rewrite: add HostScopeCookieRewriter and make it default -- set cookie scope to rewritten host

wombat: ensure rewrite html is using a string
bump to 0.10.5!

Author: ikreymer

README.rst

@@ -1,4 +1,4 @@
-PyWb 0.10.3
+PyWb 0.10.5
 ===========
 
 .. image:: https://travis-ci.org/ikreymer/pywb.svg?branch=master

pywb/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '0.10.3-dev'
+__version__ = '0.10.5'
 
 DEFAULT_CONFIG = 'pywb/default_config.yaml'
 

pywb/rewrite/cookie_rewriter.py

@@ -66,6 +66,40 @@ def rewrite_cookie(self, name, morsel):
         return morsel
 
 
+#=================================================================
+class HostScopeCookieRewriter(WbUrlBaseCookieRewriter):
+    """
+    Attempt to rewrite cookies to current host url..
+
+    If path present, rewrite path to current host. Only makes sense in live
+    proxy or no redirect mode, as otherwise timestamp may change.
+
+    If domain present, remove domain and set to path prefix
+    """
+
+    def rewrite_cookie(self, name, morsel):
+        new_path = None
+
+        # if domain set, expand cookie to host prefix
+        if morsel.get('domain'):
+            del morsel['domain']
+            new_path = self.url_rewriter.rewrite('/')
+
+        # set cookie to rewritten path
+        elif morsel.get('path'):
+            new_path = self.url_rewriter.rewrite(morsel['path'])
+
+        if new_path:
+            inx = new_path.find(self.url_rewriter.rel_prefix)
+            if inx > 0:
+                new_path = new_path[inx:]
+
+            morsel['path'] = new_path
+
+        self._remove_age_opts(morsel)
+        return morsel
+
+
 #=================================================================
 class ExactPathCookieRewriter(WbUrlBaseCookieRewriter):
     """
@@ -112,7 +146,11 @@ def get_cookie_rewriter(cookie_scope):
         return RootScopeCookieRewriter
     elif cookie_scope == 'exact':
         return ExactPathCookieRewriter
+    elif cookie_scope == 'host':
+        return HostScopeCookieRewriter
     elif cookie_scope == 'removeall':
         return RemoveAllCookiesRewriter
-    else:
+    elif cookie_scope == 'coll':
         return MinimalScopeCookieRewriter
+    else:
+        return HostScopeCookieRewriter

pywb/rewrite/test/test_cookie_rewriter.py

@@ -1,27 +1,27 @@
 r"""
-# Default -- MinimalScopeRewriter
+# Default -- MinimalScopeRewriter (Collection scope)
 # No rewriting
 >>> rewrite_cookie('a=b; c=d;')
 [('Set-Cookie', 'a=b'), ('Set-Cookie', 'c=d')]
 
->>> rewrite_cookie('some=value; Path=/;')
+>>> rewrite_cookie('some=value; Path=/;', urlrewriter, 'coll')
 [('Set-Cookie', 'some=value; Path=/pywb/20131226101010/http://example.com/')]
 
->>> rewrite_cookie('some=value; Path=../;', rewriter=urlrewriter2)
+>>> rewrite_cookie('some=value; Path=../;', urlrewriter2, 'coll')
 [('Set-Cookie', 'some=value; Path=/preview/em_/http://example.com/')]
 
->>> rewrite_cookie('some=value; Path=/diff/path/;')
+>>> rewrite_cookie('some=value; Path=/diff/path/;', urlrewriter, 'coll')
 [('Set-Cookie', 'some=value; Path=/pywb/20131226101010/http://example.com/diff/path/')]
 
 # if domain set, set path to root
->>> rewrite_cookie('some=value; Domain=.example.com; Path=/diff/path/; Max-Age=1500')
+>>> rewrite_cookie('some=value; Domain=.example.com; Path=/diff/path/; Max-Age=1500', urlrewriter, 'coll')
 [('Set-Cookie', 'some=value; Path=/pywb/')]
 
->>> rewrite_cookie('abc=def; Path=file.html; Expires=Wed, 13 Jan 2021 22:23:01 GMT')
+>>> rewrite_cookie('abc=def; Path=file.html; Expires=Wed, 13 Jan 2021 22:23:01 GMT', urlrewriter, 'coll')
 [('Set-Cookie', 'abc=def; Path=/pywb/20131226101010/http://example.com/some/path/file.html')]
 
 # Cookie with invalid chars, not parsed
->>> rewrite_cookie('abc@def=123')
+>>> rewrite_cookie('abc@def=123', urlrewriter, 'coll')
 []
 
 
@@ -33,6 +33,14 @@
 [('Set-Cookie', 'some=value')]
 
 
+# HostCookieRewriter -- set path to host
+>>> rewrite_cookie('some=value; Path=/diff/path/;', urlrewriter, 'host')
+[('Set-Cookie', 'some=value; Path=/pywb/20131226101010/http://example.com/diff/path/')]
+
+>>> rewrite_cookie('some=value; Domain=.example.com; Path=/diff/path/; Max-Age=1500', urlrewriter, 'host')
+[('Set-Cookie', 'some=value; Path=/pywb/20131226101010/http://example.com/')]
+
+
 # RootCookieRewriter -- always sets Path=/, removes Domain
 >>> rewrite_cookie('some=value; Path=/diff/path/;', urlrewriter, 'root')
 [('Set-Cookie', 'some=value; Path=/')]