Check regular expressions from users for errors.

doc/changelog.txt

@@ -6,18 +6,20 @@ Features:
   checks into plugins (see upgrading.txt for more info).
 - checking: Add options to limit the number of requests per second,
   allowed URL schemes and maximum file or download size.
-  Closes: GH bug #397, #465
+  Closes: GH bug #397, #465, #420
 - gui: UI language can be changed dynamically.
   Closes: GH bug #391
 
 Changes:
 - checking: Use the Python requests module for HTTP and HTTPS requests.
-  Closes: GH bug #393, #463
+  Closes: GH bug #393, #463, #417
 - logging: Removed download, domains and robots.txt statistics.
 - logging: HTML output is now in HTML5.
 - checking: Removed 301 warning since 301 redirects are used
   a lot without updating the old URL links.
-  Closes: GH bug #444
+  Also, recursive redirection is not checked any more since there
+  is a maximum redirection limit anyway.
+  Closes: GH bug #444, #419
 - checking: Disallowed access by robots.txt is an info now, not
   a warning. Otherwise it produces a lot of warnings which
   is counter-productive.
@@ -46,6 +48,8 @@ Fixes:
 - checking: For login pages, use twill form field counter if
   the field has neither name nor id.
   Closes: GH bug #428
+- configuration: Check regular expressions for errors.
+  Closes: GH bug #410
 
 
 8.6 "About Time" (released 8.1.2014)

linkcheck/__init__.py

@@ -121,6 +121,7 @@ def get_link_pat (arg, strict=False):
     @type strict: bool
     @return: dictionary with keys 'pattern', 'negate' and 'strict'
     @rtype: dict
+    @raises: re.error on invalid regular expressions
     """
     log.debug(LOG_CHECK, "Link pattern %r strict=%s", arg, strict)
     if arg.startswith('!'):
@@ -129,8 +130,13 @@ def get_link_pat (arg, strict=False):
     else:
         pattern = arg
         negate = False
+    try:
+        regex = re.compile(pattern)
+    except re.error as msg:
+        log.warn(LOG_CHECK, "invalid regular expression %r: %s" % (pattern, msg))
+        raise
     return {
-        "pattern": re.compile(pattern),
+        "pattern": regex,
         "negate": negate,
         "strict": strict,
     }

linkcheck/gui/__init__.py

@@ -261,16 +261,26 @@ class LinkCheckerMain (QtGui.QMainWindow, Ui_MainWindow):
             self.config.reset_loglevel()
         if data["warninglines"]:
             lines = data["warninglines"].splitlines()
-            ro = re.compile(warninglines2regex(lines))
-            self.backup_config("warningregex", ro)
+            pattern = warninglines2regex(lines)
+            try:
+                ro = re.compile(pattern)
+                self.backup_config("warningregex", ro)
+            except re.error as err:
+                msg = _("Invalid regular expression %r: %s" % (pattern, err))
+                self.set_statusmsg(msg)
+
         # set ignore patterns
         ignorepats = data["ignorelines"].strip()
         if ignorepats:
             self.backup_config("externlinks")
             lines = ignorepats.splitlines()
             for line in lines:
-                pat = get_link_pat(line, strict=1)
-                self.config["externlinks"].append(pat)
+                try:
+                    pat = get_link_pat(line, strict=1)
+                    self.config["externlinks"].append(pat)
+                except re.error as err:
+                    msg = _("Invalid regular expression %r: %s" % (pat, err))
+                    self.set_statusmsg(msg)
 
     def backup_config (self, key, value=None):
         """Backup config key if not already done and set given value."""

linkcheck/plugins/regexcheck.py

@@ -37,10 +37,13 @@ class RegexCheck(_ContentPlugin):
     def __init__(self, config):
         """Set warning regex from config."""
         super(RegexCheck, self).__init__(config)
-        if config["warningregex"]:
-            self.warningregex = re.compile(config["warningregex"])
-        else:
-            self.warningregex = None
+        self.warningregex = None
+        pattern = config["warningregex"]
+        if pattern:
+            try:
+                self.warningregex = re.compile(pattern)
+            except re.error as msg:
+                log.warn(LOG_PLUGIN, "Invalid regex pattern %r: %s" % (pattern, msg))
 
     def check(self, url_data):
         """Check content."""