Add function to generate benchmark results csv anywhere anyname defined

src/additional_tools/benchmark_macro.jl

@@ -1,12 +1,20 @@
 @doc draw"""
-    macro benchmarked(args...)
-    This macro is needed for getting both benchmark results and a return value from a function.
+    @benchmarked <expr to benchmark> [setup=<setup expr>] [other keyword parameters...]
+
+    Run benchmarked on a given expression.
+
+    This macro is a customization of BenchmarkTools' `@benchmark`.
+    `@benchmark` only provides benchmark results.
+    With this customization, we can benchmark a function that has a return value.
+    Customization uses internals from BenchmarkTools. 
+    Updates to this macro is needed when internals get changed from BenchmarkTools.
+
+    #Example
 """
 
 using BenchmarkTools
 using DataFrames
 
-
 macro benchmarked(args...)
     _, params = BenchmarkTools.prunekwargs(args...)
     bench, trial, result = gensym(), gensym(), gensym()
@@ -19,15 +27,35 @@ macro benchmarked(args...)
             $tune_phase
             local $trial, $result = $BenchmarkTools.run_result($bench)
             local $trialmin = $BenchmarkTools.minimum($trial)
-
-            display($trial)
+            local $trialallocs = $BenchmarkTools.allocs($trialmin) 
+
+             println(
+                "  ",
+                $BenchmarkTools.prettytime($BenchmarkTools.time($trialmin)),
+                " (",
+                $trialallocs,
+                " allocation",
+                $trialallocs == 1 ? "" : "s",
+                ": ",
+                $BenchmarkTools.prettymemory($BenchmarkTools.memory($trialmin)),
+                ")",
+            )
 
             $result, $trial
         end,
     )
 end
 
-
+function generate_benchmark_csv(path::String, bm_file, bm_results::BenchmarkTools.Trial)
+    bm_df = DataFrame(
+    time_ms = bm_results.times ./ 1e6,
+    gctime_ms = bm_results.gctimes ./ 1e6,
+    bm_nt_ms = (bm_results.times - bm_results.gctimes) ./ 1e6,
+    memory_mb = bm_results.memory ./ 1e6,
+    allocs = bm_results.allocs
+    )
+    CSV.write(joinpath(path, bm_file, bm_df)
+end