From c1f23c578881b85ae79b524a60160d8f4fb7151b Mon Sep 17 00:00:00 2001
From: Luca Casonato <hello@lcas.dev>
Date: Sat, 8 Jun 2024 18:36:13 +0200
Subject: fix(ext/node): lossy UTF-8 read node_modules files (#24140)

Previously various reads of files in `node_modules` would error on
invalid UTF-8. These were cases involving:

- reading package.json from Rust
- reading package.json from JS
- reading CommonJS files from JS
- reading CommonJS files from Rust (for ESM translation)
- reading ESM files from Rust
---
 cli/node.rs           | 2 +-
 cli/resolver.rs       | 7 ++++++-
 cli/util/gitignore.rs | 2 +-
 3 files changed, 8 insertions(+), 3 deletions(-)

(limited to 'cli')

diff --git a/cli/node.rs b/cli/node.rs
index c696fcac9..5ecbacdc7 100644
--- a/cli/node.rs
+++ b/cli/node.rs
@@ -125,7 +125,7 @@ impl CjsCodeAnalyzer for CliCjsCodeAnalyzer {
       None => {
         self
           .fs
-          .read_text_file_async(specifier.to_file_path().unwrap(), None)
+          .read_text_file_lossy_async(specifier.to_file_path().unwrap(), None)
           .await?
       }
     };
diff --git a/cli/resolver.rs b/cli/resolver.rs
index 301cd0666..3edc6f429 100644
--- a/cli/resolver.rs
+++ b/cli/resolver.rs
@@ -320,7 +320,12 @@ impl NpmModuleLoader {
 
     let code = if self.cjs_resolutions.contains(specifier) {
       // translate cjs to esm if it's cjs and inject node globals
-      let code = String::from_utf8(code)?;
+      let code = match String::from_utf8_lossy(&code) {
+        Cow::Owned(code) => code,
+        // SAFETY: `String::from_utf8_lossy` guarantees that the result is valid
+        // UTF-8 if `Cow::Borrowed` is returned.
+        Cow::Borrowed(_) => unsafe { String::from_utf8_unchecked(code) },
+      };
       ModuleSourceCode::String(
         self
           .node_code_translator
diff --git a/cli/util/gitignore.rs b/cli/util/gitignore.rs
index 12a450d64..4538e0912 100644
--- a/cli/util/gitignore.rs
+++ b/cli/util/gitignore.rs
@@ -105,7 +105,7 @@ impl GitIgnoreTree {
     });
     let current = self
       .fs
-      .read_text_file_sync(&dir_path.join(".gitignore"), None)
+      .read_text_file_lossy_sync(&dir_path.join(".gitignore"), None)
       .ok()
       .and_then(|text| {
         let mut builder = ignore::gitignore::GitignoreBuilder::new(dir_path);
-- 
cgit v1.2.3