treat zero benchmark baselines as missing

clabby · clabby · commit 7e7a952963b0 · 2026-05-04T14:24:02.000-04:00
diff --git a/.github/scripts/benchmark-tracking.py b/.github/scripts/benchmark-tracking.py
@@ -314,7 +314,7 @@ def compare_results(
     comparisons = []
     for result in current:
         baseline = find_baseline(baseline_data, result["baseline_suite"], result["name"])
-        if baseline is None:
+        if baseline is None or baseline["value"] == 0:
             comparisons.append(
                 {
                     **result,

Original file line number	Diff line number	Diff line change
`@@ -314,7 +314,7 @@ def compare_results(`
`314`	`314`	`comparisons = []`
`315`	`315`	`for result in current:`
`316`	`316`	`baseline = find_baseline(baseline_data, result["baseline_suite"], result["name"])`
`317`		`- if baseline is None:`
	`317`	`+ if baseline is None or baseline["value"] == 0:`
`318`	`318`	`comparisons.append(`
`319`	`319`	`{`
`320`	`320`	`**result,`