feat: 将 ripgrep 二进制文件通过 base64 编码嵌入到编译产物

Bot · Bot · commit fb1fae35461f · 2026-04-23T19:11:15.000+08:00
compile.ts 构建时编码各平台 ripgrep 二进制文件为 base64，
运行时解码到临时文件（编译模式），开发模式回退 SDK 内置路径。
package.json 添加 compile 脚本。
diff --git a/compile.ts b/compile.ts
@@ -0,0 +1,198 @@
+import { getMacroDefines } from "./scripts/defines.ts";
+import { exit } from "process";
+import { join, resolve } from "path";
+import { readFile, writeFile } from "fs/promises";
+
+const outfile = process.platform === "win32" ? "claude.exe" : "claude";
+
+// Use the currently running bun executable
+const bunExe = process.execPath;
+
+// Collect FEATURE_* env vars from environment
+const features = Object.keys(process.env)
+    .filter(k => k.startsWith("FEATURE_"))
+    .map(k => k.replace("FEATURE_", ""));
+
+// Auto-enable CHICAGO_MCP so @ant packages (computer-use-mcp, etc.)
+// are bundled into the standalone exe. Without this flag, the feature-gated
+// dynamic imports are tree-shaken and the native .node files are not embedded.
+if (!features.includes("CHICAGO_MCP")) {
+    features.push("CHICAGO_MCP");
+}
+
+const defines = getMacroDefines();
+
+// Build --define flags
+const defineArgs = Object.entries(defines).flatMap(([k, v]) => [
+    "--define",
+    `${k}:${v}`,
+]);
+
+// Pass BUNDLED_MODE flag so ripgrepAsset.ts knows we're in compiled mode
+const defineArgsWithBundled = [
+    ...defineArgs,
+    "--define",
+    `BUNDLED_MODE:"true"`,
+];
+
+// Build --feature flags
+const featureArgs = features.flatMap(f => ["--feature", f]);
+
+// ─── Native module embedding ──────────────────────────────────────────────────
+// bun build --compile embeds .node files as assets. When the bundler sees
+// process.env.XXX_NODE_PATH with the var set to an absolute .node path,
+// it rewrites the string to the bunfs asset path. This lets the runtime
+// require() the embedded .node from within the compiled exe.
+//
+// Paths must use forward slashes and be absolute at compile time.
+const repoRoot = resolve(__dirname);
+
+const nativeNodePaths: Record<string, string> = {
+    // @ant packages — macOS only. Path is used at compile time for Bun asset embedding.
+    // Runtime: TS files check process.platform !== "darwin" and skip native load.
+    COMPUTER_USE_INPUT_NODE_PATH: join(repoRoot,
+        "packages/@ant/computer-use-input/prebuilds/arm64-darwin/computer-use-input.node"),
+    COMPUTER_USE_SWIFT_NODE_PATH: join(repoRoot,
+        "packages/@ant/computer-use-swift/prebuilds/arm64-darwin/computer_use.node"),
+
+    // vendor modules — cross-platform (win32/linux/darwin)
+    AUDIO_CAPTURE_NODE_PATH: join(repoRoot,
+        `vendor/audio-capture/${process.arch}-${process.platform}/audio-capture.node`),
+    IMAGE_PROCESSOR_NODE_PATH: join(repoRoot,
+        `vendor/image-processor/${process.arch}-${process.platform}/image-processor.node`),
+    // modifiers and url-handler are macOS only — paths point to darwin builds
+    MODIFIERS_NODE_PATH: join(repoRoot,
+        `vendor/modifiers-napi/${process.arch}-darwin/modifiers.node`),
+    URL_HANDLER_NODE_PATH: join(repoRoot,
+        `vendor/url-handler/${process.arch}-darwin/url-handler.node`),
+};
+
+// Build env with native paths (forward slashes for Bun compatibility)
+const compileEnv: Record<string, string> = {
+    ...process.env,
+    ...Object.fromEntries(
+        Object.entries(nativeNodePaths).map(([k, v]) => [k, v.replace(/\\/g, "/")]),
+    ),
+};
+
+// ─── Step 0: Generate ripgrep base64 asset ───────────────────────────────────
+// Bun's bundler does not support ?url imports or arbitrary file embedding
+// for non-.node files. The only reliable way to embed a binary into the
+// compiled exe is to base64-encode it and store it as a JS string constant.
+// At runtime, we decode to a temp file and execute.
+async function generateRipgrepAsset() {
+    const rgCache = join(repoRoot,
+        `node_modules/.bun/@anthropic-ai+claude-agent-sdk@0.2.87+3c5d820c62823f0b/node_modules/@anthropic-ai/claude-agent-sdk/vendor/ripgrep`);
+
+    const ripgrepBinaries: Record<string, string> = {}
+
+    // Map platform+arch to filename
+    const allPlatforms: Array<{ key: string; subdir: string; file: string }> = [
+        { key: 'windows_x64',    subdir: 'x64-win32',     file: 'rg.exe' },
+        { key: 'darwin_x64',    subdir: 'x64-darwin',    file: 'rg'     },
+        { key: 'darwin_arm64',  subdir: 'arm64-darwin',  file: 'rg'     },
+        { key: 'linux_x64',     subdir: 'x64-linux',     file: 'rg'     },
+        { key: 'linux_arm64',   subdir: 'arm64-linux',   file: 'rg'     },
+    ];
+
+    // Only embed the current platform's binary to minimize exe size.
+    // The other platforms are available in the SDK for dev-mode fallback.
+    const currentPlatformKey = (() => {
+        if (process.platform === 'win32') return 'windows_x64'
+        if (process.platform === 'darwin') return process.arch === 'arm64' ? 'darwin_arm64' : 'darwin_x64'
+        return process.arch === 'arm64' ? 'linux_arm64' : 'linux_x64'
+    })()
+
+    for (const { key, subdir, file } of allPlatforms) {
+        if (key !== currentPlatformKey) continue  // Skip other platforms
+        const binPath = join(rgCache, subdir, file);
+        try {
+            const data = await readFile(binPath);
+            ripgrepBinaries[key] = data.toString('base64');
+            console.log(`Encoded ${key}: ${data.length} bytes -> ${Math.round(data.length * 1.37)} chars`);
+        } catch (e) {
+            console.warn(`Warning: could not read ${binPath}: ${e}`);
+        }
+    }
+
+    // Generate TypeScript asset file
+    const assetFile = join(repoRoot, "src", "utils", "ripgrepAssetBase64.ts");
+    const content = `/**
+ * AUTO-GENERATED by compile.ts — do not edit manually.
+ * Ripgrep binaries encoded as base64 strings.
+ * Decoded at runtime to temp files for execution.
+ */
+export const RIPGREP_BINARIES: Record<string, string> = ${JSON.stringify(ripgrepBinaries, null, 2)};
+`;
+    await writeFile(assetFile, content);
+    console.log(`Generated ${assetFile}`);
+}
+
+// ─── Step 1: Patch SDK ripgrep path ───────────────────────────────────────────
+// The SDK's cli.js computes dy_ from import.meta.url which points to B:\~BUN\root\...
+// in --compile mode. Patch it to use path.dirname(process.execPath) instead.
+async function patchRipgrepPaths() {
+    // --- Patch bun cache SDK cli.js ---
+    const sdkCachePath = join(repoRoot,
+        "node_modules/.bun/@anthropic-ai+claude-agent-sdk@0.2.87+3c5d820c62823f0b/node_modules/@anthropic-ai/claude-agent-sdk/cli.js");
+    const sdkContent = await readFile(sdkCachePath, "utf-8");
+    const patchedSdk = sdkContent
+        .replace(
+            /import\{fileURLToPath as Uy_\}from"url";/,
+            ";",
+        )
+        .replace(
+            /dy_=Uy_\(import\.meta\.url\),dy_=Z16\.join\(dy_,"\.\/"\)/,
+            "dy_=Z16.dirname(process.execPath)",
+        );
+    if (patchedSdk === sdkContent) {
+        console.warn("Warning: SDK patch did not match");
+    } else {
+        await writeFile(sdkCachePath, patchedSdk);
+        console.log("Patched SDK cli.js (bun cache)");
+    }
+}
+
+// ─── Step 2: Run the compile ───────────────────────────────────────────────────
+async function run() {
+    await generateRipgrepAsset();
+    await patchRipgrepPaths();
+
+    console.log("\nCompiling standalone executable with native modules...");
+    console.log(`Outfile: ${outfile}`);
+    console.log(`Defines: ${Object.keys(defines).join(", ")}`);
+    console.log(`Native modules:`);
+    for (const [k, v] of Object.entries(nativeNodePaths)) {
+        console.log(`  ${k}=${v}`);
+    }
+
+    // Use Bun.spawn with CLI because Bun.build({ outfile, compile: true })
+    // does not reliably place the output file on Windows.
+    const result = Bun.spawnSync(
+        [
+            bunExe,
+            "build",
+            "--compile",
+            "--outfile=" + outfile,
+            ...defineArgsWithBundled,
+            ...featureArgs,
+            "src/entrypoints/cli.tsx",
+        ],
+        {
+            stdio: ["inherit", "inherit", "inherit"],
+            env: compileEnv,
+        },
+    );
+
+    if (result.exitCode !== 0) {
+        console.error("Compile failed with exit code:", result.exitCode);
+        exit(1);
+    }
+
+    console.log(`\nCompiled standalone executable: ${outfile}`);
+    if (features.length > 0) {
+        console.log(`Features enabled: ${features.join(", ")}`);
+    }
+}
+
+run();
diff --git a/src/utils/bundledMode.ts b/src/utils/bundledMode.ts
@@ -11,9 +11,14 @@ export function isRunningWithBun(): boolean {
 
 /**
  * Detects if running as a Bun-compiled standalone executable.
- * This checks for embedded files which are present in compiled binaries.
+ *
+ * Primary check: Bun.embeddedFiles (present in compiled binaries).
+ * Fallback: BUNDLED_MODE compile-time constant injected by compile.ts.
  */
+// BUNDLED_MODE is injected at compile time by compile.ts --define flag.
+declare const BUNDLED_MODE: string | undefined
 export function isInBundledMode(): boolean {
+  if (typeof BUNDLED_MODE !== 'undefined') return true
   return (
     typeof Bun !== 'undefined' &&
     Array.isArray(Bun.embeddedFiles) &&
diff --git a/src/utils/ripgrep.ts b/src/utils/ripgrep.ts
@@ -6,6 +6,7 @@ import * as path from 'path'
 import { logEvent } from 'src/services/analytics/index.js'
 import { fileURLToPath } from 'url'
 import { isInBundledMode } from './bundledMode.js'
+import { getRipgrepBinaryPath } from './ripgrepAsset.js'
 import { logForDebugging } from './debug.js'
 import { isEnvDefinedFalsy } from './envUtils.js'
 import { execFileNoThrow } from './execFileNoThrow.js'
@@ -44,15 +45,11 @@ const getRipgrepConfig = memoize((): RipgrepConfig => {
     }
   }
 
-  // In bundled (native) mode, ripgrep is statically compiled into bun-internal
-  // and dispatches based on argv[0]. We spawn ourselves with argv0='rg'.
+  // In bundled mode (compiled exe), ripgrep is embedded via base64.
+  // Extract to temp and execute from there.
   if (isInBundledMode()) {
-    return {
-      mode: 'embedded',
-      command: process.execPath,
-      args: ['--no-config'],
-      argv0: 'rg',
-    }
+    const rgPath = getRipgrepBinaryPath()
+    return { mode: 'builtin', command: rgPath, args: [] }
   }
 
   const rgRoot = path.resolve(__dirname, 'vendor', 'ripgrep')
diff --git a/src/utils/ripgrepAsset.ts b/src/utils/ripgrepAsset.ts
@@ -0,0 +1,122 @@
+/**
+ * Gets the ripgrep binary path for the current platform/arch.
+ *
+ * In compiled mode: decodes base64 from ripgrepAssetBase64.ts, writes to temp,
+ * and caches on disk so subsequent starts skip the decode.
+ *
+ * In dev mode: falls back to SDK's bundled ripgrep path.
+ *
+ * BUNDLED_MODE is a compile-time constant injected by compile.ts --define flag.
+ */
+import { writeFileSync, readFileSync } from 'fs'
+import { mkdirSync } from 'fs'
+import { tmpdir } from 'os'
+import { join } from 'path'
+import { getPlatform } from './platform.js'
+
+// In-memory cache: platform+arch -> absolute path to extracted temp file
+const extractedPaths: Record<string, string> = {}
+
+// Global base64 data — loaded once on first access
+let globalBase64: Record<string, string> | null = null
+
+// SDK's bundled ripgrep path (used as fallback in dev mode)
+function getSdkRipgrepPath(): string {
+  const p = getPlatform()
+  const arch = process.arch
+  if (p === 'windows') return join(process.cwd(), 'node_modules/.bun/@anthropic-ai+claude-agent-sdk@0.2.87+3c5d820c62823f0b/node_modules/@anthropic-ai/claude-agent-sdk/vendor/ripgrep/x64-win32/rg.exe')
+  if (p === 'macos') return join(process.cwd(), 'node_modules/.bun/@anthropic-ai+claude-agent-sdk@0.2.87+3c5d820c62823f0b/node_modules/@anthropic-ai/claude-agent-sdk/vendor/ripgrep', arch === 'arm64' ? 'arm64-darwin/rg' : 'x64-darwin/rg')
+  return join(process.cwd(), 'node_modules/.bun/@anthropic-ai+claude-agent-sdk@0.2.87+3c5d820c62823f0b/node_modules/@anthropic-ai/claude-agent-sdk/vendor/ripgrep', arch === 'arm64' ? 'arm64-linux/rg' : 'x64-linux/rg')
+}
+
+function getPlatformKey(): string {
+  const platform = getPlatform()
+  const arch = process.arch
+  if (platform === 'windows') return 'windows_x64'
+  if (platform === 'macos') return arch === 'arm64' ? 'darwin_arm64' : 'darwin_x64'
+  return arch === 'arm64' ? 'linux_arm64' : 'linux_x64'
+}
+
+// BUNDLED_MODE is injected at compile time by compile.ts --define flag.
+// In dev mode, this variable is undefined.
+declare const BUNDLED_MODE: string | undefined
+
+/**
+ * Load base64 data asynchronously (first call only).
+ * Subsequent calls use the cached global.
+ */
+async function ensureBase64Loaded(): Promise<Record<string, string>> {
+  if (globalBase64 !== null) return globalBase64
+  // Dynamic import so the 6.9MB base64 string isn't loaded in dev mode
+  const mod = await import('./ripgrepAssetBase64.js')
+  globalBase64 = mod.RIPGREP_BINARIES ?? {}
+  return globalBase64
+}
+
+/**
+ * Get the ripgrep binary path for the current platform/arch.
+ * In compiled mode: decodes base64, extracts to temp, caches by version fingerprint.
+ * In dev mode: returns SDK path directly.
+ */
+export function getRipgrepBinaryPath(): string {
+  const key = getPlatformKey()
+  if (extractedPaths[key]) return extractedPaths[key]
+
+  const tmpDir = join(tmpdir(), 'claude-code-ripgrep')
+  const filename = key === 'windows_x64' ? 'rg.exe' : 'rg'
+  const filePath = join(tmpDir, filename)
+  const versionPath = join(tmpDir, `${key}.version`)
+
+  // Dev mode: use SDK path directly
+  if (typeof BUNDLED_MODE === 'undefined') {
+    const sdkPath = getSdkRipgrepPath()
+    extractedPaths[key] = sdkPath
+    return sdkPath
+  }
+
+  // Compiled mode: must use base64 decode (synchronous path — loaded eagerly from embedded module)
+  // In the compiled exe, require() resolves to the embedded ripgrepAssetBase64.js
+  let base64Data: string | undefined
+  try {
+    // eslint-disable-next-line @typescript-eslint/no-var-requires
+    const RIPGREP_BINARIES: Record<string, string> = require('./ripgrepAssetBase64.js').RIPGREP_BINARIES
+    base64Data = RIPGREP_BINARIES[key]
+  } catch {
+    // require failed — fall back to SDK path
+  }
+
+  if (!base64Data) {
+    const sdkPath = getSdkRipgrepPath()
+    extractedPaths[key] = sdkPath
+    return sdkPath
+  }
+
+  const versionTag = `b64:${base64Data.length}:${base64Data.slice(0, 16)}:${base64Data.slice(-16)}`
+
+  // Fast cache check: read only the version tag (~50 bytes)
+  try {
+    const storedTag = readFileSync(versionPath, 'utf8')
+    if (storedTag === versionTag && readFileSync(filePath)) {
+      extractedPaths[key] = filePath
+      return filePath
+    }
+  } catch {
+    // Cache miss or stale
+  }
+
+  // Decode and extract
+  mkdirSync(tmpDir, { recursive: true })
+  const buffer = Buffer.from(base64Data, 'base64')
+  writeFileSync(filePath, buffer, { mode: 0o755 })
+  writeFileSync(versionPath, versionTag, 'utf8')
+  extractedPaths[key] = filePath
+  return filePath
+}
+
+/**
+ * Async version — preloads base64 data before extracting.
+ * Call this early (e.g., during startup) to avoid decode delay on first grep.
+ */
+export async function preloadRipgrepBinary(): Promise<void> {
+  getRipgrepBinaryPath()
+}
diff --git a/src/utils/ripgrepAssetBase64.ts b/src/utils/ripgrepAssetBase64.ts
