Make backend a required parameter in get_extractor

Make the `backend` argument required in the `get_extractor` internal
routine. Specify a backend in the scripts which call this function. Add
a CLI backend option in capa/features/freeze.py as well.

capa/features/freeze.py

@@ -264,6 +264,14 @@ def main(argv=None):
     parser.add_argument(
         "-f", "--format", choices=[f[0] for f in formats], default="auto", help="Select sample format, %s" % format_help
     )
+    parser.add_argument(
+        "-b",
+        "--backend",
+        type=str,
+        help="select the backend to use in Python 3 (this option is ignored in Python 2)",
+        choices=(capa.main.BACKEND_VIV, capa.main.BACKEND_SMDA),
+        default=capa.main.BACKEND_VIV,
+    )
     args = parser.parse_args(args=argv)
 
     if args.quiet:
@@ -276,7 +284,7 @@ def main(argv=None):
         logging.basicConfig(level=logging.INFO)
         logging.getLogger().setLevel(logging.INFO)
 
-    extractor = capa.main.get_extractor(args.sample, args.format)
+    extractor = capa.main.get_extractor(args.sample, args.format, args.backend)
     with open(args.output, "wb") as f:
         f.write(dump(extractor))
 

capa/main.py

@@ -335,7 +335,7 @@ def get_extractor_py3(path, format, backend, disable_progress=False):
         return capa.features.extractors.viv.VivisectFeatureExtractor(vw, path)
 
 
-def get_extractor(path, format, backend=BACKEND_VIV, disable_progress=False):
+def get_extractor(path, format, backend, disable_progress=False):
     """
     raises:
       UnsupportedFormatError:

scripts/bulk-process.py

@@ -95,7 +95,7 @@ def get_capa_results(args):
     rules, format, path = args
     logger.info("computing capa results for: %s", path)
     try:
-        extractor = capa.main.get_extractor(path, format, disable_progress=True)
+        extractor = capa.main.get_extractor(path, format, capa.main.BACKEND_VIV, disable_progress=True)
     except capa.main.UnsupportedFormatError:
         # i'm 100% sure if multiprocessing will reliably raise exceptions across process boundaries.
         # so instead, return an object with explicit success/failure status.

scripts/capa_as_library.py

@@ -191,7 +191,7 @@ def render_dictionary(doc):
 def capa_details(file_path, output_format="dictionary"):
 
     # extract features and find capabilities
-    extractor = capa.main.get_extractor(file_path, "auto", disable_progress=True)
+    extractor = capa.main.get_extractor(file_path, "auto", capa.main.BACKEND_VIV, disable_progress=True)
     capabilities, counts = capa.main.find_capabilities(rules, extractor, disable_progress=True)
 
     # collect metadata (used only to make rendering more complete)

scripts/lint.py

@@ -201,7 +201,7 @@ class DoesntMatchExample(Lint):
                 continue
 
             try:
-                extractor = capa.main.get_extractor(path, "auto", disable_progress=True)
+                extractor = capa.main.get_extractor(path, "auto", capa.main.BACKEND_VIV, disable_progress=True)
                 capabilities, meta = capa.main.find_capabilities(ctx["rules"], extractor, disable_progress=True)
             except Exception as e:
                 logger.error("failed to extract capabilities: %s %s %s", rule.name, path, e)

scripts/show-capabilities-by-function.py

@@ -199,7 +199,7 @@ def main(argv=None):
         else:
             format = args.format
             try:
-                extractor = capa.main.get_extractor(args.sample, args.format)
+                extractor = capa.main.get_extractor(args.sample, args.format, capa.main.BACKEND_VIV)
             except capa.main.UnsupportedFormatError:
                 logger.error("-" * 80)
                 logger.error(" Input file does not appear to be a PE file.")

scripts/show-features.py

@@ -125,7 +125,7 @@ def main(argv=None):
             extractor = capa.features.freeze.load(f.read())
     else:
         try:
-            extractor = capa.main.get_extractor(args.sample, args.format)
+            extractor = capa.main.get_extractor(args.sample, args.format, capa.main.BACKEND_VIV)
         except capa.main.UnsupportedFormatError:
             logger.error("-" * 80)
             logger.error(" Input file does not appear to be a PE file.")