You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm satisfied with current provisioning, tests and benchmark chain, written in POSIX shell script and Lua. Using unit-test library like Busted and/or provisioning tool like Ansible would be more modern, but this is Lua world, where minimalism is appreciated. Surely, I want to add Windows support, but I want to keep busybox support as well.
For plotting, YAML-formatted benchmark results are generated by Lua. It is converted to gnuplot input by a Ruby script. Changing the intermediate format to JSON and using Lunajson itself for plotting have some beauty, but who cares. I want to escape from Python at a hobby-project, because I'm working on so-called "data science" for a living.
The text was updated successfully, but these errors were encountered:
Now Lunajson uses matplotlib to plot benchmark results. Benchmark procedure is simplified and automated. (I hadn't personally liked Python, but now I'm forgetting Ruby...)
I'll add some docs about testing and benchmark to README
Minor: typo in issue title: s/docuements/documents/
grafi-tt
changed the title
add docuements about test, benchmark, plotting and utility scripts
add documents about test, benchmark, plotting and utility scripts
Nov 5, 2023
Some usage information should be added.
I'm satisfied with current provisioning, tests and benchmark chain, written in POSIX shell script and Lua. Using unit-test library like Busted and/or provisioning tool like Ansible would be more modern, but this is Lua world, where minimalism is appreciated. Surely, I want to add Windows support, but I want to keep busybox support as well.
For plotting, YAML-formatted benchmark results are generated by Lua. It is converted to gnuplot input by a Ruby script. Changing the intermediate format to JSON and using Lunajson itself for plotting have some beauty, but who cares. I want to escape from Python at a hobby-project, because I'm working on so-called "data science" for a living.
The text was updated successfully, but these errors were encountered: