Is there an existing issue or pull request for this?
Feature description
Currently, there are no standardized benchmarks to measure git-cliff's performance across changes and releases.
This makes it difficult to:
- Detect performance regressions early
- Compare different implementations or configurations objectively
- Validate optimizations with reproducible data
My investigations showed that real-world performance can vary significantly depending on execution environment and subprocess resolution, which further highlights the need for reliable benchmarks.
Desired solution
Introduce a benchmark suite to measure git-cliff's performance in a reproducible way.
Ideally, this would:
- Track execution time and resource usage across versions
- Be runnable in CI or locally in a consistent environment
- Make it easier to validate performance-related changes
One possible direction is to follow an approach similar to ratatui's benchmarking setup, which uses historical commits and realistic workloads.
Alternatives considered
- Relying on ad-hoc local timing (time, profiler tools)
- Useful for investigation, but not reproducible or comparable over time.
- Simple microbenchmarks
- Easier to set up, but may not reflect real-world usage or regressions.
Additional context
N/A
Is there an existing issue or pull request for this?
Feature description
Currently, there are no standardized benchmarks to measure git-cliff's performance across changes and releases.
This makes it difficult to:
My investigations showed that real-world performance can vary significantly depending on execution environment and subprocess resolution, which further highlights the need for reliable benchmarks.
Desired solution
Introduce a benchmark suite to measure git-cliff's performance in a reproducible way.
Ideally, this would:
One possible direction is to follow an approach similar to ratatui's benchmarking setup, which uses historical commits and realistic workloads.
Alternatives considered
Additional context
N/A