I would like to have a PGraphics / framebuffer stack that works analogously to the matrix stack such that all the default drawing functions apply to the framebuffer currently at the top of the stack without having to reference it explicitly.
All too often, I write code that draws some complex object to the screen and then later end up wanting to draw it to a PGraphics instead so that I can, for instance, generate the image at a much higher resolution. This requires re-writing the code to pass in a PGraphics parameter (or using a global variable) and changing all of the drawing functions to use that PGraphics instead, such as line() to pg.line(). It's messy, error-prone (I always seem to miss changing a width to pg.width), and clutters up the code with all the pg. prefixes.
A far simpler mechanism would use a framebuffer stack with a pushFB( PGraphics pg ) / popFB() pair such that all drawing commands (and width and height constants) apply only to the framebuffer at the top of the stack. pushFB( pg ) would call endDraw() on the current top of stack, push pg on the stack, and then pg.beginDraw(). popFB() would endDraw() the top FB, pop it from the stack, and beginDraw() the new top. The stack is initialized with the default framebuffer and popping it either does nothing or throws an error.
All drawing functions would then be modified (entirely internally and invisible to the users, without any changes to the current API) to apply to the top-of-stack framebufffer. Anyone not using the FB stack wouldn't even know it's there, exactly analogous with the matrix stack.
I would like to have a PGraphics / framebuffer stack that works analogously to the matrix stack such that all the default drawing functions apply to the framebuffer currently at the top of the stack without having to reference it explicitly.
All too often, I write code that draws some complex object to the screen and then later end up wanting to draw it to a PGraphics instead so that I can, for instance, generate the image at a much higher resolution. This requires re-writing the code to pass in a PGraphics parameter (or using a global variable) and changing all of the drawing functions to use that PGraphics instead, such as
line()
topg.line()
. It's messy, error-prone (I always seem to miss changing awidth
topg.width
), and clutters up the code with all thepg.
prefixes.A far simpler mechanism would use a framebuffer stack with a
pushFB( PGraphics pg )
/popFB()
pair such that all drawing commands (andwidth
andheight
constants) apply only to the framebuffer at the top of the stack.pushFB( pg )
would callendDraw()
on the current top of stack, pushpg
on the stack, and thenpg.beginDraw()
.popFB()
wouldendDraw()
the top FB, pop it from the stack, andbeginDraw()
the new top. The stack is initialized with the default framebuffer and popping it either does nothing or throws an error.All drawing functions would then be modified (entirely internally and invisible to the users, without any changes to the current API) to apply to the top-of-stack framebufffer. Anyone not using the FB stack wouldn't even know it's there, exactly analogous with the matrix stack.