Open isabelle-dr opened 2 years ago
Issue by barbeau Dec 09, 2020 Originally opened as https://github.com/CUTR-at-USF/gtfs-realtime-validator/issues/392
Summary:
Benchmarks when working on gtfs-validator show that computeIfAbsent() is about 20% more expensive than calling get() and put().: https://github.com/MobilityData/gtfs-validator/pull/546/files#diff-ab042c46ddc495ec40e8c9a80fdd5ea27dfbac6582dc76a973e523dd34c7b0f9R57
We should optimize the RT validator as necessary.
Steps to reproduce:
Run the validator
Expected behavior:
Have ideal performance when adding to HashMap
Observed behavior:
Benchmarks in context of static GTFS show we're using a slower method
Issue by barbeau Dec 09, 2020 Originally opened as https://github.com/CUTR-at-USF/gtfs-realtime-validator/issues/392
Summary:
Benchmarks when working on gtfs-validator show that computeIfAbsent() is about 20% more expensive than calling get() and put().: https://github.com/MobilityData/gtfs-validator/pull/546/files#diff-ab042c46ddc495ec40e8c9a80fdd5ea27dfbac6582dc76a973e523dd34c7b0f9R57
We should optimize the RT validator as necessary.
Steps to reproduce:
Run the validator
Expected behavior:
Have ideal performance when adding to HashMap
Observed behavior:
Benchmarks in context of static GTFS show we're using a slower method