Currently parsing creates a lot of new strings as it goes. Might it be possible to avoid this by storing start/end positions instead? How would that affect memory usage and performance?
Currently parsing creates a lot of new strings as it goes.
Might it be possible to avoid this by storing start/end positions instead?
How would that affect memory usage and performance?