Copy potential BOM to the output of PyString_FromString

filmor · filmor · commit dc6f5efb7905 · 2024-05-05T20:42:33.000+02:00
The documentation of the used `PyUnicode_DecodeUTF16` states that not
passing `*byteorder` or passing a 0 results in the first two bytes, if
they are the BOM (U+FEFF, zero-width no-break space), to be interpreted
and skipped, which is incorrect when we convert a known "non BOM"
string, which all strings from C# are.
diff --git a/src/runtime/Runtime.cs b/src/runtime/Runtime.cs
@@ -1252,12 +1252,14 @@ internal static bool PyString_CheckExact(BorrowedReference ob)
 
         internal static NewReference PyString_FromString(string value)
         {
+            int byteorder = BitConverter.IsLittleEndian ? -1 : 1;
+            int* byteorderPtr = &byteorder;
             fixed(char* ptr = value)
                 return Delegates.PyUnicode_DecodeUTF16(
                     (IntPtr)ptr,
                     value.Length * sizeof(Char),
                     IntPtr.Zero,
-                    IntPtr.Zero
+                    (IntPtr)byteorderPtr
                 );
         }
 

Original file line number	Diff line number	Diff line change
`@@ -1252,12 +1252,14 @@ internal static bool PyString_CheckExact(BorrowedReference ob)`
`1252`	`1252`
`1253`	`1253`	`internal static NewReference PyString_FromString(string value)`
`1254`	`1254`	`{`
	`1255`	`+ int byteorder = BitConverter.IsLittleEndian ? -1 : 1;`
	`1256`	`+ int* byteorderPtr = &byteorder;`
`1255`	`1257`	`fixed(char* ptr = value)`
`1256`	`1258`	`return Delegates.PyUnicode_DecodeUTF16(`
`1257`	`1259`	`(IntPtr)ptr,`
`1258`	`1260`	`value.Length * sizeof(Char),`
`1259`	`1261`	`IntPtr.Zero,`
`1260`		`- IntPtr.Zero`
	`1262`	`+ (IntPtr)byteorderPtr`
`1261`	`1263`	`);`
`1262`	`1264`	`}`
`1263`	`1265`