In the latest refactoring pass, I focused on simplifying and solidifying the rendering pipeline by introducing a dedicated PIXEL class. The goal was to make pixel handling explicit and predictable, instead of spreading low level color and attribute logic across multiple stages.
The PIXEL abstraction now acts as the smallest atomic unit in the pipeline, encapsulating color, alpha, and character data in a single structure. This change reduced coupling between rendering steps and made the pipeline easier to reason about, debug, and extend.
A direct consequence of this refactor was a more structured approach to Unicode character handling. Supporting extended character sets required introducing an explicit font dependency, rather than relying on implicit or system level assumptions. Each rendered character now maps cleanly to a font glyph, making Unicode behavior deterministic and consistent across platforms.
This refactoring improved clarity, reduced edge case bugs, and laid the groundwork for future features such as richer text rendering and better localization support.
More iterations ahead, but the pipeline is now simpler, stricter, and more honest about what it actually needs to work.