I am a big fan of the finite state machine, the first time I got to know this was learning algorithms for LCS, and after that I become more and more fascinate about it.
This repository contains two implementation of validating the given time string is a valid time or not, for example: "20:04" is valid while "24+7" is not.
And I want to let you know how fast FSM can be.
The 1 million strings are generated randomly to feed as test input. And FSM implementation is about 3X as fast as the common one.
it_works is to prove that fsm and common produce identical results.
✔ ~/projects/github/fsm-perf [main|✔]
pseudoc $ cargo test --tests --release -- --show-output
Finished release [optimized] target(s) in 0.00s
Running unittests src/lib.rs (target/release/deps/fsm_perf-ce885a8409f4abb3)
running 3 tests
test tests::fsm ... ok
test tests::it_works ... ok
test tests::common ... ok
successes:
---- tests::fsm stdout ----
The test took 26.386528ms to finish.
---- tests::common stdout ----
The test took 74.52521ms to finish.
successes:
tests::common
tests::fsm
tests::it_works
test result: ok. 3 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.20s