Embed tree-sitter.wasm in CLI binary to fix bun --compile resolution

web-tree-sitter@0.25.10 split its package exports into separate `import`
and `require` conditions. In bun --compile binaries, require.resolve from
init-node.ts now returns the build-time absolute path of tree-sitter.cjs,
which doesn't exist on user machines, causing freebuff to crash on
startup with "Cannot find module .../tree-sitter.cjs".

Fix by embedding tree-sitter.wasm into the binary via Bun's
\`import ... with { type: 'file' }\` and stashing the resulting bunfs path
in process.env so all copies of init-node.ts (including the SDK pre-built
bundle's inlined copy) can pick it up.

Author: jahooma

cli/package.json

@@ -23,7 +23,9 @@
     "test:tmux-poc": "bun run src/__tests__/tmux-poc.ts",
     "typecheck": "tsc --noEmit -p ."
   },
-  "sideEffects": false,
+  "sideEffects": [
+    "./src/pre-init/*.ts"
+  ],
   "engines": {
     "bun": "1.3.11"
   },

cli/src/index.tsx

@@ -1,5 +1,12 @@
 #!/usr/bin/env bun
 
+// Embed tree-sitter.wasm into the bun-compile binary at a bunfs path the runtime
+// can find. Without this, web-tree-sitter resolves the wasm via require.resolve,
+// which (since 0.25.10's split exports map) returns the build-time absolute path
+// of tree-sitter.cjs and fails on user machines. Must run before the SDK / code-map
+// import chain triggers Parser.init.
+import './pre-init/tree-sitter-wasm'
+
 import fs from 'fs'
 import { createRequire } from 'module'
 import os from 'os'

cli/src/pre-init/tree-sitter-wasm.ts

@@ -0,0 +1,16 @@
+// Embed tree-sitter.wasm into the bun-compile binary at a bunfs path the runtime
+// can find. Must be the very first import in `index.tsx`: subsequent imports
+// (the SDK / code-map) eagerly construct a tree-sitter parser singleton, and its
+// `locateFile` callback reads `CODEBUFF_TREE_SITTER_WASM_PATH` from `process.env`.
+//
+// Without this, web-tree-sitter@0.25.10 falls back to `require.resolve` which —
+// per the package's split `import`/`require` exports map — returns the build-time
+// absolute path of `tree-sitter.cjs` and fails on user machines.
+
+import treeSitterWasmPath from 'web-tree-sitter/tree-sitter.wasm' with {
+  type: 'file',
+}
+
+if (treeSitterWasmPath) {
+  process.env.CODEBUFF_TREE_SITTER_WASM_PATH = treeSitterWasmPath
+}

packages/code-map/src/index.ts

@@ -1,3 +1,4 @@
 import './types'
 export * from './parse'
 export * from './languages'
+export { setTreeSitterWasmPath } from './init-node'

packages/code-map/src/init-node.ts

@@ -3,6 +3,53 @@ import * as path from 'path'
 
 import { Parser } from 'web-tree-sitter'
 
+const TREE_SITTER_WASM_ENV_VAR = 'CODEBUFF_TREE_SITTER_WASM_PATH'
+
+/**
+ * Override the path to `tree-sitter.wasm` used during {@link initTreeSitterForNode}.
+ *
+ * Needed for `bun build --compile` binaries: the embedded `tree-sitter.js` reports a
+ * `scriptDir` like `/$bunfs/root/`, but the runtime wasm isn't auto-embedded next to
+ * it, and `require.resolve('web-tree-sitter')` resolves to the build-time absolute
+ * path of `tree-sitter.cjs` (per the package's `require` exports condition added in
+ * 0.25.10), which doesn't exist on the end user's machine. Callers building binaries
+ * should embed the wasm via Bun's `import ... with { type: 'file' }` and pass the
+ * resulting path here before any tree-sitter use.
+ *
+ * Stored on `process.env` so it reaches every copy of this module — the SDK
+ * pre-built bundle inlines its own copy of `init-node.ts`, so a module-level
+ * variable here wouldn't be visible to the singleton initialized via the SDK.
+ */
+export function setTreeSitterWasmPath(wasmPath: string): void {
+  process.env[TREE_SITTER_WASM_ENV_VAR] = wasmPath
+}
+
+function resolveTreeSitterWasm(scriptDir: string): string {
+  const override = process.env[TREE_SITTER_WASM_ENV_VAR]
+  if (override && fs.existsSync(override)) {
+    return override
+  }
+
+  const fallback = path.join(scriptDir, 'tree-sitter.wasm')
+  if (fs.existsSync(fallback)) {
+    return fallback
+  }
+
+  try {
+    const pkgDir = path.dirname(require.resolve('web-tree-sitter'))
+    const wasm = path.join(pkgDir, 'tree-sitter.wasm')
+    if (fs.existsSync(wasm)) {
+      return wasm
+    }
+  } catch {
+    // Package not resolvable; fall through.
+  }
+
+  throw new Error(
+    `Internal error: tree-sitter.wasm not found (looked at scriptDir=${scriptDir} and via web-tree-sitter package). Set ${TREE_SITTER_WASM_ENV_VAR} or ensure the file is included in your deployment bundle.`,
+  )
+}
+
 /**
  * Initialize web-tree-sitter for Node.js environments with proper WASM file location
  */
@@ -11,22 +58,7 @@ export async function initTreeSitterForNode(): Promise<void> {
   await Parser.init({
     locateFile: (name: string, scriptDir: string) => {
       if (name === 'tree-sitter.wasm') {
-        // Fallback to script directory
-        const fallback = path.join(scriptDir, name)
-        if (fs.existsSync(fallback)) {
-          return fallback
-        }
-
-        // Find the installed package root
-        const pkgDir = path.dirname(require.resolve('web-tree-sitter'))
-        // The wasm ships at: node_modules/web-tree-sitter/tree-sitter.wasm
-        const wasm = path.join(pkgDir, 'tree-sitter.wasm')
-        if (fs.existsSync(wasm)) {
-          return wasm
-        }
-        throw new Error(
-          `Internal error: web-tree-sitter/tree-sitter.wasm not found at ${wasm}. Ensure the file is included in your deployment bundle.`,
-        )
+        return resolveTreeSitterWasm(scriptDir)
       }
 
       // For other files, use default behavior

sdk/src/index.ts

@@ -82,7 +82,11 @@ export {
 export type { CodebuffFileSystem } from '@codebuff/common/types/filesystem'
 
 // Tree-sitter / code-map exports
-export { getFileTokenScores, setWasmDir } from '@codebuff/code-map'
+export {
+  getFileTokenScores,
+  setWasmDir,
+  setTreeSitterWasmPath,
+} from '@codebuff/code-map'
 export type { FileTokenData, TokenCallerMap } from '@codebuff/code-map'
 
 export { runTerminalCommand } from './tools/run-terminal-command'