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
{{ message }}
This repository was archived by the owner on Mar 24, 2026. It is now read-only.
Use an exact version, minimum MAJOR.MINOR (and the .PATCH if not use semver)
Not permitted images: :latest, :testing, :rolling, :nightly, :devel, ...
And also :slim, :alpine, ... without any version. As by default is :latest without any version.
Perhaps we can include a github action to check it.
Example:
❌ rust without any version, by default is latest !!!
✅ rust:1.93
❌ rust:slim is also latest.
✅ rust:1.93-slim
Headers
No constants or pre-calculated header for the Content-Length.
Need to be calculated for each request.
Only permitted in the plaintext test.
Header for Server need minimum 3 ( or 4 ? ) characters.
A lot of frameworks only use 1 character only for the bench (not realistic).
We can add it in the tests.
Fortune test
The new fortune object must be constructed in the scope of the request, not as a constant:
Within the scope of the request, a new Fortune object must be
constructed and added to the list.
Another suggestion, no framework can separate the plaintext/json bench from the database.
Please use the same variant for all the tests, the runs will be faster. Thank you !!
With a github action we can check it.
Docker images
Use an exact version, minimum MAJOR.MINOR (and the .PATCH if not use semver)
Not permitted images:
:latest,:testing,:rolling,:nightly,:devel, ...And also
:slim,:alpine, ... without any version. As by default is:latestwithout any version.Perhaps we can include a github action to check it.
Example:
❌
rustwithout any version, by default is latest !!!✅
rust:1.93❌
rust:slimis also latest.✅
rust:1.93-slimHeaders
Content-Length.Need to be calculated for each request.
Only permitted in the plaintext test.
Serverneed minimum 3 ( or 4 ? ) characters.A lot of frameworks only use 1 character only for the bench (not realistic).
We can add it in the tests.
Fortune test
The new fortune object must be constructed in the scope of the request, not as a constant:
https://github.com/TechEmpower/FrameworkBenchmarks/wiki/Project-Information-Framework-Tests-Overview#fortunes
by @p8
Join plaintext/json with database tests
Please use the same variant for all the tests, the runs will be faster. Thank you !!
With a github action we can check it.
Add maintainers
benchmark_config.jsonto be notified with any change.We can add it in the github action than now send the notifications, if no maintainers are found, please ask to add maintainers in the PRs.
https://github.com/TechEmpower/FrameworkBenchmarks/blob/master/toolset/github_actions/get_maintainers.py#L56-L57
And update https://github.com/TechEmpower/FrameworkBenchmarks/blob/master/toolset/scaffolding/benchmark_config.jsonMade in [ci-skip] Add maintainers and tags fields to benchmark config #10721Open to more discussion about that !!