From 44a963aa695409f7b9c9c69f6c2abf79c22e4003 Mon Sep 17 00:00:00 2001
From: Konstantin Konstantinov <konstantin@mach5technology.com>
Date: Fri, 14 Aug 2026 14:14:04 +0300
Subject: [PATCH] fix(scripts): stop the snippet sync walking node_modules and
 cloned fixture repos

findPackageSrcDirs walked packages/ with readdirSync's recursive option and
filtered node_modules out of the result. The filter came too late: recursive
follows symlinks, so the walk entered every packages/*/node_modules link back
into the pnpm store and re-traversed the dependency graph along every path,
collecting each entry into one array until it exhausted the V8 heap. It also
descended into packages/codemod/batch-test/repos and returned src dirs from
the cloned fixture monorepos, whose sources the script would then process.

Descend explicitly and skip node_modules, batch-test and dist before entering
them. pnpm-workspace.yaml already excludes batch-test from the workspace.

Severity depends on the local checkout: batch-test/repos is gitignored, so a
developer who has run the codemod batch tests carries whole external monorepos
under packages/, and there the check aborts out of memory and takes the
pre-push hook down with it.
---
 scripts/sync-snippets.ts | 51 +++++++++++++++++++++++++++++-----------
 1 file changed, 37 insertions(+), 14 deletions(-)

diff --git a/scripts/sync-snippets.ts b/scripts/sync-snippets.ts
index 33258f4414..fd1ffc6932 100644
--- a/scripts/sync-snippets.ts
+++ b/scripts/sync-snippets.ts
@@ -504,31 +504,54 @@ function findMarkdownFiles(dir: string): string[] {
   return files;
 }
 
+/**
+ * Directory names that are never descended into when looking for package sources.
+ *
+ * `node_modules` is the critical one: pnpm fills it with symlinks pointing back
+ * into the store, whose packages link onward in turn. Entering it means walking
+ * that graph along every distinct link path, re-visiting the same directories
+ * over and over.
+ *
+ * `batch-test` holds the codemod's cloned fixture repositories — whole external
+ * monorepos that `pnpm-workspace.yaml` already excludes from the workspace, and
+ * whose sources this script must never rewrite.
+ */
+const SKIPPED_DIRS = new Set(['node_modules', 'dist', 'batch-test']);
+
 /**
  * Find all package src directories under the packages directory.
+ *
+ * Descends explicitly rather than using `readdirSync`'s `recursive` option. That
+ * option follows symlinks and collects every entry it visits into a single array
+ * before returning, so filtering unwanted directories out of the result is too
+ * late to keep the walk bounded — the traversal has already happened.
+ *
  * @param packagesDir The packages directory
  * @returns Array of absolute paths to src directories
  */
 function findPackageSrcDirs(packagesDir: string): string[] {
   const srcDirs: string[] = [];
-  const entries = readdirSync(packagesDir, {
-    withFileTypes: true,
-    recursive: true,
-  });
 
-  for (const entry of entries) {
-    if (!entry.isDirectory()) continue;
-    if (entry.name !== 'src') continue;
+  const descend = (dir: string): void => {
+    for (const entry of readdirSync(dir, { withFileTypes: true })) {
+      // isDirectory() is false for a symlink, so pnpm's links are never entered.
+      if (!entry.isDirectory()) continue;
+      if (SKIPPED_DIRS.has(entry.name)) continue;
 
-    const fullPath = join(entry.parentPath, entry.name);
+      const fullPath = join(dir, entry.name);
 
-    // Only include src dirs that are direct children of a package
-    // (e.g., packages/core-internal/src, packages/middleware/express/src)
-    // Skip nested src dirs like node_modules/*/src
-    if (fullPath.includes('node_modules')) continue;
+      // A package owns a single src dir; everything below it is that package's
+      // own tree, which findSourceFiles walks.
+      if (entry.name === 'src') {
+        srcDirs.push(fullPath);
+        continue;
+      }
 
-    srcDirs.push(fullPath);
-  }
+      descend(fullPath);
+    }
+  };
+
+  descend(packagesDir);
 
   return srcDirs;
 }