This adds UTF-8 variants of the rasterize functions.
- They are slower than their UTF-32 counterpart, since they convert from UTF-8 to UTF-32 every time they're called
- They: do very basic codepoint validation, but does not guarantee all invalid UTF-8 sequences are detected. This isn't much different from the UTF-32 variants, where we also do not try to detect, or reject, invalid codepoints.
- If an invalid sequence is detected, nothing is rasterized and NULL is returned.
Closes #46
This adds UTF-8 variants of the rasterize functions.
* They are slower than their UTF-32 counterpart, since they convert from UTF-8 to UTF-32 every time they're called
* They: do very basic codepoint validation, but does not guarantee all invalid UTF-8 sequences are detected. This isn't much different from the UTF-32 variants, where we also do not try to detect, or reject, invalid codepoints.
* **If** an invalid sequence is detected, nothing is rasterized and NULL is returned.
Closes #46