From b2f95339ce3f42f8724fdf524004b41abcfa793a Mon Sep 17 00:00:00 2001
From: Jimmy <jimmy@spalge.com>
Date: Wed, 27 Dec 2017 22:04:47 +1300
Subject: [PATCH] Greasemonkey: support regexes in @include and @exclude.

Like the spec says, if a value for the @include or @exclude rules starts
and ends with a '/' it should be parsed as a regular expression.
Technically a ECMAScript syntax regular expression, but I am not sure of
the differences and I assume they are far fewer than the similarities.
One that I did see mentioned was that javascript RegExp doesn't support
unicode. Although it apparently does support a 'u' flag now.

Note that code will only be ran for QtWebkit and QWebEngine < 5.8
we rely on the builtin support for metadata it QWebEngine for most
things greasemonkey related. Sadly it seems that they missed the regex
requirement too. I've opened a ticket to track that https://bugreports.qt.io/browse/QTBUG-65484
---
 qutebrowser/browser/greasemonkey.py        | 19 +++++++++++----
 tests/unit/javascript/test_greasemonkey.py | 28 +++++++++++++++++++++-
 2 files changed, 41 insertions(+), 6 deletions(-)

diff --git a/qutebrowser/browser/greasemonkey.py b/qutebrowser/browser/greasemonkey.py
index 9a82d6a93..579cca82b 100644
--- a/qutebrowser/browser/greasemonkey.py
+++ b/qutebrowser/browser/greasemonkey.py
@@ -23,7 +23,6 @@ import re
 import os
 import json
 import fnmatch
-import functools
 import glob
 
 import attr
@@ -196,11 +195,21 @@ class GreasemonkeyManager(QObject):
         """
         if url.scheme() not in self.greaseable_schemes:
             return MatchingScripts(url, [], [], [])
-        match = functools.partial(fnmatch.fnmatch,
-                                  url.toString(QUrl.FullyEncoded))
+
+        def _match(pattern):
+            # For include and exclude rules if they start and end with '/' they
+            # should be treated as a (ecma syntax) regular expression.
+            string_url = url.toString(QUrl.FullyEncoded)
+            if pattern.startswith('/') and pattern.endswith('/'):
+                return re.search(pattern[1:-1], string_url) is not None
+
+            # Otherwise they are glob expressions.
+            return fnmatch.fnmatch(string_url, pattern)
+
         tester = (lambda script:
-                  any(match(pat) for pat in script.includes) and
-                  not any(match(pat) for pat in script.excludes))
+                  any(_match(pat) for pat in script.includes) and
+                  not any(_match(pat) for pat in script.excludes))
+
         return MatchingScripts(
             url,
             [script for script in self._run_start if tester(script)],
diff --git a/tests/unit/javascript/test_greasemonkey.py b/tests/unit/javascript/test_greasemonkey.py
index 0f5fe476c..1aaa3380f 100644
--- a/tests/unit/javascript/test_greasemonkey.py
+++ b/tests/unit/javascript/test_greasemonkey.py
@@ -19,6 +19,7 @@
 """Tests for qutebrowser.browser.greasemonkey."""
 
 import logging
+import textwrap
 
 import pytest
 import py.path  # pylint: disable=no-name-in-module
@@ -26,7 +27,7 @@ from PyQt5.QtCore import QUrl
 
 from qutebrowser.browser import greasemonkey
 
-test_gm_script = """
+test_gm_script = r"""
 // ==UserScript==
 // @name qutebrowser test userscript
 // @namespace invalid.org
@@ -75,6 +76,31 @@ def test_get_scripts_by_url(url, expected_matches):
             expected_matches)
 
 
+@pytest.mark.parametrize("url, expected_matches", [
+    # included
+    ('https://github.com/qutebrowser/qutebrowser/', 1),
+    # neither included nor excluded
+    ('http://aaaaaaaaaa.com/', 0),
+    # excluded takes priority
+    ('http://github.com/foo', 0),
+])
+def test_regex_includes_scripts_for(url, expected_matches):
+    """Ensure our GM @*clude support supports regular expressions."""
+    gh_dark_example = textwrap.dedent(r"""
+        // ==UserScript==
+        // @include     /^https?://((gist|guides|help|raw|status|developer)\.)?github\.com/((?!generated_pages\/preview).)*$/
+        // @exclude     /https?://github\.com/foo/
+        // @run-at document-start
+        // ==/UserScript==
+    """)
+    _save_script(gh_dark_example, 'test.user.js')
+    gm_manager = greasemonkey.GreasemonkeyManager()
+
+    scripts = gm_manager.scripts_for(QUrl(url))
+    assert (len(scripts.start + scripts.end + scripts.idle) ==
+            expected_matches)
+
+
 def test_no_metadata(caplog):
     """Run on all sites at document-end is the default."""
     _save_script("var nothing = true;\n", 'nothing.user.js')