Better Py2/3 cross compatibility in modified_utf8

and added bytes_char in utils

Author: tcalmant

javaobj/modifiedutf8.py

@@ -15,6 +15,11 @@
 :status: Alpha
 """
 
+from __future__ import unicode_literals
+
+import sys
+
+
 # Module version
 __version_info__ = (0, 3, 0)
 __version__ = ".".join(str(x) for x in __version_info__)
@@ -27,6 +32,36 @@
 
 # ------------------------------------------------------------------------------
 
+if sys.version_info[0] >= 3:
+    unicode_char = chr
+
+    def byte_to_int(data):
+        # type: (bytes) -> int
+        """
+        Converts the first byte of the given data to an integer
+        """
+        if isinstance(data, int):
+            return data
+        elif isinstance(data, bytes):
+            return data[0]
+
+
+else:
+    unicode_char = unichr  # pylint:disable=undefined-variable
+
+    def byte_to_int(data):
+        # type: (bytes) -> int
+        """
+        Converts the first byte of the given data to an integer
+        """
+        if isinstance(data, int):
+            return data
+        elif isinstance(data, str):
+            return ord(data[0])
+
+
+# ------------------------------------------------------------------------------
+
 
 class DecodeMap(object):
     """
@@ -70,7 +105,11 @@ def apply(self, byte, value, data, i, count):
             value |= byte & self.mask2
         else:
             raise UnicodeDecodeError(
-                NAME, data, i, i + count, "invalid {}-byte sequence".format(self.count)
+                NAME,
+                data,
+                i,
+                i + count,
+                "invalid {}-byte sequence".format(self.count),
             )
         return value
 
@@ -171,8 +210,8 @@ def decode_modified_utf8(data, errors="strict"):
     :return: unicode text and length
     :raises UnicodeDecodeError: sequence is invalid.
     """
-    value, length = u"", 0
-    it = iter(decoder(data))
+    value, length = "", 0
+    it = iter(decoder(byte_to_int(d) for d in data))
     while True:
         try:
             value += next(it)
@@ -185,10 +224,10 @@ def decode_modified_utf8(data, errors="strict"):
             elif errors == "ignore":
                 pass
             elif errors == "replace":
-                value += u"\uFFFD"
+                value += "\uFFFD"
                 length += 1
     return value, length
 
 
 def mutf8_unichr(value):
-    return chr(value)
+    return unicode_char(value)

javaobj/utils.py

@@ -141,6 +141,9 @@ def hexdump(src, start_offset=0, length=16):
     UNICODE_TYPE = str
     unicode_char = chr
 
+    def bytes_char(c):
+        return bytes((c,))
+
     # Python 3 interpreter : bytes & str
     def to_bytes(data, encoding="UTF-8"):
         """
@@ -186,6 +189,7 @@ def read_to_str(data):
 else:
     UNICODE_TYPE = unicode  # pylint:disable=undefined-variable
     unicode_char = unichr  # pylint:disable=undefined-variable
+    bytes_char = chr
 
     # Python 2 interpreter : str & unicode
     def to_str(data, encoding="UTF-8"):