Currently, we have a single cursor implementation which must guard against / optimize for different scenarios such as:
- Multi-byte vs ASCII strings
- Lines with tabs
- etc
Perhaps we should instead switch to an interface with multiple specialized implementations so that we can optimize the line-parsing based on these factors without needing to have multiple repeated checks across the various functions.
Currently, we have a single cursor implementation which must guard against / optimize for different scenarios such as:
Perhaps we should instead switch to an interface with multiple specialized implementations so that we can optimize the line-parsing based on these factors without needing to have multiple repeated checks across the various functions.