Simplify escape()

The fancy regex isn't really justified. Much faster with replace() too,
though this is an unlikely hotspot.

Could have used a \g<0> backreference to refer to the entire match
instead of using a capturing group too. Hadn't discovered that.

Add some selftests for escape() and unescape() too.

2 files changed, 27 insertions, 6 deletions

diff --git a/kconfiglib.py b/kconfiglib.py
index f1a254d..550c347 100644
--- a/kconfiglib.py
+++ b/kconfiglib.py
@@ -3671,17 +3671,17 @@ def expr_str(expr):
                              _REL_TO_STR[expr[0]],
                              expr_str(expr[2]))
 
-# escape()/unescape() helpers
-_escape_re_sub = re.compile(r'(["\\])').sub
-_unescape_re_sub = re.compile(r"\\(.)").sub
-
 def escape(s):
     r"""
     Escapes the string 's' in the same fashion as is done for display in
     Kconfig format and when writing strings to a .config file. " and \ are
     replaced by \" and \\, respectively.
     """
-    return _escape_re_sub(r"\\\1", s)
+    # \ must be escaped before " to avoid double escaping
+    return s.replace("\\", r"\\").replace('"', r'\"')
+
+# unescape() helper
+_unescape_re_sub = re.compile(r"\\(.)").sub
 
 def unescape(s):
     r"""
diff --git a/testsuite.py b/testsuite.py
index 0d892ed..9d7dda2 100644
--- a/testsuite.py
+++ b/testsuite.py
@@ -41,7 +41,7 @@
 from kconfiglib import Kconfig, Symbol, Choice, COMMENT, MENU, \
                        BOOL, TRISTATE, HEX, STRING, \
                        TRI_TO_STR, \
-                       KconfigSyntaxError, expr_value
+                       KconfigSyntaxError, expr_value, escape, unescape
 import difflib
 import errno
 import os
@@ -241,6 +241,27 @@ def run_selftests():
     verify_string_bad(r""" 'foo """)
 
 
+    print("Testing escape() and unescape()")
+
+    def verify_escape_unescape(s, sesc):
+        # Verify that 's' escapes to 'sesc' and that 'sesc' unescapes to 's'
+        verify_equal(escape(s), sesc)
+        verify_equal(unescape(sesc), s)
+
+    verify_escape_unescape(r''          , r''              )
+    verify_escape_unescape(r'foo'       , r'foo'           )
+    verify_escape_unescape(r'"'         , r'\"'            )
+    verify_escape_unescape(r'""'        , r'\"\"'          )
+    verify_escape_unescape('\\'         , r'\\'            )
+    verify_escape_unescape(r'\\'        , r'\\\\'          )
+    verify_escape_unescape(r'\"'        , r'\\\"'          )
+    verify_escape_unescape(r'"ab\cd"ef"', r'\"ab\\cd\"ef\"')
+
+    # Backslashes before any character should be unescaped, not just before "
+    # and \
+    verify_equal(unescape(r"\afoo\b\c\\d\\\e\\\\f"), r"afoobc\d\e\\f")
+
+
     print("Testing expression evaluation")
 
     c = Kconfig("Kconfiglib/tests/Keval")