Benchmarks for generated parser
See #49
Here is a list of benchmarks to make and remake into figures:
- Prototype benchmarks with 8.6 and latest GHC 8.10 (improved GC). Both time and memory use
- Benchmarks of parsers with and without lazy output constructionn(3 vs 3A etc)
- Benchmarks with and without null byte guardian (Xeno VectorisedString class.)
- Comparison of Xeno, XML TypeLift, pugixml, and hexml on some XML document.
Maybe we can make shell script to make .csv
input for each one, and the I make gnuplot publication-ready charts.
Edited by Dmitry Krylov