Add in protections against call to eval(expression)

numexpr/necompiler.py

@@ -13,6 +13,7 @@
 import sys
 import numpy
 import threading
+import re
 
 is_cpu_amd_intel = False # DEPRECATION WARNING: WILL BE REMOVED IN FUTURE RELEASE
 from numexpr import interpreter, expressions, use_vml
@@ -259,10 +260,17 @@ def __init__(self, astnode):
     def __str__(self):
         return 'Immediate(%d)' % (self.node.value,)
 
-
+_forbidden_re = re.compile('[\;[\:]|__')
 def stringToExpression(s, types, context):
     """Given a string, convert it to a tree of ExpressionNode's.
     """
+    # sanitize the string for obvious attack vectors that NumExpr cannot 
+    # parse into its homebrew AST. This is to protect the call to `eval` below.
+    # We forbid `;`, `:`. `[` and `__`
+    # We would like to forbid `.` but it is both a reference and decimal point.
+    if _forbidden_re.search(s) is not None:
+        raise ValueError(f'Expression {s} has forbidden control characters.')
+
     old_ctx = expressions._context.get_current_context()
     try:
         expressions._context.set_new_context(context)
@@ -285,8 +293,10 @@ def stringToExpression(s, types, context):
                 t = types.get(name, default_type)
                 names[name] = expressions.VariableNode(name, type_to_kind[t])
         names.update(expressions.functions)
+
         # now build the expression
         ex = eval(c, names)
+
         if expressions.isConstant(ex):
             ex = expressions.ConstantNode(ex, expressions.getKind(ex))
         elif not isinstance(ex, expressions.ExpressionNode):
@@ -611,9 +621,7 @@ def NumExpr(ex, signature=(), **kwargs):
 
     Returns a `NumExpr` object containing the compiled function.
     """
-    # NumExpr can be called either directly by the end-user, in which case
-    # kwargs need to be sanitized by getContext, or by evaluate,
-    # in which case kwargs are in already sanitized.
+
     # In that case _frame_depth is wrong (it should be 2) but it doesn't matter
     # since it will not be used (because truediv='auto' has already been
     # translated to either True or False).
@@ -758,7 +766,7 @@ def getArguments(names, local_dict=None, global_dict=None, _frame_depth: int=2):
 _names_cache = CacheDict(256)
 _numexpr_cache = CacheDict(256)
 _numexpr_last = {}
-
+_numexpr_sanity = set()
 evaluate_lock = threading.Lock()
 
 # MAYBE: decorate this function to add attributes instead of having the 
@@ -861,7 +869,7 @@ def evaluate(ex: str,
              out: numpy.ndarray = None, 
              order: str = 'K', 
              casting: str = 'safe', 
-             _frame_depth: int=3,
+             _frame_depth: int = 3,
              **kwargs) -> numpy.ndarray:
     """
     Evaluate a simple array expression element-wise using the virtual machine.
@@ -909,6 +917,8 @@ def evaluate(ex: str,
     _frame_depth: int
         The calling frame depth. Unless you are a NumExpr developer you should 
         not set this value.
+
+    
     """
     # We could avoid code duplication if we called validate and then re_evaluate 
     # here, but they we have difficulties with the `sys.getframe(2)` call in
@@ -921,10 +931,6 @@ def evaluate(ex: str,
     else:
         raise e
 
-
-
-
-
 def re_evaluate(local_dict: Optional[Dict] = None, 
                 _frame_depth: int=2) -> numpy.ndarray:
     """

numexpr/tests/test_numexpr.py

@@ -373,8 +373,9 @@ def test_re_evaluate_dict(self):
         a1 = array([1., 2., 3.])
         b1 = array([4., 5., 6.])
         c1 = array([7., 8., 9.])
-        x = evaluate("2*a + 3*b*c", local_dict={'a': a1, 'b': b1, 'c': c1})
-        x = re_evaluate()
+        local_dict={'a': a1, 'b': b1, 'c': c1}
+        x = evaluate("2*a + 3*b*c", local_dict=local_dict)
+        x = re_evaluate(local_dict=local_dict)
         assert_array_equal(x, array([86., 124., 168.]))
 
     def test_validate(self):
@@ -400,9 +401,10 @@ def test_validate_dict(self):
         a1 = array([1., 2., 3.])
         b1 = array([4., 5., 6.])
         c1 = array([7., 8., 9.])
-        retval = validate("2*a + 3*b*c", local_dict={'a': a1, 'b': b1, 'c': c1})
+        local_dict={'a': a1, 'b': b1, 'c': c1}
+        retval = validate("2*a + 3*b*c", local_dict=local_dict)
         assert(retval is None)
-        x = re_evaluate()
+        x = re_evaluate(local_dict=local_dict)
         assert_array_equal(x, array([86., 124., 168.]))
 
     # Test for issue #22
@@ -502,11 +504,49 @@ def test_illegal_value(self):
         a = arange(3)
         try:
             evaluate("a < [0, 0, 0]")
-        except TypeError:
+        except (ValueError, TypeError):
+            pass
+        else:
+            self.fail()
+
+    def test_forbidden_tokens(self):
+        # Forbid dunder
+        try:
+            evaluate('__builtins__')
+        except ValueError:
+            pass
+        else:
+            self.fail()
+
+        # Forbid colon for lambda funcs
+        try: 
+            evaluate('lambda x: x')
+        except ValueError:
+            pass
+        else:
+            self.fail()
+
+        # Forbid indexing
+        try:
+            evaluate('locals()[]')
+        except ValueError:
             pass
         else:
             self.fail()
 
+        # Forbid semicolon
+        try:
+            evaluate('import os; os.cpu_count()')
+        except ValueError:
+            pass
+        else:
+            self.fail()
+
+        # I struggle to come up with cases for our ban on `'` and `"`
+
+
+
+
     def test_disassemble(self):
         assert_equal(disassemble(NumExpr(
             "where(m, a, -1)", [('m', bool), ('a', float)])),