Bug 149944 – Extend run-benchmark script to support human-readable results conversion.

• Component: New Bugs
• Status: RESOLVED FIXED

Extend run-benchmark script to support human-readable results conversion.

Created attachment 262740 [details]
Patch

[Ryosuke Niwa]

Comment on attachment 262740 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=262740&action=review

> Tools/Scripts/webkitpy/benchmark_runner/benchmark_results.py:98
> +            return ('{mean:.3f}{unit} stdev={delta:.1%}').format(
> +                mean=mean, delta=sample_stdev / mean, unit=unit)

This should fit in one line.

> Tools/Scripts/webkitpy/benchmark_runner/run_benchmark.py:27
> +    parser.add_argument('--raw', action='store_true')

I don't think this is a descriptive name.
I'd call this --no-adjust-unit or --no-scale-unit and store the value in scaleUnit which defaults to True.

> Tools/Scripts/webkitpy/benchmark_runner/run_benchmark.py:29
> +    mutual_group.add_argument('--format-result', dest='format_result', help='Specify file you want to format')

--format-results sound as if we wouldn't format results if we don't set this option.
--read-results-json?  --results-json?

Created attachment 262744 [details]
Patch

[WebKit Commit Bot]

Comment on attachment 262744 [details]
Patch

Clearing flags on attachment: 262744

Committed r190779: <http://trac.webkit.org/changeset/190779>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Csaba Osztrogonác]

(In reply to comment #4)
> Comment on attachment 262744 [details]
> Patch
> 
> Clearing flags on attachment: 262744
> 
> Committed r190779: <http://trac.webkit.org/changeset/190779>

It broke 6 webkitpy tests, see bug149949 for details.