docs: drop the Go port framing, stand on its own - #4
Merged
Conversation
The README explained this project as a port of its Go sibling, which made the Go repository a prerequisite for reading it. Nothing here depends on that repository, so the framing only sent readers somewhere else. The intro now names pytest-benchmark as the metric source directly. The status note, the decision-procedure note and the Windows limits section make the same points without the comparison; two code comments that justified a choice by pointing at the Go implementation now carry the reasoning themselves. A test guards the README against the framing coming back. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VaGHC3qsuNHw3gX8EvGgxY
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VaGHC3qsuNHw3gX8EvGgxY
The hoisted constant put "GO" in the vocabulary of a project that has none. Two of its entries are not Go anyway — config.yaml and allocs_delta are leftovers of an earlier config format and metric set. What the guard actually checks is that the agent-facing template and the README name nothing absent from this project, so ABSENT_NAMES says that; the README's prose check moves to PORT_FRAMING beside it. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VaGHC3qsuNHw3gX8EvGgxY
The benchtime error told the user their value 'uses go test's fixed-iteration-count form' — a Go reference in output this project prints at runtime. The form is refused for its own reasons, which the rest of the message already gives, so naming another tool's flag adds nothing. Three docstrings explained a gate or a fixture by analogy to go build and go test; they now say what the thing does. The guard extends over src/ as well, since error messages reach users the same way the README does. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VaGHC3qsuNHw3gX8EvGgxY
g4lb
added a commit
that referenced
this pull request
Sep 10, 2026
The "4" in out check for the experiment total also matched row #4's listing and the 0.4500 cumulative figure, so it passed even against a hardcoded count. Assert the full rendered fields instead.
g4lb
added a commit
that referenced
this pull request
Sep 10, 2026
The README explained this project as a port of its Go sibling, and a runtime error message told the user their benchtime value used "go test's" flag form. Nothing here depends on that project, so the references only sent readers somewhere else. - README: pytest-benchmark named directly as the metric source; the status note, decision-procedure note and Windows limits section make their points without the comparison. - config.py: the benchtime error names the fixed-iteration-count form without attributing it to another tool. - runner.py and two test docstrings say what a gate or fixture does rather than which Go command it mirrors. - A guard over the template, the README and src/ keeps absent names and port framing from coming back. Docs, comments and one error string; no behavior change.
g4lb
added a commit
that referenced
this pull request
Sep 11, 2026
The README explained this project as a port of its Go sibling, and a runtime error message told the user their benchtime value used "go test's" flag form. Nothing here depends on that project, so the references only sent readers somewhere else. - README: pytest-benchmark named directly as the metric source; the status note, decision-procedure note and Windows limits section make their points without the comparison. - config.py: the benchtime error names the fixed-iteration-count form without attributing it to another tool. - runner.py and two test docstrings say what a gate or fixture does rather than which Go command it mirrors. - A guard over the template, the README and src/ keeps absent names and port framing from coming back. Docs, comments and one error string; no behavior change.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The README explained this project as a port of its Go sibling, which made that repository a prerequisite for reading it. Nothing here depends on it, so the framing only sent readers somewhere else.
README
go test -benchandns/op.Ported from the Go original:becomesAccepted limits of the design:.Code comments
stats.pyjustified implementing the statistics here by reference togolang.org/x/perf/benchmath; it now gives the reason on its own terms (no scipy dependency, small-sample behavior under this project's tests).test_e2e.pyjustified the raisedmin_effect_pctfloor by citing the Go original's 15%; it now explains the floor from this fixture's own two effects and the gap between them.Test
test_readme_names_no_go_artifactsguards the README against the framing returning. The existing token list is hoisted into a sharedGO_ARTIFACTSconstant and extended for the README withGo original,Go sibling,Go harness, andPython port.Docs and comments only — no behavior change. Full suite: 509 passed, 2 skipped.
🤖 Generated with Claude Code
https://claude.ai/code/session_01VaGHC3qsuNHw3gX8EvGgxY