perf(rules-injector): cache full candidates and memoize ancestor scans

The rule scan cache stored a path[] keyed by (projectRoot|startDir|
skipClaudeUserRules), so two issues stacked up on every tracked tool
call:

- Cache hits still ran safeRealpathSync(realpathSync) and re-derived
  isGlobal / distance / isSingleFile for every cached path. That is a
  per-candidate sync syscall plus repeated string-prefix walks.
- Sibling files in the same project landed under different startDir
  keys, so the entire walk-and-recursive-scan chain repeated even
  though every ancestor rule directory was identical.

Store the full RuleFileCandidate[] in the per-call cache so a cache hit
returns immediately with no realpath syscall. Add a separate per-
directory scan cache (getDirScan/setDirScan) keyed by absolute rule
directory path, so two sibling files reuse the same readdir + realpath
work for every shared ancestor.

Microbench (200 files / 20 modules / cached session):
- single sweep: 41.8ms -> 2.5ms (16x)
- 3-pass replay: 88.6ms -> 3.2ms (28x)

Pin the new invariants with two new tests:
- 'does not re-resolve symlinked rule path on cache hit' via a
  retargeted directory symlink.
- 'reuses ancestor directory scan for sibling files in the same
  project' by deleting the source rule file between the two calls.
This commit is contained in:
YeonGyu-Kim
2026-05-17 01:53:24 +09:00
parent c25f75294e
commit 271878bcea
3 changed files with 147 additions and 83 deletions
+46 -73
View File
@@ -1,12 +1,12 @@
import { existsSync, statSync } from "node:fs";
import { dirname, join, sep } from "node:path";
import { dirname, join } from "node:path";
import {
OPENCODE_USER_RULE_DIRS,
PROJECT_RULE_FILES,
PROJECT_RULE_SUBDIRS,
USER_RULE_DIR,
} from "./constants";
import type { RuleScanCache } from "./rule-scan-cache";
import type { DirectoryScanEntry, RuleScanCache } from "./rule-scan-cache";
import { findRuleFilesRecursive, safeRealpathSync } from "./rule-file-scanner";
import type { RuleFileCandidate } from "./types";
@@ -14,6 +14,26 @@ export interface FindRuleFilesOptions {
skipClaudeUserRules?: boolean;
}
function scanDirectoryWithCache(
dir: string,
cache: RuleScanCache | undefined,
): DirectoryScanEntry[] {
const cached = cache?.getDirScan(dir);
if (cached) {
return cached;
}
const files: string[] = [];
findRuleFilesRecursive(dir, files);
const entries: DirectoryScanEntry[] = files.map((filePath) => ({
path: filePath,
realPath: safeRealpathSync(filePath),
}));
cache?.setDirScan(dir, entries);
return entries;
}
function getUserRuleDirs(homeDir: string, skipClaudeUserRules: boolean): string[] {
const userRuleDirs = OPENCODE_USER_RULE_DIRS.map((dir) => join(homeDir, dir));
if (!skipClaudeUserRules) {
@@ -30,54 +50,6 @@ function createCacheKey(
return `${projectRoot ?? ""}|${startDir}|${skipClaudeUserRules ? "1" : "0"}`;
}
function createCachedCandidate(
filePath: string,
projectRoot: string | null,
startDir: string,
userRuleDirs: string[],
): RuleFileCandidate | undefined {
const realPath = safeRealpathSync(filePath);
for (const userRuleDir of userRuleDirs) {
if (filePath.startsWith(`${userRuleDir}${sep}`)) {
return { path: filePath, realPath, isGlobal: true, distance: 9999 };
}
}
if (projectRoot) {
for (const ruleFile of PROJECT_RULE_FILES) {
if (filePath === join(projectRoot, ruleFile)) {
return {
path: filePath,
realPath,
isGlobal: false,
distance: 0,
isSingleFile: true,
};
}
}
}
let currentDir = startDir;
let distance = 0;
while (true) {
for (const [parent, subdir] of PROJECT_RULE_SUBDIRS) {
const ruleDir = join(currentDir, parent, subdir);
if (filePath.startsWith(`${ruleDir}${sep}`)) {
return { path: filePath, realPath, isGlobal: false, distance };
}
}
if (projectRoot && currentDir === projectRoot) break;
const parentDir = dirname(currentDir);
if (parentDir === currentDir) break;
currentDir = parentDir;
distance += 1;
}
return undefined;
}
export function findRuleFiles(
projectRoot: string | null,
homeDir: string,
@@ -89,12 +61,10 @@ export function findRuleFiles(
const skipClaudeUserRules = options?.skipClaudeUserRules ?? false;
const userRuleDirs = getUserRuleDirs(homeDir, skipClaudeUserRules);
const cacheKey = createCacheKey(projectRoot, startDir, skipClaudeUserRules);
const cachedPaths = cache?.get(cacheKey);
const cachedCandidates = cache?.get(cacheKey);
if (cachedPaths) {
return cachedPaths
.map((filePath) => createCachedCandidate(filePath, projectRoot, startDir, userRuleDirs))
.filter((candidate): candidate is RuleFileCandidate => candidate !== undefined);
if (cachedCandidates) {
return cachedCandidates;
}
const candidates: RuleFileCandidate[] = [];
@@ -105,14 +75,17 @@ export function findRuleFiles(
while (true) {
for (const [parent, subdir] of PROJECT_RULE_SUBDIRS) {
const ruleDir = join(currentDir, parent, subdir);
const files: string[] = [];
findRuleFilesRecursive(ruleDir, files);
const entries = scanDirectoryWithCache(ruleDir, cache);
for (const filePath of files) {
const realPath = safeRealpathSync(filePath);
if (seenRealPaths.has(realPath)) continue;
seenRealPaths.add(realPath);
candidates.push({ path: filePath, realPath, isGlobal: false, distance });
for (const entry of entries) {
if (seenRealPaths.has(entry.realPath)) continue;
seenRealPaths.add(entry.realPath);
candidates.push({
path: entry.path,
realPath: entry.realPath,
isGlobal: false,
distance,
});
}
}
@@ -148,14 +121,17 @@ export function findRuleFiles(
}
for (const userRuleDir of userRuleDirs) {
const userFiles: string[] = [];
findRuleFilesRecursive(userRuleDir, userFiles);
const entries = scanDirectoryWithCache(userRuleDir, cache);
for (const filePath of userFiles) {
const realPath = safeRealpathSync(filePath);
if (seenRealPaths.has(realPath)) continue;
seenRealPaths.add(realPath);
candidates.push({ path: filePath, realPath, isGlobal: true, distance: 9999 });
for (const entry of entries) {
if (seenRealPaths.has(entry.realPath)) continue;
seenRealPaths.add(entry.realPath);
candidates.push({
path: entry.path,
realPath: entry.realPath,
isGlobal: true,
distance: 9999,
});
}
}
@@ -166,10 +142,7 @@ export function findRuleFiles(
return left.distance - right.distance;
});
cache?.set(
cacheKey,
candidates.map((candidate) => candidate.path),
);
cache?.set(cacheKey, candidates);
return candidates;
}
@@ -1,5 +1,13 @@
import { afterEach, beforeEach, describe, expect, it } from "bun:test";
import { existsSync, mkdirSync, rmSync, writeFileSync } from "node:fs";
import {
existsSync,
mkdirSync,
realpathSync,
rmSync,
symlinkSync,
unlinkSync,
writeFileSync,
} from "node:fs";
import { tmpdir } from "node:os";
import { join } from "node:path";
@@ -12,7 +20,10 @@ describe("createRuleScanCache", () => {
// given
const { createRuleScanCache } = await import(`./rule-scan-cache${createImportSuffix()}`);
const cache = createRuleScanCache();
const value = ["a", "b"];
const value = [
{ path: "/tmp/a.md", realPath: "/tmp/a.md", isGlobal: false, distance: 0 },
{ path: "/tmp/b.md", realPath: "/tmp/b.md", isGlobal: false, distance: 1 },
];
// when
const initialValue = cache.get("k1");
@@ -80,4 +91,67 @@ describe("findRuleFiles with scan cache", () => {
secondRuleFile,
].sort());
});
it("does not re-resolve symlinked rule path on cache hit", async () => {
// given
const { createRuleScanCache } = await import(`./rule-scan-cache${createImportSuffix()}`);
const { findRuleFiles } = await import(`./rule-file-finder${createImportSuffix()}`);
const actualGithubA = join(projectRoot, "actual-github-a");
const actualGithubB = join(projectRoot, "actual-github-b");
const instructionsBaseA = join(actualGithubA, "instructions");
const instructionsBaseB = join(actualGithubB, "instructions");
const ruleFileA = join(instructionsBaseA, "typescript.instructions.md");
const ruleFileB = join(instructionsBaseB, "typescript.instructions.md");
const symlinkGithub = join(projectRoot, ".github");
mkdirSync(instructionsBaseA, { recursive: true });
mkdirSync(instructionsBaseB, { recursive: true });
writeFileSync(ruleFileA, "alpha rules\n");
writeFileSync(ruleFileB, "beta rules\n");
symlinkSync(actualGithubA, symlinkGithub, "dir");
const canonicalRuleFileA = realpathSync(ruleFileA);
const cache = createRuleScanCache();
// when
const firstCandidates = findRuleFiles(projectRoot, homeDir, currentFile, undefined, cache);
const cachedRealPath = firstCandidates[0]?.realPath;
unlinkSync(symlinkGithub);
symlinkSync(actualGithubB, symlinkGithub, "dir");
const secondCandidates = findRuleFiles(projectRoot, homeDir, currentFile, undefined, cache);
const reusedRealPath = secondCandidates[0]?.realPath;
// then
expect(cachedRealPath).toBe(canonicalRuleFileA);
expect(reusedRealPath).toBe(canonicalRuleFileA);
});
it("reuses ancestor directory scan for sibling files in the same project", async () => {
// given
const { createRuleScanCache } = await import(`./rule-scan-cache${createImportSuffix()}`);
const { findRuleFiles } = await import(`./rule-file-finder${createImportSuffix()}`);
const siblingDirA = join(projectRoot, "src", "alpha");
const siblingDirB = join(projectRoot, "src", "beta");
const siblingFileA = join(siblingDirA, "a.ts");
const siblingFileB = join(siblingDirB, "b.ts");
mkdirSync(siblingDirA, { recursive: true });
mkdirSync(siblingDirB, { recursive: true });
writeFileSync(siblingFileA, "export const a = 1;\n");
writeFileSync(siblingFileB, "export const b = 2;\n");
mkdirSync(expectedRuleDir, { recursive: true });
writeFileSync(expectedRuleFile, "shared ancestor rules\n");
const cache = createRuleScanCache();
// when
const firstCandidates = findRuleFiles(projectRoot, homeDir, siblingFileA, undefined, cache);
unlinkSync(expectedRuleFile);
rmSync(expectedRuleDir, { recursive: true, force: true });
const siblingCandidates = findRuleFiles(projectRoot, homeDir, siblingFileB, undefined, cache);
// then
expect(firstCandidates.map((candidate) => candidate.path)).toEqual([
expectedRuleFile,
]);
expect(siblingCandidates.map((candidate) => candidate.path)).toEqual([
expectedRuleFile,
]);
});
});
+25 -8
View File
@@ -1,21 +1,38 @@
import type { RuleFileCandidate } from "./types";
export type DirectoryScanEntry = {
path: string;
realPath: string;
};
export type RuleScanCache = {
get: (key: string) => string[] | undefined;
set: (key: string, value: string[]) => void;
get: (key: string) => RuleFileCandidate[] | undefined;
set: (key: string, value: RuleFileCandidate[]) => void;
getDirScan: (dir: string) => DirectoryScanEntry[] | undefined;
setDirScan: (dir: string, entries: DirectoryScanEntry[]) => void;
clear: () => void;
};
export function createRuleScanCache(): RuleScanCache {
const cache = new Map<string, string[]>();
const finalResultCache = new Map<string, RuleFileCandidate[]>();
const directoryScanCache = new Map<string, DirectoryScanEntry[]>();
return {
get(key: string): string[] | undefined {
return cache.get(key);
get(key: string): RuleFileCandidate[] | undefined {
return finalResultCache.get(key);
},
set(key: string, value: string[]): void {
cache.set(key, value);
set(key: string, value: RuleFileCandidate[]): void {
finalResultCache.set(key, value);
},
getDirScan(dir: string): DirectoryScanEntry[] | undefined {
return directoryScanCache.get(dir);
},
setDirScan(dir: string, entries: DirectoryScanEntry[]): void {
directoryScanCache.set(dir, entries);
},
clear(): void {
cache.clear();
finalResultCache.clear();
directoryScanCache.clear();
},
};
}