssassert(false, "Unexpected unit");
}
}
double Canvas::Stroke::StippleScaleMm(const Camera &camera) const {
switch(unit) {
case Canvas::Unit::MM:
return stippleScale;
case Canvas::Unit::PX:
return stippleScale / camera.scale;
default:
ssassert(false, "Unexpected unit");
}
}
double Canvas::Stroke::StippleScalePx(const Camera &camera) const {
switch(unit) {
case Canvas::Unit::MM:
return stippleScale * camera.scale;
case Canvas::Unit::PX:
return stippleScale;
default:
ssassert(false, "Unexpected unit");
}
}
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
bool Canvas::Fill::Equals(const Fill &other) const {
return (layer == other.layer &&
zIndex == other.zIndex &&
color.Equals(other.color) &&
Implement an OpenGL 2 renderer. There are two main reasons to desire an OpenGL 2 renderer: 1. Compatibility. The compatibility profile, ironically, does not offer a lot of compatibility, and our OpenGL 1 renderer will not run on Android, iOS, or WebGL. 2. Performance. The immediate mode does not scale, and in fact becomes very slow with only a moderate amount of lines on screen, and only a somewhat large amount of triangles. This commit implements a basic OpenGL 2 renderer that uses only features from the (OpenGL 3.2) core profile. It is not yet faster than the OpenGL 1 renderer, primarily because it uses a lot of small draw calls. This commit uses OpenGL 2 on Linux and Mac OS X directly (i.e. links to the GL symbols from version 2+); on Windows this is impossible with the default drivers, so for now OpenGL 1 is still used there.
2016-06-30 23:54:35 +08:00
pattern == other.pattern &&
texture == other.texture);
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
}
Eliminate several memory leaks. All leaks found with valgrind while running the test suite. This commit also clears the Cairo cache to improve SNR of Valgrind output.
2016-08-01 11:52:12 +08:00
void Canvas::Clear() {
strokes.Clear();
fills.Clear();
}
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
Canvas::hStroke Canvas::GetStroke(const Stroke &stroke) {
for(const Stroke &s : strokes) {
if(s.Equals(stroke)) return s.h;
}
Stroke strokeCopy = stroke;
return strokes.AddAndAssignId(&strokeCopy);
}
Canvas::hFill Canvas::GetFill(const Fill &fill) {
for(const Fill &f : fills) {
if(f.Equals(fill)) return f.h;
}
Fill fillCopy = fill;
return fills.AddAndAssignId(&fillCopy);
}
Use a dedicated BitmapFont instance per Canvas. Before this commit, updates to the bitmap font in the graphics window cause missing characters in the text window and vice versa. This is because BitmapFont contains a texture that's mutated, and sharing it would also require sharing display lists between GL contexts, which is not done (and overly complicated for this case anyway).
2017-01-10 10:52:51 +08:00
BitmapFont *Canvas::GetBitmapFont() {
if(bitmapFont.IsEmpty()) {
bitmapFont = BitmapFont::Create();
}
return &bitmapFont;
}
Add an interface for view-independent rendering. To actually achieve improved performance with the OpenGL 2 renderer, we have to cache geometry that doesn't change when the viewport does (note that the rendered pixels can change quite dramatically because we can reconfigure shaders; e.g. stippling can be drawn in screen coordinates). This commit adds a BatchCanvas interface that can be implemented by renderers, and uses it for drawing entities such as lines and points.
2016-06-30 23:54:35 +08:00
std::shared_ptr<BatchCanvas> Canvas::CreateBatch() {
return std::shared_ptr<BatchCanvas>();
}
//-----------------------------------------------------------------------------
// An interface for view-independent visualization
//-----------------------------------------------------------------------------
const Camera &BatchCanvas::GetCamera() const {
ssassert(false, "Geometry drawn on BatchCanvas must be independent from camera");
}
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
//-----------------------------------------------------------------------------
// A wrapper around Canvas that simplifies drawing UI in screen coordinates
//-----------------------------------------------------------------------------
Allow using z-index in UiCanvas, and use it for tooltips. Before this commit, tooltips in the text window are drawn under the red "X" indicating a disabled button. After this commit, they are moved on top of that. This commit also alters the OpenGL renderers' SetCamera() method to clear the depth buffer, as that would interfere with drawing the UI; the toolbar would get occluded by geometry.
2016-11-16 05:35:06 +08:00
void UiCanvas::DrawLine(int x1, int y1, int x2, int y2, RgbaColor color, int width, int zIndex) {
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
Vector va = { (double)x1 + 0.5, (double)Flip(y1) + 0.5, 0.0 },
vb = { (double)x2 + 0.5, (double)Flip(y2) + 0.5, 0.0 };
Canvas::Stroke stroke = {};
Allow using z-index in UiCanvas, and use it for tooltips. Before this commit, tooltips in the text window are drawn under the red "X" indicating a disabled button. After this commit, they are moved on top of that. This commit also alters the OpenGL renderers' SetCamera() method to clear the depth buffer, as that would interfere with drawing the UI; the toolbar would get occluded by geometry.
2016-11-16 05:35:06 +08:00
stroke.layer = Canvas::Layer::NORMAL;
stroke.zIndex = zIndex;
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
stroke.width = (double)width;
stroke.color = color;
Fix many rendering bugs introduced in df83ee4 and 9f97e9a.
2016-10-13 21:23:22 +08:00
stroke.unit = Canvas::Unit::PX;
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
Canvas::hStroke hcs = canvas->GetStroke(stroke);
canvas->DrawLine(va, vb, hcs);
}
Allow using z-index in UiCanvas, and use it for tooltips. Before this commit, tooltips in the text window are drawn under the red "X" indicating a disabled button. After this commit, they are moved on top of that. This commit also alters the OpenGL renderers' SetCamera() method to clear the depth buffer, as that would interfere with drawing the UI; the toolbar would get occluded by geometry.
2016-11-16 05:35:06 +08:00
void UiCanvas::DrawRect(int l, int r, int t, int b, RgbaColor fillColor, RgbaColor outlineColor,
int zIndex) {
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
Vector va = { (double)l + 0.5, (double)Flip(b) + 0.5, 0.0 },
vb = { (double)l + 0.5, (double)Flip(t) + 0.5, 0.0 },
vc = { (double)r + 0.5, (double)Flip(t) + 0.5, 0.0 },
vd = { (double)r + 0.5, (double)Flip(b) + 0.5, 0.0 };
if(!fillColor.IsEmpty()) {
Canvas::Fill fill = {};
Allow using z-index in UiCanvas, and use it for tooltips. Before this commit, tooltips in the text window are drawn under the red "X" indicating a disabled button. After this commit, they are moved on top of that. This commit also alters the OpenGL renderers' SetCamera() method to clear the depth buffer, as that would interfere with drawing the UI; the toolbar would get occluded by geometry.
2016-11-16 05:35:06 +08:00
fill.layer = Canvas::Layer::NORMAL;
fill.zIndex = zIndex;
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
fill.color = fillColor;
Canvas::hFill hcf = canvas->GetFill(fill);
canvas->DrawQuad(va, vb, vc, vd, hcf);
}
if(!outlineColor.IsEmpty()) {
Canvas::Stroke stroke = {};
Allow using z-index in UiCanvas, and use it for tooltips. Before this commit, tooltips in the text window are drawn under the red "X" indicating a disabled button. After this commit, they are moved on top of that. This commit also alters the OpenGL renderers' SetCamera() method to clear the depth buffer, as that would interfere with drawing the UI; the toolbar would get occluded by geometry.
2016-11-16 05:35:06 +08:00
stroke.layer = Canvas::Layer::NORMAL;
stroke.zIndex = zIndex;
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
stroke.width = 1.0;
stroke.color = outlineColor;
Fix many rendering bugs introduced in df83ee4 and 9f97e9a.
2016-10-13 21:23:22 +08:00
stroke.unit = Canvas::Unit::PX;
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
Canvas::hStroke hcs = canvas->GetStroke(stroke);
canvas->DrawLine(va, vb, hcs);
canvas->DrawLine(vb, vc, hcs);
canvas->DrawLine(vc, vd, hcs);
canvas->DrawLine(vd, va, hcs);
}
}
Allow using z-index in UiCanvas, and use it for tooltips. Before this commit, tooltips in the text window are drawn under the red "X" indicating a disabled button. After this commit, they are moved on top of that. This commit also alters the OpenGL renderers' SetCamera() method to clear the depth buffer, as that would interfere with drawing the UI; the toolbar would get occluded by geometry.
2016-11-16 05:35:06 +08:00
void UiCanvas::DrawPixmap(std::shared_ptr<const Pixmap> pm, int x, int y, int zIndex) {
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
Canvas::Fill fill = {};
Allow using z-index in UiCanvas, and use it for tooltips. Before this commit, tooltips in the text window are drawn under the red "X" indicating a disabled button. After this commit, they are moved on top of that. This commit also alters the OpenGL renderers' SetCamera() method to clear the depth buffer, as that would interfere with drawing the UI; the toolbar would get occluded by geometry.
2016-11-16 05:35:06 +08:00
fill.layer = Canvas::Layer::NORMAL;
fill.zIndex = zIndex;
Use transparent white fill color for drawing pixmaps, not black. Textures can interact with selected color. This commit makes it a no-op.
2016-06-30 23:54:35 +08:00
fill.color = { 255, 255, 255, 255 };
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
Canvas::hFill hcf = canvas->GetFill(fill);
canvas->DrawPixmap(pm,
{ (double)x, (double)(flip ? Flip(y) - pm->height : y), 0.0 },
{ (double)pm->width, 0.0, 0.0 },
{ 0.0, (double)pm->height, 0.0 },
{ 0.0, 1.0 },
{ 1.0, 0.0 },
hcf);
}
Allow using z-index in UiCanvas, and use it for tooltips. Before this commit, tooltips in the text window are drawn under the red "X" indicating a disabled button. After this commit, they are moved on top of that. This commit also alters the OpenGL renderers' SetCamera() method to clear the depth buffer, as that would interfere with drawing the UI; the toolbar would get occluded by geometry.
2016-11-16 05:35:06 +08:00
void UiCanvas::DrawBitmapChar(char32_t codepoint, int x, int y, RgbaColor color, int zIndex) {
Use a dedicated BitmapFont instance per Canvas. Before this commit, updates to the bitmap font in the graphics window cause missing characters in the text window and vice versa. This is because BitmapFont contains a texture that's mutated, and sharing it would also require sharing display lists between GL contexts, which is not done (and overly complicated for this case anyway).
2017-01-10 10:52:51 +08:00
BitmapFont *font = canvas->GetBitmapFont();
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
Canvas::Fill fill = {};
Allow using z-index in UiCanvas, and use it for tooltips. Before this commit, tooltips in the text window are drawn under the red "X" indicating a disabled button. After this commit, they are moved on top of that. This commit also alters the OpenGL renderers' SetCamera() method to clear the depth buffer, as that would interfere with drawing the UI; the toolbar would get occluded by geometry.
2016-11-16 05:35:06 +08:00
fill.layer = Canvas::Layer::NORMAL;
fill.zIndex = zIndex;
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
fill.color = color;
Canvas::hFill hcf = canvas->GetFill(fill);
if(codepoint >= 0xe000 && codepoint <= 0xefff) {
// Special character, like a checkbox or a radio button
x -= 3;
}
double s0, t0, s1, t1;
size_t w, h;
Fix BitmapFont to not lose texture updates. Texture could also be updated by GetWidth(), which calls GetGlyph() internally, and then the next LocateGlyph() call would return false.
2016-06-30 23:54:35 +08:00
font->LocateGlyph(codepoint, &s0, &t0, &s1, &t1, &w, &h);
if(font->textureUpdated) {
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
// LocateGlyph modified the texture, reload it.
canvas->InvalidatePixmap(font->texture);
Fix BitmapFont to not lose texture updates. Texture could also be updated by GetWidth(), which calls GetGlyph() internally, and then the next LocateGlyph() call would return false.
2016-06-30 23:54:35 +08:00
font->textureUpdated = false;
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
}
canvas->DrawPixmap(font->texture,
{ (double)x, (double)Flip(y), 0.0 },
{ (double)w, 0.0, 0.0 },
{ 0.0, (double) h, 0.0 },
{ s0, t1 },
{ s1, t0 },
hcf);
}
Allow using z-index in UiCanvas, and use it for tooltips. Before this commit, tooltips in the text window are drawn under the red "X" indicating a disabled button. After this commit, they are moved on top of that. This commit also alters the OpenGL renderers' SetCamera() method to clear the depth buffer, as that would interfere with drawing the UI; the toolbar would get occluded by geometry.
2016-11-16 05:35:06 +08:00
void UiCanvas::DrawBitmapText(const std::string &str, int x, int y, RgbaColor color, int zIndex) {
Use a dedicated BitmapFont instance per Canvas. Before this commit, updates to the bitmap font in the graphics window cause missing characters in the text window and vice versa. This is because BitmapFont contains a texture that's mutated, and sharing it would also require sharing display lists between GL contexts, which is not done (and overly complicated for this case anyway).
2017-01-10 10:52:51 +08:00
BitmapFont *font = canvas->GetBitmapFont();
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
for(char32_t codepoint : ReadUTF8(str)) {
Allow using z-index in UiCanvas, and use it for tooltips. Before this commit, tooltips in the text window are drawn under the red "X" indicating a disabled button. After this commit, they are moved on top of that. This commit also alters the OpenGL renderers' SetCamera() method to clear the depth buffer, as that would interfere with drawing the UI; the toolbar would get occluded by geometry.
2016-11-16 05:35:06 +08:00
DrawBitmapChar(codepoint, x, y, color, zIndex);
Fix type conversion warnings.
2017-03-11 03:39:55 +08:00
x += (int)font->GetWidth(codepoint) * 8;
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
}
}
//-----------------------------------------------------------------------------
// A canvas that performs picking against drawn geometry.
//-----------------------------------------------------------------------------
void ObjectPicker::DoCompare(double distance, int zIndex, int comparePosition) {
if(distance > selRadius) return;
if((zIndex == maxZIndex && distance < minDistance) || (zIndex > maxZIndex)) {
minDistance = distance;
maxZIndex = zIndex;
position = comparePosition;
}
}
void ObjectPicker::DoQuad(const Vector &a, const Vector &b, const Vector &c, const Vector &d,
int zIndex, int comparePosition) {
Point2d corners[4] = {
camera.ProjectPoint(a),
camera.ProjectPoint(b),
camera.ProjectPoint(c),
camera.ProjectPoint(d)
};
double minNegative = VERY_NEGATIVE,
maxPositive = VERY_POSITIVE;
for(int i = 0; i < 4; i++) {
Point2d ap = corners[i],
bp = corners[(i + 1) % 4];
double distance = point.DistanceToLineSigned(ap, bp.Minus(ap), /*asSegment=*/true);
if(distance < 0) minNegative = std::max(minNegative, distance);
if(distance > 0) maxPositive = std::min(maxPositive, distance);
}
bool insideQuad = (minNegative == VERY_NEGATIVE || maxPositive == VERY_POSITIVE);
if(insideQuad) {
DoCompare(0.0, zIndex, comparePosition);
} else {
double distance = std::min(fabs(minNegative), fabs(maxPositive));
DoCompare(distance, zIndex, comparePosition);
}
}
void ObjectPicker::DrawLine(const Vector &a, const Vector &b, hStroke hcs) {
Stroke *stroke = strokes.FindById(hcs);
Point2d ap = camera.ProjectPoint(a);
Point2d bp = camera.ProjectPoint(b);
double distance = point.DistanceToLine(ap, bp.Minus(ap), /*asSegment=*/true);
DoCompare(distance - stroke->width / 2.0, stroke->zIndex);
}
void ObjectPicker::DrawEdges(const SEdgeList &el, hStroke hcs) {
Stroke *stroke = strokes.FindById(hcs);
for(const SEdge &e : el.l) {
Point2d ap = camera.ProjectPoint(e.a);
Point2d bp = camera.ProjectPoint(e.b);
double distance = point.DistanceToLine(ap, bp.Minus(ap), /*asSegment=*/true);
DoCompare(distance - stroke->width / 2.0, stroke->zIndex, e.auxB);
}
}
Unify displayEdges and displayOutlines. This has the following benefits: * Less geometry to generate; we can do both in one pass; * Less geometry to draw; * Eliminate overdraw of outlines on top of emphasized edges; * In future, being able to seamlessly stitch stippled lines. The contour edges are now also drawn before emphasized edges; this makes intersections of contour and emphasized edges look better as the thinner emphasized edge doesn't clobber the depth buffer.
2016-06-26 14:39:27 +08:00
void ObjectPicker::DrawOutlines(const SOutlineList &ol, hStroke hcs, DrawOutlinesAs drawAs) {
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
ssassert(false, "Not implemented");
}
void ObjectPicker::DrawVectorText(const std::string &text, double height,
const Vector &o, const Vector &u, const Vector &v,
hStroke hcs) {
Stroke *stroke = strokes.FindById(hcs);
double w = VectorFont::Builtin()-> GetWidth(height, text),
h = VectorFont::Builtin()->GetHeight(height);
DoQuad(o,
o.Plus(v.ScaledBy(h)),
o.Plus(u.ScaledBy(w)).Plus(v.ScaledBy(h)),
o.Plus(u.ScaledBy(w)),
stroke->zIndex);
}
void ObjectPicker::DrawQuad(const Vector &a, const Vector &b, const Vector &c, const Vector &d,
hFill hcf) {
Fill *fill = fills.FindById(hcf);
DoQuad(a, b, c, d, fill->zIndex);
}
Use Canvas::Stroke, not Canvas::Fill, for drawing a point. Our points are more like fat lines than actual quads, in that they are scale-invariant.
2016-07-17 22:46:38 +08:00
void ObjectPicker::DrawPoint(const Vector &o, Canvas::hStroke hcs) {
Stroke *stroke = strokes.FindById(hcs);
double distance = point.DistanceTo(camera.ProjectPoint(o)) - stroke->width / 2;
DoCompare(distance, stroke->zIndex);
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
}
void ObjectPicker::DrawPolygon(const SPolygon &p, hFill hcf) {
ssassert(false, "Not implemented");
}
Reimplement drawing of the mesh wireframe. OpenGL 2 and newer do not have the glPolygonMode(..., GL_LINES) API, so produce the wireframe on our side. It's somewhat slow, and draws every line three times, but it is cached when the OpenGL 2 renderer is used, and this should do for a debugging feature.
2016-11-18 00:47:45 +08:00
void ObjectPicker::DrawMesh(const SMesh &m, hFill hcfFront, hFill hcfBack) {
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
ssassert(false, "Not implemented");
}
void ObjectPicker::DrawFaces(const SMesh &m, const std::vector<uint32_t> &faces, hFill hcf) {
ssassert(false, "Not implemented");
}
void ObjectPicker::DrawPixmap(std::shared_ptr<const Pixmap> pm,
const Vector &o, const Vector &u, const Vector &v,
const Point2d &ta, const Point2d &tb, Canvas::hFill hcf) {
Implement an image request.
2016-11-30 00:49:20 +08:00
DrawQuad(o, o.Plus(u), o.Plus(u).Plus(v), o.Plus(v), hcf);
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
}
bool ObjectPicker::Pick(std::function<void()> drawFn) {
minDistance = VERY_POSITIVE;
Choose entities to select in a way appropriate for the operation. Before this commit, when an entity is clicked at or dragged, and it shares a place with other entities, which of them is selected is decided more or less at random. This is particularly annoying when dragging. After this commit, when clicking, an entity from the current group is given preference, and when dragging, an entity from a request is given preference. This allows e.g. dragging points of a sketch even when an extrusion of that sketch is active.
2016-07-22 01:58:18 +08:00
maxZIndex = INT_MIN;
Abstract all (ex-OpenGL) drawing operations into a Canvas interface. This has several desirable consequences: * It is now possible to port SolveSpace to a later version of OpenGL, such as OpenGLES 2, so that it runs on platforms that only have that OpenGL version; * The majority of geometry is now rendered without references to the camera in C++ code, so a renderer can now submit it to the video card once and re-rasterize with a different projection matrix every time the projection is changed, avoiding expensive reuploads; * The DOGD (draw or get distance) interface is now a straightforward Canvas implementation; * There are no more direct references to SS.GW.(projection) in sketch rendering code, which allows rendering to multiple viewports; * There are no more unnecessary framebuffer flips on CPU on Cocoa and GTK; * The platform-dependent GL code is now confined to rendergl1.cpp. * The Microsoft and Apple headers required by it that are prone to identifier conflicts are no longer included globally; * The rendergl1.cpp implementation can now be omitted from compilation to run SolveSpace headless or with a different OpenGL version. Note these implementation details of Canvas: * GetCamera currently always returns a reference to the field `Camera camera;`. This is so that a future renderer that caches geometry in the video memory can define it as asserting, which would provide assurance against code that could accidentally put something projection-dependent in the cache; * Line and triangle rendering is specified through a level of indirection, hStroke and hFill. This is so that a future renderer that batches geometry could cheaply group identical styles. * DrawPixmap and DrawVectorText accept a (o,u,v) and not a matrix. This is so that a future renderer into an output format that uses 2d transforms (e.g. SVG) could easily derive those. Some additional internal changes were required to enable this: * Pixmap is now always passed as std::shared_ptr<{const ,}Pixmap>. This is so that the renderer could cache uploaded textures between API calls, which requires it to capture a (weak) reference. * The PlatformPathEqual function was properly extracted into platform-specific code. This is so that the <windows.h> header could be included only where needed (in platform/w32* as well as rendergl1.cpp). * The SBsp{2,3}::DebugDraw functions were removed. They can be rewritten using the Canvas API if they are ever needed. While no visual changes were originally intended, some minor fixes happened anyway: * The "emphasis" yellow line from top-left corner is now correctly rendered much wider. * The marquee rectangle is now pixel grid aligned. * The hidden entities now do not clobber the depth buffer, removing some minor artifacts. * The workplane "tab" now scales with the font used to render the workplane name. * The workplane name font is now taken from the normals style. * Workplane and constraint line stipple is insignificantly different. This is so that it can reuse the existing stipple codepaths; rendering of workplanes and constraints predates those. Some debug functionality was added: * In graphics window, an fps counter that becomes red when rendering under 60fps is drawn.
2016-05-31 08:55:13 +08:00
drawFn();
return minDistance < selRadius;
}
}
Reference in New Issue Copy Permalink