perf(test): add gh run ingestion for memory hotspots (#60187)

* perf(test): add gh run ingestion for memory hotspots

* perf(test): harden gh run hotspot ingestion

scripts/test-update-memory-hotspots-sources.mjs

@@ -5,6 +5,8 @@ import path from "node:path";
 export function loadHotspotInputTexts({
   logPaths = [],
   ghJobs = [],
+  ghRuns = [],
+  ghRunJobMatches = [],
   readFileSyncImpl = fs.readFileSync,
   execFileSyncImpl = execFileSync,
 }) {
@@ -15,7 +17,66 @@ export function loadHotspotInputTexts({
       text: readFileSyncImpl(logPath, "utf8"),
     });
   }
-  for (const ghJobId of ghJobs) {
+  const normalizedRunJobMatches = ghRunJobMatches
+    .map((match) => (typeof match === "string" ? match.trim() : String(match ?? "").trim()))
+    .filter((match) => match.length > 0)
+    .map((match) => match.toLowerCase());
+  const shouldIncludeRunJob = (jobName) => {
+    if (normalizedRunJobMatches.length === 0) {
+      return true;
+    }
+    if (typeof jobName !== "string") {
+      return false;
+    }
+    const normalizedName = jobName.toLowerCase();
+    return normalizedRunJobMatches.some((match) => normalizedName.includes(match));
+  };
+  // Deduplicate explicit and run-derived job ids so repeated inputs do not refetch the same log.
+  const ghJobIds = new Set(
+    ghJobs
+      .map((jobId) => (typeof jobId === "string" ? jobId.trim() : String(jobId ?? "").trim()))
+      .filter((jobId) => jobId.length > 0),
+  );
+  for (const ghRunId of ghRuns) {
+    const normalizedRunId =
+      typeof ghRunId === "string" ? ghRunId.trim() : String(ghRunId ?? "").trim();
+    if (normalizedRunId.length === 0) {
+      continue;
+    }
+    let rawJobs;
+    try {
+      rawJobs = execFileSyncImpl("gh", ["run", "view", normalizedRunId, "--json", "jobs"], {
+        encoding: "utf8",
+        maxBuffer: 8 * 1024 * 1024,
+      });
+    } catch (error) {
+      throw new Error(
+        `[test-update-memory-hotspots] failed to fetch gh run ${normalizedRunId} jobs`,
+        { cause: error },
+      );
+    }
+    let jobs = [];
+    try {
+      const parsed = JSON.parse(rawJobs);
+      jobs = Array.isArray(parsed?.jobs) ? parsed.jobs : [];
+    } catch (error) {
+      throw new Error(
+        `[test-update-memory-hotspots] failed to parse gh run ${normalizedRunId} jobs json`,
+        { cause: error },
+      );
+    }
+    for (const job of jobs) {
+      if (!shouldIncludeRunJob(job?.name)) {
+        continue;
+      }
+      const jobId = job?.databaseId;
+      if (!Number.isFinite(jobId)) {
+        continue;
+      }
+      ghJobIds.add(String(jobId));
+    }
+  }
+  for (const ghJobId of ghJobIds) {
     inputs.push({
       sourceName: `gh-job-${String(ghJobId)}`,
       text: execFileSyncImpl("gh", ["run", "view", "--job", String(ghJobId), "--log"], {

scripts/test-update-memory-hotspots.mjs

@@ -19,6 +19,8 @@ if (process.argv.slice(2).includes("--help")) {
       "  --lane-prefix <prefix>     Additional lane prefixes to include (repeatable)",
       "  --log <path>               Memory trace log to ingest (repeatable, required)",
       "  --gh-job <id>              GitHub Actions job id to ingest via gh (repeatable)",
+      "  --gh-run <id>              GitHub Actions run id to ingest via gh (repeatable)",
+      "  --gh-run-job-match <text>  Filter gh-run jobs by name substring (repeatable)",
       "  --min-delta-kb <kb>        Minimum RSS delta to retain (default: 262144)",
       "  --limit <count>            Max hotspot entries to retain (default: 64)",
       "  --help                     Show this help text",
@@ -27,6 +29,7 @@ if (process.argv.slice(2).includes("--help")) {
       "  node scripts/test-update-memory-hotspots.mjs --log /tmp/unit-fast.log",
       "  node scripts/test-update-memory-hotspots.mjs --log a.log --log b.log --lane-prefix unit-fast-batch-",
       "  node scripts/test-update-memory-hotspots.mjs --gh-job 69804189668 --gh-job 69804189672",
+      "  node scripts/test-update-memory-hotspots.mjs --gh-run 23933168654 --gh-run-job-match extensions",
     ].join("\n"),
   );
   process.exit(0);
@@ -42,6 +45,8 @@ function parseArgs(argv) {
       lanePrefixes: [],
       logs: [],
       ghJobs: [],
+      ghRuns: [],
+      ghRunJobMatches: [],
       minDeltaKb: 256 * 1024,
       limit: 64,
     },
@@ -52,6 +57,8 @@ function parseArgs(argv) {
       stringListFlag("--lane-prefix", "lanePrefixes"),
       stringListFlag("--log", "logs"),
       stringListFlag("--gh-job", "ghJobs"),
+      stringListFlag("--gh-run", "ghRuns"),
+      stringListFlag("--gh-run-job-match", "ghRunJobMatches"),
       intFlag("--min-delta-kb", "minDeltaKb", { min: 1 }),
       intFlag("--limit", "limit", { min: 1 }),
     ],
@@ -95,8 +102,10 @@ function mergeHotspotEntry(aggregated, file, value) {
 
 const opts = parseArgs(process.argv.slice(2));
 
-if (opts.logs.length === 0 && opts.ghJobs.length === 0) {
-  console.error("[test-update-memory-hotspots] pass at least one --log <path> or --gh-job <id>.");
+if (opts.logs.length === 0 && opts.ghJobs.length === 0 && opts.ghRuns.length === 0) {
+  console.error(
+    "[test-update-memory-hotspots] pass at least one --log <path>, --gh-job <id>, or --gh-run <id>.",
+  );
   process.exit(2);
 }
 
@@ -107,7 +116,12 @@ if (existing) {
     mergeHotspotEntry(aggregated, file, value);
   }
 }
-for (const input of loadHotspotInputTexts({ logPaths: opts.logs, ghJobs: opts.ghJobs })) {
+for (const input of loadHotspotInputTexts({
+  logPaths: opts.logs,
+  ghJobs: opts.ghJobs,
+  ghRuns: opts.ghRuns,
+  ghRunJobMatches: opts.ghRunJobMatches,
+})) {
   const text = input.text;
   const summaries = parseMemoryTraceSummaryLines(text).filter((summary) =>
     matchesHotspotSummaryLane(summary.lane, opts.lane, opts.lanePrefixes),