Issue 6701 - Implement CSP support in python-abp

abp/filters/parser.py

 # This file is part of Adblock Plus <https://adblockplus.org/>,
 # Copyright (C) 2006-present eyeo GmbH
 #
 # Adblock Plus is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License version 3 as
 # published by the Free Software Foundation.
 #
 # Adblock Plus is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
@@ skipping 82 matching lines @@
     XBL = 'xbl'
     DTD = 'dtd'
 
     # Other options.
     MATCH_CASE = 'match-case'
     DOMAIN = 'domain'
     THIRD_PARTY = 'third-party'
     COLLAPSE = 'collapse'
     SITEKEY = 'sitekey'
     DONOTTRACK = 'donottrack'
+    CSP = 'csp'
 
 
 def _line_type(name, field_names, format_string):
     """Define a line type.
 
     Parameters
     ----------
     name: str
         The name of the line type to define.
     field_names: str or list
@@ skipping 25 matching lines @@
 Include = _line_type('Include', 'target', '%include {0.target}%')
 
 
 METADATA_REGEXP = re.compile(r'!\s*(\w+)\s*:\s*(.*)')
 METADATA_KEYS = {'Homepage', 'Title', 'Expires', 'Checksum', 'Redirect',
                  'Version'}
 INCLUDE_REGEXP = re.compile(r'%include\s+(.+)%')
 HEADER_REGEXP = re.compile(r'\[(Adblock(?:\s*Plus\s*[\d\.]+?)?)\]', flags=re.I)
 HIDING_FILTER_REGEXP = re.compile(r'^([^/*|@"!]*?)#([@?])?#(.+)$')
 FILTER_OPTIONS_REGEXP = re.compile(
-    r'\$(~?[\w-]+(?:=[^,\s]+)?(?:,~?[\w-]+(?:=[^,\s]+)?)*)$'
+    r'\$(~?[\w-]+(?:=[^,]+)?(?:,~?[\w-]+(?:=[^,]+)?)*)$'
+)
+
+# Regular expression that matches an invalid Content Security Policy

[rhowell, 2018/06/07 18:16:55]

Removing this, since we aren't currently checking validity.

+INVALID_CSP_REGEXP = re.compile(
+    r'(;|^) ?(base-uri|referrer|report-to|report-uri|upgrade-insecure-requests)\b'
 )
 
 
 def _parse_comment(text):
     match = METADATA_REGEXP.match(text)
     if match and match.group(1) in METADATA_KEYS:
         return Metadata(match.group(1), match.group(2))
     return Comment(text[1:].strip())
 
 
 def _parse_header(text):
     match = HEADER_REGEXP.match(text)
     if not match:
         raise ParseError('Malformed header', text)
     return Header(match.group(1))
 
 
 def _parse_instruction(text):
     match = INCLUDE_REGEXP.match(text)
     if not match:
         raise ParseError('Unrecognized instruction', text)
     return Include(match.group(1))
 
 
 def _parse_option(option):
     if '=' in option:
         return option.split('=', 1)
     if option.startswith('~'):

[Vasily Kuznetsov, 2018/05/30 19:49:26]

If we want to be able to handle the form "~csp=xxxx" this line should be moved
to before line 180. I'm not sure if we do, or what such form means ("~csp" seems
sufficient to negate CSP resource type) but I might be missing something so
better check with Dave, who wrote the code in abpcore (see
https://hg.adblockplus.org/adblockpluscore/rev/f9a48feb592c#l2.84).

[rhowell, 2018/06/07 18:16:55]

I talked to Dave, and he said that ~csp=xxxx and ~csp are both effectively
ignored, and not expected to be used, but the filters are considered valid. Do
you think we should still implement this functionality?

[Vasily Kuznetsov, 2018/06/07 18:28:31]

Let's switch the ifs -- this way both ~csp=xxx and ~csp will produce ('csp',
False) in the return value, which seems like the best way to interpret it. What
to do with this information will be the worry of the user (user of the library
in this case), but at least here we produce the output that makes sense (I think
returning ('~csp', 'xxx') would be error-prone).

[rhowell, 2018/06/07 23:13:06]

Done.

         return option[1:], False
     return option, True
 
 
 def _parse_filter_option(option):
     name, value = _parse_option(option)
 
     # Handle special cases of multivalued options.
     if name == FILTER_OPTION.DOMAIN:
         value = [_parse_option(o) for o in value.split('|')]
@@ skipping 125 matching lines @@
     Raises
     ------
     ParseError
         Thrown during iteration for invalid filter list lines.
     TypeError
         If `lines` is not iterable.
 
     """
     for line in lines:
         yield parse_line(line)