fix(app): preserve file line endings and UTF-8 BOM (#2277)

* fix(app): preserve file line endings and UTF-8 BOM

* refactor(app): simplify line ending preservation

Let the editor parse newline variants and serialize with the file's first separator. Mixed-ending files become uniform on edit without a separate normalization layer.

---------

Co-authored-by: Mohamed Boudra <boudra.moha@gmail.com>

packages/app/src/file-explorer/read-result.test.ts

@@ -0,0 +1,34 @@
+import type { FileReadResult } from "@getpaseo/client/internal/daemon-client";
+import { describe, expect, it } from "vitest";
+import { explorerFileFromReadResult } from "./read-result";
+
+function textRead(bytes: Uint8Array): FileReadResult {
+  return {
+    bytes,
+    mime: "text/plain",
+    size: bytes.byteLength,
+    path: "notes.txt",
+    kind: "text",
+    modifiedAt: "2026-07-21T00:00:00.000Z",
+  };
+}
+
+describe("explorerFileFromReadResult", () => {
+  it("records and hides a leading UTF-8 BOM", () => {
+    const file = explorerFileFromReadResult(
+      textRead(new Uint8Array([0xef, 0xbb, 0xbf, 0x68, 0x69])),
+    );
+
+    expect(file).toMatchObject({ content: "hi", hasBom: true });
+  });
+
+  it("does not mark BOM-free text or non-leading U+FEFF as BOM files", () => {
+    const plain = explorerFileFromReadResult(textRead(new TextEncoder().encode("hi")));
+    const embedded = explorerFileFromReadResult(
+      textRead(new Uint8Array([0x68, 0x69, 0xef, 0xbb, 0xbf])),
+    );
+
+    expect(plain.hasBom).toBe(false);
+    expect(embedded.hasBom).toBe(false);
+  });
+});

packages/app/src/file-explorer/read-result.ts

@@ -8,8 +8,13 @@ export function explorerFileFromReadResult(file: FileReadResult): ExplorerFile {
     kind: file.kind,
     encoding: isText ? "utf-8" : "none",
     content: isText ? new TextDecoder().decode(file.bytes) : undefined,
+    hasBom: isText && hasUtf8Bom(file.bytes),
     mimeType: file.mime,
     size: file.size,
     modifiedAt: file.modifiedAt,
   };
 }
+
+function hasUtf8Bom(bytes: Uint8Array): boolean {
+  return bytes.length >= 3 && bytes[0] === 0xef && bytes[1] === 0xbb && bytes[2] === 0xbf;
+}