https_everywhere_checker syntax fixes (#18987)

* https_everywhere_checker: Fix regex syntax

Fixes #18985

* check_rules.py: Fix unclosed files

Fixes #18986

Author: jayvdb

test/rules/src/https_everywhere_checker/check_rules.py

@@ -237,7 +237,7 @@ def disableRuleset(ruleset, problemRules, urlCount):
             disableMessage = "The following targets have been disabled at {}:\n".format(timestamp)
             host = urllib.parse.urlparse(rule)
             logging.info("Disabling target {}".format(host.netloc))
-            contents = re.sub('<[ \n]*target[ \n]+host[ \n]*=[ \n]*"{}"[ \n]*\/?[ \n]*>'.format(host.netloc),
+            contents = re.sub('<[ \n]*target[ \n]+host[ \n]*=[ \n]*"{}"[ \n]*/?[ \n]*>'.format(host.netloc),
                 '<!-- target host="{}" /-->'.format(host.netloc), contents);
 
     # Since the problems are going to be inserted into an XML comment, they cannot
@@ -266,7 +266,8 @@ def disableRuleset(ruleset, problemRules, urlCount):
 
 def skipFile(filename):
     hasher = hashlib.new('sha256')
-    hasher.update(open(filename, 'rb').read())
+    with open(filename, 'rb') as f:
+        hasher.update(f.read())
     if hasher.digest() in skipdict:
         return True
     else:
@@ -388,7 +389,8 @@ def cli():
                 "Skipping rule file '{}', matches skiplist.".format(xmlFname))
             continue
 
-        ruleset = Ruleset(etree.parse(open(xmlFname, "rb")).getroot(), xmlFname)
+        with open(xmlFname, "rb") as f:
+            ruleset = Ruleset(etree.parse(f).getroot(), xmlFname)
         if ruleset.defaultOff and not includeDefaultOff:
             logging.debug("Skipping rule '{}', reason: {}".format(
                           ruleset.name, ruleset.defaultOff))

test/rules/src/https_everywhere_checker/rules.py

@@ -324,18 +324,18 @@ def getCoverageProblems(self):
             # According to the logic in rules.js available at
             # EFForg/https-everywhere/blob/07fe9bd51456cc963c2d99e327f3183e032374ee/chromium/rules.js#L404
             #
-            pattern = target.replace('.', '\.')  # .replace('*', '.+')
+            pattern = target.replace('.', r'\.')  # .replace('*', '.+')
 
             # `*.example.com` matches `bar.example.com` and `foo.bar.example.com` etc.
             if pattern[0] == '*':
                 pattern = pattern.replace('*', '.+')
 
             # however, `example.*` match `example.com` but not `example.co.uk`
             if pattern[-1] == '*':
-                pattern = pattern.replace('*', '[^\.]+')
+                pattern = pattern.replace('*', '[^.]+')
 
             # `www.*.example.com` match `www.image.example.com` but not `www.ssl.image.example.com`
-            pattern = pattern.replace('*', '[^\.]+')
+            pattern = pattern.replace('*', '[^.]+')
 
             pattern = '^' + pattern + '$'
 
@@ -356,15 +356,15 @@ def getCoverageProblems(self):
             # Don't treat the question mark in non-capturing and lookahead groups as increasing the
             # number of required tests.
             needed_count = needed_count - \
-                len(regex.findall("\(\?:", rule.fromPattern))
+                len(regex.findall(r"\(\?:", rule.fromPattern))
             needed_count = needed_count - \
-                len(regex.findall("\(\?!", rule.fromPattern))
+                len(regex.findall(r"\(\?!", rule.fromPattern))
             needed_count = needed_count - \
-                len(regex.findall("\(\?=", rule.fromPattern))
+                len(regex.findall(r"\(\?=", rule.fromPattern))
             # Don't treat escaped questions marks as increasing the number of required
             # tests.
             needed_count = needed_count - \
-                len(regex.findall("\\?", rule.fromPattern))
+                len(regex.findall(r"\?", rule.fromPattern))
             actual_count = len(rule.tests)
             if actual_count < needed_count:
                 problems.append("{}: Not enough tests ({} vs {}) for {}".format(
@@ -374,9 +374,9 @@ def getCoverageProblems(self):
             needed_count = 1 + \
                 len(regex.findall("[+*?|]", exclusion.exclusionPattern))
             needed_count = needed_count - \
-                len(regex.findall("\(\?:", exclusion.exclusionPattern))
+                len(regex.findall(r"\(\?:", exclusion.exclusionPattern))
             needed_count = needed_count - \
-                len(regex.findall("\\?", rule.fromPattern))
+                len(regex.findall(r"\?", rule.fromPattern))
             actual_count = len(exclusion.tests)
             if actual_count < needed_count:
                 problems.append("{}: Not enough tests ({} vs {}) for {}".format(