Python3 support for vivisect

Vivisect has moved to Python3. Allow to run vivisect with Python3 in
capa.

I am using the following version of vivisect (which includes fixes for
some bugs I have found and some open PRs in vivisect):
https://github.com/Ana06/vivisect/tree/py-3

capa/features/extractors/viv/__init__.py

@@ -8,11 +8,7 @@
 
 import types
 
-import file
-import insn
-import function
 import viv_utils
-import basicblock
 
 import capa.features.extractors
 import capa.features.extractors.viv.file
@@ -42,7 +38,7 @@ def add_va_int_cast(o):
     this bit of skullduggery lets use cast viv-utils objects as ints.
     the correct way of doing this is to update viv-utils (or subclass the objects here).
     """
-    setattr(o, "__int__", types.MethodType(get_va, o, type(o)))
+    setattr(o, "__int__", types.MethodType(get_va, o))
     return o
 
 

capa/features/extractors/viv/basicblock.py

@@ -125,11 +125,16 @@ def get_printable_len(oper):
 
 
 def is_printable_ascii(chars):
-    return all(ord(c) < 127 and c in string.printable for c in chars)
+    try:
+        chars_str = chars.decode("ascii")
+    except UnicodeDecodeError:
+        return False
+    else:
+        return all(c in string.printable for c in chars_str)
 
 
 def is_printable_utf16le(chars):
-    if all(c == "\x00" for c in chars[1::2]):
+    if all(c == b"\x00" for c in chars[1::2]):
         return is_printable_ascii(chars[::2])
 
 

capa/main.py

@@ -32,7 +32,7 @@ import capa.features.extractors
 from capa.helpers import oint, get_file_taste
 
 RULES_PATH_DEFAULT_STRING = "(embedded rules)"
-SUPPORTED_FILE_MAGIC = set(["MZ"])
+SUPPORTED_FILE_MAGIC = set([b"MZ"])
 
 
 logger = logging.getLogger("capa")
@@ -304,6 +304,7 @@ class UnsupportedRuntimeError(RuntimeError):
 
 
 def get_extractor_py3(path, format, disable_progress=False):
+    if False:
         from smda.SmdaConfig import SmdaConfig
         from smda.Disassembler import Disassembler
 
@@ -317,6 +318,19 @@ def get_extractor_py3(path, format, disable_progress=False):
             smda_report = smda_disasm.disassembleFile(path)
 
         return capa.features.extractors.smda.SmdaFeatureExtractor(smda_report, path)
+    else:
+        import capa.features.extractors.viv
+
+        with halo.Halo(text="analyzing program", spinner="simpleDots", stream=sys.stderr, enabled=not disable_progress):
+            vw = get_workspace(path, format, should_save=False)
+
+            try:
+                vw.saveWorkspace()
+            except IOError:
+                # see #168 for discussion around how to handle non-writable directories
+                logger.info("source directory is not writable, won't save intermediate workspace")
+
+        return capa.features.extractors.viv.VivisectFeatureExtractor(vw, path)
 
 
 def get_extractor(path, format, disable_progress=False):

setup.py

@@ -27,6 +27,8 @@ if sys.version_info >= (3, 0):
     # py3
     requirements.append("halo")
     requirements.append("networkx")
+    requirements.append("vivisect @ git+https://github.com/Ana06/vivisect@py-3#egg=vivisect")
+    requirements.append("viv-utils==0.3.19")
     requirements.append("smda==1.5.13")
 else:
     # py2