Drop UTF-8 BOM when present while decoding UTF-8 bytes into String (#1165)

Author: jmschonfeld

Sources/FoundationEssentials/String/String+IO.swift

@@ -101,6 +101,10 @@ extension String {
             }
         case .utf8:
             func makeString(buffer: UnsafeBufferPointer<UInt8>) -> String? {
+                var buffer = buffer
+                if buffer.starts(with: [0xEF, 0xBB, 0xBF]) {
+                    buffer = UnsafeBufferPointer(rebasing: buffer.suffix(from: 3))
+                }
                 if let string = String._tryFromUTF8(buffer) {
                     return string
                 }

Tests/FoundationEssentialsTests/StringTests.swift

@@ -1004,6 +1004,13 @@ final class StringTests : XCTestCase {
 
         let utf32BEBOMStringMismatch = String(bytes: utf32BEWithBOM, encoding: String._Encoding.utf32LittleEndian)
         XCTAssertNil(utf32BEBOMStringMismatch)
+        
+        // UTF-8 With BOM
+        
+        let utf8BOM = Data([0xEF, 0xBB, 0xBF])
+        let helloWorld = Data("Hello, world".utf8)
+        XCTAssertEqual(String(bytes: utf8BOM + helloWorld, encoding: String._Encoding.utf8), "Hello, world")
+        XCTAssertEqual(String(bytes: helloWorld + utf8BOM, encoding: String._Encoding.utf8), "Hello, world\u{FEFF}")
     }
 
     func test_dataUsingEncoding_preservingBOM() {