From 492e5e77d58e9bf999a9c293d03d7c78974dbafc Mon Sep 17 00:00:00 2001
From: willem <itsme@xs4all.nl>
Date: Thu, 29 Jun 2017 15:10:13 +0200
Subject: [PATCH] Add support for decoding CESU-8 encoded strings. This works
 around java's broken utf-8 implementation.

---
 javaobj.py | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/javaobj.py b/javaobj.py
index 51e320d..9fec6f4 100644
--- a/javaobj.py
+++ b/javaobj.py
@@ -47,6 +47,12 @@
     # Python 3+
     from io import BytesIO
 
+try:
+    import ftfy.bad_codecs
+    javacodec = "utf-8-var"
+except ImportError:
+    javacodec = "utf-8"
+
 # ------------------------------------------------------------------------------
 
 # Module version
@@ -641,7 +647,7 @@ def _readString(self, length_fmt="H"):
         """
         (length,) = self._readStruct(">{0}".format(length_fmt))
         ba = self.object_stream.read(length)
-        return to_str(ba)
+        return to_str(ba.decode(javacodec))
 
     def do_classdesc(self, parent=None, ident=0):
         """