- From: Ville Skytta via cvs-syncmail <cvsmail@w3.org>
- Date: Mon, 14 Dec 2009 22:39:20 +0000
- To: www-validator-cvs@w3.org
Update of /sources/public/validator/httpd/cgi-bin In directory hutz:/tmp/cvs-serv20479/httpd/cgi-bin Modified Files: check Log Message: Clean up XML declaration regex. Index: check =================================================================== RCS file: /sources/public/validator/httpd/cgi-bin/check,v retrieving revision 1.756 retrieving revision 1.757 diff -u -d -r1.756 -r1.757 --- check 14 Dec 2009 22:19:09 -0000 1.756 +++ check 14 Dec 2009 22:39:18 -0000 1.757 @@ -647,7 +647,7 @@ # handle a structured error (XML::LibXML::Error object) # (lib XML::LibXML > 0.66, but will work MUCH better > 0.69 ) - my $err_obj = $@; + my $err_obj = $@; my $num_xmlwf_error = 0; while ($err_obj) { my $err; @@ -2742,22 +2742,22 @@ # the 10 first lines should be safe my $parseModeFromXMLDecl = ( $begincontent =~ - /^ [\x20|\x09|\x0D|\x0A]* # whitespace before the decl should not be happening + /^ [\x20\x09\x0D\x0A]* # whitespace before the decl should not be happening # but we are greedy for the sake of detection, not validation <\?xml # start matching an XML Declaration - [\x20|\x09|\x0D|\x0A]+ # x20, x09, xD and xA are the allowed "xml white space" - version [\x20|\x09|\x0D|\x0A]* = # for documents, version info is mandatory - [\x20|\x09|\x0D|\x0A]* ("1.0"|"1.1"|'1.0'|'1.1') # hardcoding the existing XML versions. + [\x20\x09\x0D\x0A]+ # x20, x09, xD and xA are the allowed "xml white space" + version [\x20\x09\x0D\x0A]* = # for documents, version info is mandatory + [\x20\x09\x0D\x0A]* (["'])1.[01]\1 # hardcoding the existing XML versions. # Maybe we should use \d\.\d - ([\x20|\x09|\x0D|\x0A]+ encoding - [\x20|\x09|\x0D|\x0A]* = [\x20|\x09|\x0D|\x0A]* - ("[A-Za-z][a-zA-Z0-9_-]+"|'[A-Za-z][a-zA-Z0-9_-]+') + (?:[\x20\x09\x0D\x0A]+ encoding + [\x20\x09\x0D\x0A]* = [\x20\x09\x0D\x0A]* + (["'])[A-Za-z][a-zA-Z0-9_-]+\2 )? # encoding info is optional - ([\x20|\x09|\x0D|\x0A]+ standalone - [\x20|\x09|\x0D|\x0A]* = [\x20|\x09|\x0D|\x0A]* - ("yes"|"no"|'yes'|'no') + (?:[\x20\x09\x0D\x0A]+ standalone + [\x20\x09\x0D\x0A]* = [\x20\x09\x0D\x0A]* + (["'])(?:yes|no)\3 )? # ditto standalone info, optional - [\x20|\x09|\x0D|\x0A]* \?> # end of XML Declaration + [\x20\x09\x0D\x0A]* \?> # end of XML Declaration /x ? 'XML' :
Received on Monday, 14 December 2009 22:39:29 UTC