Python 3 decompilation from Python2

test/Makefile

@@ -9,30 +9,38 @@ NATIVE_CHECK = check-$(PYTHON_VERSION)
 # Set COMPILE='--compile' to force compilation before check
 COMPILE ?=
 
-#: Run working tests from Python 2.7
+# Run short tests
-check-2.7: check-short-2.7 check-bytecode-2.5 check-2.7-ok
+check-short:
+	@$(PYTHON) -V && PYTHON_VERSION=`$(PYTHON) -V 2>&1 | cut -d ' ' -f 2 | cut -d'.' -f1,2`; \
+	$(MAKE) check-bytecode
 
-#: Run working tests from Python 3.4
+# Run all tests
-check-3.4: check-short-3.4 check-short-2.7 check-bytecode-2.5 check-bytecode-3.2
+check:
-
-check: check-short
 	@$(PYTHON) -V && PYTHON_VERSION=`$(PYTHON) -V 2>&1 | cut -d ' ' -f 2 | cut -d'.' -f1,2`; \
 	$(MAKE) check-$$PYTHON_VERSION
 
-check-short:
+#: Run working tests from Python 2.7
-	@$(PYTHON) -V && PYTHON_VERSION=`$(PYTHON) -V 2>&1 | cut -d ' ' -f 2 | cut -d'.' -f1,2`; \
+check-2.7: check-bytecode check-2.7-ok
-	$(MAKE) check-short-$$PYTHON_VERSION
+
+#: Run working tests from Python 3.4
+check-3.4: check-bytecode
 
 #: Check deparsing only, but from a different Python version
 check-disasm:
 	$(PYTHON) dis-compare.py
 
-#: Check deparsing only, from Python 2.5
+#: Check deparsing bytecode only
+check-bytecode: check-bytecode-2.5 check-bytecode-2.5 check-bytecode-3.2
+
+#: Check deparsing Python 2.5
 check-bytecode-2.5:
 	$(PYTHON) test_pythonlib.py --bytecode-2.5
 
+#: Check deparsing Python 2.7
+check-bytecode-2.7:
+	$(PYTHON) test_pythonlib.py --bytecode-2.7
 
-#: Check deparsing only, from Python 3.2
+#: Check deparsing Python 3.2
 check-bytecode-3.2:
 	$(PYTHON) test_pythonlib.py --bytecode-3.2
 
@@ -40,14 +48,6 @@ check-bytecode-3.2:
 check-native-short:
 	$(PYTHON) test_pythonlib.py --bytecode-$(PYTHON_VERSION) --verify $(COMPILE)
 
-#: Short, quickly-running Python 2.7 programs. Useful in debugging grammar problems
-check-short-2.7:
-	$(PYTHON) test_pythonlib.py --bytecode-2.7 --verify $(COMPILE)
-
-#: Short, quickly-running Python 2.7 programs. Useful in debugging grammar problems
-check-short-3.4:
-	$(PYTHON) test_pythonlib.py --bytecode-3.4 --verify $(COMPILE)
-
 #: Run longer Python 2.7's lib files known to be okay
 check-2.7-ok:
 	$(PYTHON) test_pythonlib.py --ok-2.7 --verify $(COMPILE)

test/simple-source/misc/assign_none.py

@@ -0,0 +1 @@
+vario = None

test/simple-source/misc/assign_none_str.py

@@ -0,0 +1 @@
+vario = 'None'

uncompyle6/marsh.py

@@ -92,6 +92,14 @@ def load_code_internal(fp, magic_int, bytes_for_s=False):
                             co_firstlineno, bytes(co_lnotab, encoding='utf-8'),
                             co_freevars, co_cellvars)
         else:
+            if (3000 < magic_int < 20121):
+                # Python 3  encodes some fields as Unicode while Python2
+                # requires the corresponding field to have string values
+                co_consts = tuple([str(s) if s else None for s in co_consts])
+                co_names  = tuple([str(s) if s else None for s in co_names])
+                co_varnames  = tuple([str(s) if s else None for s in co_varnames])
+                co_filename = str(co_filename)
+                co_name = str(co_name)
             return Code(co_argcount, co_nlocals, co_stacksize, co_flags, co_code,
                         co_consts, co_names, co_varnames, co_filename, co_name,
                         co_firstlineno, co_lnotab, co_freevars, co_cellvars)

uncompyle6/scanners/scanner32.py

@@ -14,6 +14,7 @@ from __future__ import print_function
 
 import dis, marshal
 from collections import namedtuple
+from array import array
 
 from uncompyle6.scanner import Token, L65536
 
@@ -42,7 +43,7 @@ class Scanner32(scan.Scanner):
         # Container for tokens
         tokens = []
         customize = {}
-        self.code = code = co.co_code
+        self.code = code = array('B', co.co_code)
         codelen = len(code)
         self.build_lines_data(co)
         self.build_prev_op()
@@ -97,7 +98,6 @@ class Scanner32(scan.Scanner):
                         free = co.co_cellvars + co.co_freevars
                     current_token.pattr = free[oparg]
             tokens.append(current_token)
-
         return tokens, customize
 
     def build_lines_data(self, code_obj):

uncompyle6/scanners/scanner34.py

@@ -163,11 +163,12 @@ class Scanner34(scan.Scanner):
     def disassemble_cross_version(self, co):
         """
         Convert code object <co> into a sequence of tokens
-        FIXME: the below code faulty in many was is probably based on older Python2's dis.disassemble() function.
+        FIXME: the below is not based on the current Python 3.4 dis.disassemble_bytes().
-        It needs to be rewritten and moduled off of Python 3.4's dis.disassemble_bytes().
+        Fix that.
         """
         # Container for tokens
         tokens = []
+        customize = {}
         self.code = code = array('B', co.co_code)
         codelen = len(code)
         self.build_lines_data(co)
@@ -223,7 +224,7 @@ class Scanner34(scan.Scanner):
                         free = co.co_cellvars + co.co_freevars
                     current_token.pattr = free[oparg]
             tokens.append(current_token)
-        return tokens, {}
+        return tokens, customize
 
     def build_lines_data(self, code_obj):
         """
@@ -620,7 +621,6 @@ class Scanner34(scan.Scanner):
         return filtered
 
 if __name__ == "__main__":
-    import inspect
     co = inspect.currentframe().f_code
     tokens, customize = Scanner34().disassemble(co)
     for t in tokens: