Allow spaces in cookie values

git-svn-id: https://linkchecker.svn.sourceforge.net/svnroot/linkchecker/trunk/linkchecker@3636 e7d03fd6-7b0d-0410-9947-9c21f3af8025

ChangeLog

@@ -82,6 +82,13 @@
     Changed: config/create.sql
     Closes: SF #1849733
 
+  * Improved cookie parsing:
+    + Allow spaces in attribute values. Example:
+      "Set-Cookie: expires=Wed, 12-Dec-2001 19:27:57 GMT"
+      is now parsed correctly
+    Type: feature
+    Changed: linkcheck/cookies.py
+
 4.7 "300" (released 17.6.2007)
 
   * Mention in the documentation that --anchors enables logging of

linkcheck/cookies.py

@@ -47,6 +47,24 @@ unquote = Cookie._unquote
 quote = Cookie._quote
 has_embedded_dot = re.compile(r"[a-zA-Z0-9]\.[a-zA-Z]").search
 
+
+# Pattern for finding cookie snatched from Pythons Cookie.py
+# Modification: allow whitespace in values.
+LegalChars  = r"\w\d!#%&'~_`><@,:/\$\*\+\-\.\^\|\)\(\?\}\{\="
+CookiePattern = re.compile(r"""
+    (?P<key>                   # Start of group 'key'
+    [%(legalchars)s]+?         # Any word of at least one letter, nongreedy
+    )                          # End of group 'key'
+    \s*=\s*                    # Equal Sign
+    (?P<val>                   # Start of group 'val'
+    "(?:[^\\"]|\\.)*"          # Any doublequoted string
+    |                          # or
+    [%(legalchars)s\s]*        # Any word or empty string
+    )                          # End of group 'val'
+    \s*;?                      # Probably ending in a semi-colon
+    """ % {"legalchars": LegalChars}, re.VERBOSE)
+
+
 class HttpCookie (object):
     """
     A cookie consists of one name-value pair with attributes.
@@ -188,7 +206,7 @@ class HttpCookie (object):
                     raise CookieError("invalid port number: %r" % port)
         self.attributes[key] = value
 
-    def parse (self, text, patt=Cookie._CookiePattern):
+    def parse (self, text, patt=CookiePattern):
         text = strformat.ascii_safe(text)
         # reset values
         self.name = None

linkcheck/tests/test_cookies.py

@@ -86,6 +86,21 @@ class TestCookies (unittest.TestCase):
         path = "/"
         cookie = linkcheck.cookies.NetscapeCookie(value, scheme, host, path)
 
+    def test_netscape_cookie5 (self):
+        data = (
+            ("Foo", "Bar"),
+            ("Expires", "Wed, 12-Dec-2001 19:27:57 GMT"),
+            ("Path", "/"),
+        )
+        # note: values are without quotes
+        parts = ['%s=%s' % (key, value) for key, value in data]
+        value = "; ".join(parts)
+        scheme = "http"
+        host = "imadoofus.org"
+        path = "/"
+        cookie = linkcheck.cookies.NetscapeCookie(value, scheme, host, path)
+        self.assert_(cookie.is_expired())
+
     def test_rfc_cookie1 (self):
         data = (
             ("Foo", "Bar"),