feat: add support for select... in judge

docs/src/jogger.md

@@ -88,8 +88,11 @@ Often it's useful to run only a subset of the full benchmarking suite.
 For example, to run only the benchmarks within `benchmark/bench_filename.jl`:
 `JogExample.run("bench_filename")`.
 
-Most jogger functions support filtering ([`judge`](@ref JogExample.judge) is a notable exception).
-Greater support is planned.
+```julia
+using PkgJogger, Example
+@jog Example
+JogExample.run("bench_filename.jl")
+```
 
 !!! compat PkgJogger 0.6.0
     Support for filtering via [`JogExample.suite`](@ref) was added in v0.6.0

src/jogger.jl

@@ -214,11 +214,14 @@ macro jog(pkg)
         load_benchmarks(id) = PkgJogger.load_benchmarks(joinpath(BENCHMARK_DIR, "trial"), id)
 
         """
-            judge(new, old; metric=Statistics.median, kwargs...)
+            judge(new, old, [select...]; metric=Statistics.median, kwargs...)
 
         Compares benchmarking results from `new` vs `old` for regressions/improvements
         using `metric` as a basis. Additional `kwargs` are passed to `BenchmarkTools.judge`
 
+        Optionally, filter results using `select...`, see [`$($mod_str).suite`](@ref) for
+        details.
+
         Identical to [`PkgJogger.judge`](@ref), but accepts any identifier supported by
         [`$($mod_str).load_benchmarks`](@ref)
 
@@ -230,6 +233,11 @@ macro jog(pkg)
         [...]
         ```
 
+        ```julia
+        # Only judge results in `bench_foo.jl`
+        $($mod_str).judge(:latest, :oldest, "bench_foo.jl")
+        ```
+
         ```julia
         # Judge results by UUID
         $($mod_str).judge("$(UUIDs.uuid4())", "$(UUIDs.uuid4())")
@@ -242,8 +250,8 @@ macro jog(pkg)
         [...]
         ```
         """
-        function judge(new, old; kwargs...)
-            PkgJogger.judge(_get_benchmarks(new), _get_benchmarks(old); kwargs...)
+        function judge(new, old, select...; kwargs...)
+            PkgJogger.judge(_get_benchmarks(new), _get_benchmarks(old), select...; kwargs...)
         end
         _get_benchmarks(b) = load_benchmarks(b)
         _get_benchmarks(b::Dict) = PkgJogger._get_benchmarks(b)

src/utils.jl

@@ -117,9 +117,9 @@ function judge(
     old_estimate = metric(old)
     BenchmarkTools.judge(new_estimate, old_estimate; kwargs...)
 end
-function judge(new, old; kwargs...)
-    new_results = _get_benchmarks(new)
-    old_results = _get_benchmarks(old)
+function judge(new, old, select...; kwargs...)
+    new_results = getsuite(_get_benchmarks(new), select...)
+    old_results = getsuite(_get_benchmarks(old), select...)
     judge(new_results, old_results; kwargs...)
 end
 

test/judge_tests.jl

@@ -12,9 +12,9 @@ function gen_example()
     return results, filename, dict, uuid
 end
 
-function test_judge(f, new, old)
-    @inferred f(new, old)
-    judgement = f(new, old)
+function test_judge(f, new, old, args...)
+    @inferred f(new, old, args...)
+    judgement = f(new, old, args...)
     @test typeof(judgement) <: BenchmarkGroup
     return judgement
 end
@@ -25,6 +25,7 @@ old = gen_example()
 
 @testset "JogPkgName.judge($(typeof(n)), $(typeof(o)))" for (n, o) in Iterators.product(new, old)
     test_judge(jogger.judge, n, o)
+    test_judge(jogger.judge, n, o, :, "1ms")
 end
 
 @testset "PkgJogger.judge($(typeof(n)), $(typeof(o)))" for (n, o) in Iterators.product(new[1:3], old[1:3])