Problem with the processing of dot when preceeded by an ambiguous esc…

…ape. Resolves #19

pyformlang/regular_expression/python_regex.py

@@ -53,6 +53,7 @@
 OCTAL = "01234567"
 ESCAPED_OCTAL = ["\\0", "\\1", "\\2", "\\3", "\\4", "\\5", "\\6", "\\7"]
 
+
 class PythonRegex(regex.Regex):
     """ Represents a regular expression as used in Python.
 
@@ -106,7 +107,6 @@ def __init__(self, python_regex):
         self._preprocess_brackets()
         self._preprocess_positive_closure()
         self._preprocess_optional()
-        self._preprocess_dot()
         self._separate()
         self._python_regex = self._python_regex.lstrip('\b')
         super().__init__(self._python_regex)
@@ -119,7 +119,13 @@ def _separate(self):
             else:
                 regex_temp.append(symbol)
         regex_temp = self._recombine(regex_temp)
-        self._python_regex = " ".join(regex_temp)
+        regex_temp_dot = []
+        for symbol in regex_temp:
+            if symbol == ".":
+                regex_temp_dot.append(DOT_REPLACEMENT)
+            else:
+                regex_temp_dot.append(symbol)
+        self._python_regex = " ".join(regex_temp_dot)
 
     def _preprocess_brackets(self):
         regex_temp = []
@@ -287,13 +293,6 @@ def _preprocess_positive_closure(self):
                 regex_temp.append("*")
         self._python_regex = "".join(regex_temp)
 
-    @staticmethod
-    def _dot_replacer(dot):
-        return DOT_REPLACEMENT
-
-    def _preprocess_dot(self):
-        self._python_regex = re.sub(r'(?<!\\)\.', self._dot_replacer, self._python_regex)
-
     def _preprocess_optional(self):
         regex_temp = []
         for symbol in self._python_regex:

pyformlang/regular_expression/tests/test_python_regex.py

@@ -256,6 +256,8 @@ def test_backslash(self):
     def test_octal(self):
         self._test_compare(r"\x10", "\x10")
         self._test_compare(r"\110", "\110")
+        self._test_compare(r"\\\\x10", "\x10")
+        self._test_compare(r"\\\\x10", "\\x10")
 
     def test_backspace(self):
         self._test_compare(r"a[b\b]", "ab")
@@ -272,4 +274,12 @@ def test_unicode_name(self):
 
     def test_unicode(self):
         self._test_compare(r"\u1111", "\u1111")
-        self._test_compare(r"\U00001111", "\U00001111")
+        self._test_compare(r"\U00001111", "\U00001111")
+
+    def test_dot_harder(self):
+        self._test_compare(r"\\.", "\\a")
+        self._test_compare(r"\\.", "\\.")
+        self._test_compare(r"\.", "a")
+        self._test_compare(r"\.", ".")
+        self._test_compare(r"\\\.", "\\a")
+        self._test_compare(r"\\\.", "\\.")

setup.py

@@ -11,7 +11,7 @@
 
 setuptools.setup(
     name='pyformlang',
-    version='1.0.6',
+    version='1.0.7',
     #scripts=['pyformlang'] ,
     author="Julien Romero",
     author_email="[email protected]",