Cache CLI extractor paths across Actions steps

src/codeql.ts

@@ -736,21 +736,26 @@ async function getCodeQLForCmd(
       await runCli(cmd, args);
     },
     async resolveLanguages() {
-      const codeqlArgs = [
-        "resolve",
-        "languages",
-        "--format=json",
-        ...getExtraOptionsFromEnv(["resolve", "languages"]),
-      ];
-      const output = await runCli(cmd, codeqlArgs);
+      let result = util.getCachedCodeQlResolveLanguages(cmd);
+      if (result === undefined) {
+        const codeqlArgs = [
+          "resolve",
+          "languages",
+          "--format=json",
+          ...getExtraOptionsFromEnv(["resolve", "languages"]),
+        ];
+        const output = await runCli(cmd, codeqlArgs);
 
-      try {
-        return JSON.parse(output) as ResolveLanguagesOutput;
-      } catch (e) {
-        throw new Error(
-          `Unexpected output from codeql resolve languages: ${e}`,
-        );
+        try {
+          result = JSON.parse(output) as ResolveLanguagesOutput;
+        } catch (e) {
+          throw new Error(
+            `Unexpected output from codeql resolve languages: ${e}`,
+          );
+        }
+        util.cacheCodeQlResolveLanguages(cmd, result);
       }
+      return result;
     },
     async betterResolveLanguages(
       {
@@ -969,32 +974,11 @@ async function getCodeQLForCmd(
       await new toolrunner.ToolRunner(cmd, args).exec();
     },
     async resolveExtractor(language: Language): Promise<string> {
-      // Request it using `format=json` so we don't need to strip the trailing new line generated by
-      // the CLI.
-      let extractorPath = "";
-      await new toolrunner.ToolRunner(
-        cmd,
-        [
-          "resolve",
-          "extractor",
-          "--format=json",
-          `--language=${language}`,
-          "--extractor-include-aliases",
-          ...getExtraOptionsFromEnv(["resolve", "extractor"]),
-        ],
-        {
-          silent: true,
-          listeners: {
-            stdout: (data) => {
-              extractorPath += data.toString();
-            },
-            stderr: (data) => {
-              process.stderr.write(data);
-            },
-          },
-        },
-      ).exec();
-      return JSON.parse(extractorPath) as string;
+      // A previous implementation executed `codeql resolve extractor`.
+      // This can be a bit slow due to the JVM startup cost. Instead, get
+      // the extractor path from resolveLanguages(), which caches its output.
+      const extractors = await this.resolveLanguages();
+      return extractors[language][0];
     },
     async resolveQueriesStartingPacks(queries: string[]): Promise<string[]> {
       const codeqlArgs = [

src/environment.ts

@@ -23,6 +23,12 @@ export enum EnvVar {
    */
   CODEQL_VERSION_INFO = "CODEQL_ACTION_CLI_VERSION_INFO",
 
+  /**
+   * `ResolveLanguagesOutput` for the CodeQL CLI, so later Actions steps can reuse it instead of
+   * invoking `codeql resolve languages` again.
+   */
+  CODEQL_RESOLVE_LANGUAGES = "CODEQL_ACTION_CLI_RESOLVE_LANGUAGES",
+
   /** Whether the CodeQL Action has invoked the Go autobuilder. */
   DID_AUTOBUILD_GOLANG = "CODEQL_ACTION_DID_AUTOBUILD_GOLANG",
 

src/util.ts

@@ -10,7 +10,7 @@ import * as yaml from "js-yaml";
 import * as semver from "semver";
 
 import * as apiCompatibility from "./api-compatibility.json";
-import type { CodeQL, VersionInfo } from "./codeql";
+import type { CodeQL, VersionInfo, ResolveLanguagesOutput } from "./codeql";
 import type { Pack } from "./config/db-config";
 import type { Config } from "./config-utils";
 import { EnvVar } from "./environment";
@@ -701,6 +701,75 @@ export function getCachedCodeQlVersion(cmd?: string): undefined | VersionInfo {
   return cachedCodeQlVersion;
 }
 
+let cachedCodeQlResolveLanguages: undefined | ResolveLanguagesOutput =
+  undefined;
+
+interface PersistedResolveLanguagesOutput {
+  cmd: string;
+  output: ResolveLanguagesOutput;
+}
+
+export function cacheCodeQlResolveLanguages(
+  cmd: string,
+  output: ResolveLanguagesOutput,
+): void {
+  if (cachedCodeQlResolveLanguages !== undefined) {
+    throw new Error("cacheCodeQlResolveLanguages() should be called only once");
+  }
+  cachedCodeQlResolveLanguages = output;
+  // Persist the output so that subsequent Actions steps, which run in separate
+  // processes, can reuse it rather than invoking `codeql resolve languages` again. We
+  // record the CLI path so that a different step using a different CodeQL bundle
+  // doesn't pick up a stale output.
+  core.exportVariable(
+    EnvVar.CODEQL_RESOLVE_LANGUAGES,
+    JSON.stringify({ cmd, output }),
+  );
+}
+
+function isPersistedResolveLanguagesOutput(
+  value: unknown,
+): value is PersistedResolveLanguagesOutput {
+  return (
+    typeof value === "object" &&
+    value !== null &&
+    typeof (value as Record<string, unknown>).cmd === "string" &&
+    typeof (value as Record<string, unknown>).output === "object" &&
+    (value as Record<string, unknown>).output !== null
+  );
+}
+
+export function getCachedCodeQlResolveLanguages(
+  cmd?: string,
+): undefined | ResolveLanguagesOutput {
+  if (cachedCodeQlResolveLanguages !== undefined) {
+    return cachedCodeQlResolveLanguages;
+  }
+  // Fall back to the value persisted by an earlier Actions step, if any. This is
+  // best-effort: any malformed or mismatched value is ignored so that the caller
+  // invokes `codeql resolve languages` instead.
+  const serialized = process.env[EnvVar.CODEQL_RESOLVE_LANGUAGES];
+  if (!serialized) {
+    return undefined;
+  }
+  let persisted: unknown;
+  try {
+    persisted = JSON.parse(serialized);
+  } catch {
+    return undefined;
+  }
+  if (
+    !isPersistedResolveLanguagesOutput(persisted) ||
+    (cmd !== undefined && persisted.cmd !== cmd)
+  ) {
+    return undefined;
+  }
+  // Memoize the parsed value so that subsequent calls in this process don't
+  // re-parse the environment variable.
+  cachedCodeQlResolveLanguages = persisted.output;
+  return cachedCodeQlResolveLanguages;
+}
+
 export async function codeQlVersionAtLeast(
   codeql: CodeQL,
   requiredVersion: string,