basic precedence algorithm - not tested yet

jedi/evaluate/precedence.py

@@ -12,37 +12,41 @@ class PythonGrammar(object):
     Some kind of mirror of http://docs.python.org/3/reference/grammar.html.
     """
 
-    class AtStart(tuple):
-        pass
-
     class MultiPart(str):
         def __new__(cls, first, second):
             self = str.__new__(cls, first)
             self.second = second
 
-    factor = AtStart(('+', '-', '~'))
+        def __str__(self):
-    power = '**',
+            return self.__str__() + ' ' + self.second
-    term = '*', '/', '%', '//'
-    arith_expr = '+', '-'
 
-    shift_expr = '<<', '>>'
+    FACTOR = '+', '-', '~'
-    and_expr = '&',
+    POWER = '**',
-    xor_expr = '^',
+    TERM = '*', '/', '%', '//'
-    expr = '|',
+    ARITH_EXPR = '+', '-'
 
-    comparison = ('<', '>', '==', '>=', '<=', '!=', 'in',
+    SHIFT_EXPR = '<<', '>>'
-                  MultiPart('not', 'in'), 'is', MultiPart('is', 'not'))
+    AND_EXPR = '&',
+    XOR_EXPR = '^',
+    EXPR = '|',
 
-    not_test = AtStart(('not',))
+    COMPARISON = ('<', '>', '==', '>=', '<=', '!=', 'in',
-    and_test = 'and',
+                  MultiPart('not', 'in'),  MultiPart('is', 'not')), 'is'
-    or_test = 'or',
 
-    #test = or_test ['if' or_test 'else' test] | lambdef
+    NOT_TEST = 'not',
+    AND_TEST = 'and',
+    OR_TEST = 'or',
 
-    #sliceop = ':' [test]
+    #TEST = or_test ['if' or_test 'else' test] | lambdef
-    #subscript = test | [test] ':' [test] [sliceop]
+
-    order = (factor, power, term, arith_expr, shift_expr, and_expr, xor_expr,
+    #SLICEOP = ':' [test]
-             expr, comparison, not_test, and_test, or_test)
+    #SUBSCRIPT = test | [test] ':' [test] [sliceop]
+    ORDER = (POWER, TERM, ARITH_EXPR, SHIFT_EXPR, AND_EXPR, XOR_EXPR,
+             EXPR, COMPARISON, AND_TEST, OR_TEST)
+
+    FACTOR_PRIORITY = 0  # highest priority
+    LOWEST_PRIORITY = len(ORDER)
+    NOT_TEST_PRIORITY = LOWEST_PRIORITY - 2  # priority only lower for `and`/`or`
 
 
 class Precedence(object):
@@ -51,58 +55,64 @@ class Precedence(object):
         self.operator = operator
         self.right = right
 
+    def __repr__(self):
+        return '(%s %s %s)' % (self.left, self.operator, self.right)
+
 
 def create_precedence(expression_list):
-    def get_op(current, expression_list):
+    iterator = PushBackIterator(expression_list)
-        for element in expression_list:
+    return _check_operator(iterator)
-            return
+
-            if not isinstance(obj, str):
+
-                debug.warning('normal element in place of operator: %s', element)
+def _syntax_error(element, msg='SyntaxError in precedence'):
+    debug.warning('%s: %s, %s' % (msg, element, element.start_pos))
+
+
+def _get_number(iterator, priority=PythonGrammar.LOWEST_PRIORITY):
+    el = next(iterator)
+    if isinstance(el, pr.Operator):
+        if el in PythonGrammar.FACTOR:
+            right = _get_number(iterator, PythonGrammar.FACTOR_PRIORITY)
+        elif el in PythonGrammar.NOT_TEST and priority <= PythonGrammar.FACTOR_PRIORITY:
+            right = _get_number(iterator, PythonGrammar.NOT_TEST_PRIORITY)
+        else:
+            _syntax_error(el)
+            return _get_number(iterator, priority)
+        return Precedence(None, el, right)
+    else:
+        return el
+
+
+def _check_operator(iterator, priority=PythonGrammar.LOWEST_PRIORITY):
+    """
+    """
+    left = _get_number(iterator, priority)
+    for el in iterator:
+        if not isinstance(el, pr.Operator):
+            _syntax_error(el)
+            continue
+
+        operator = None
+        for check_prio, check in enumerate(PythonGrammar.ORDER):
+            if check_prio >= priority:
+                break  # respect priorities.
+
+            try:
+                match = check.index(el)
+            except ValueError:
                 continue
 
-    def is_operator(obj):
+            if isinstance(match, PythonGrammar.MultiPart):
-        if isinstance(obj, pr.Call):
+                next_tok = next(iterator)
-            element = str(obj.name)
+                if next_tok != match.second:
-        for term in PythonGrammar.order:
+                    iterator.push_back(next_tok)
-            pass
+                continue
-        return element
 
-    def tokenize():
+            operator = check
-        pass
+            break
 
-    new
+        if operator is None:
-    current = None
+            _syntax_error(el)
-    exp_iterator = PushBackIterator(expression_list)
+            continue
-    for element in expression_list:
+        left = Precedence(left, operator, _check_operator(iterator, check_prio))
-        op_name = element
+    return left
-        if isinstance(element, pr.Call):
-            op_name = str(element.name)
-        current = s
-        if isinstance(element, pr.Call):
-            x = str(element.name)
-        new = _process_element(element, expression_list)
-        if current is None:
-            current = new
-        else:
-            current = _process_element(element, expression_list)
-    return current
-
-
-def _process_element(element, expression_list, previous=None):
-    if element is None:
-        return previous
-
-    if isinstance(element, pr.Call):
-        element = str(element.name)
-    elif not isinstance(element, str):
-        if previous is None:
-            previous = element
-        return _process_element(expression_list, previous=previous)
-
-    for term in PythonGrammar.order:
-        if element in term:
-            right = _process_element(expression_list)
-            if right is None:
-                return previous
-            else:
-                return Precedence(previous, element, right)