XML:Simple read tag value with regex

filipebean has asked for the wisdom of the Perl Monks concerning the following question:

Hi all,

I'm using XML:Simple to read a configuration from a XML file. The code is working, but I’m facing now a problem when it read the regex that includes an xml tag. xml file below:

Config file:

<sourcetype>
  <name>G1</name>
  <desc>group 1 to decode</desc>
  <rules>
    <rule>['^\d(.)','14']</rule>
    <rule>['^(<xyz>)']</rule>
    <rule>['^(</xyz>)']</rule>
  </rules>
</sourcetype>
[download]

Script:

my $xml = new XML::Simple( KeyAttr=>[] );
my $data = $xml->XMLin( $config_file );
foreach my $sourcetype ( @{$data->{sourcetypes}{sourcetype}} )
{
  print "  " . $sourcetype->{name} . "\t\t" . $sourcetype->{desc} . "\
+n";
}
[download]

When I run the script it complains:

Opening and ending tag mismatch: xyz line 10 and rule Opening and ending tag mismatch: rule line 11 and rule at …/XML/LibXML/SAX.pm line 64 at …/5.8.4/XML/Simple.pm line 362

Is it possible to have regex as a tag value like “'^(<xyz>)'” and read it as a string?

Thank you in advance, Best regards.

Comment on XML:Simple read tag value with regex Select or Download Code

Replies are listed 'Best First'.
Re: XML:Simple read tag value with regex by toolic (Bishop) on May 07, 2013 at 13:04 UTC
Your Perl code is probably fine, but that is not valid XML syntax. Use XML entity references (something like this): `<rule>['^(<xyz>)']</rule>` [download]	[reply] [d/l]
Re: XML:Simple read tag value with regex by mirod (Canon) on May 07, 2013 at 13:19 UTC
As said before, you can escape the `<` by using `<` or you can use CDATA sections: `<sourcetype> <name>G1</name> <desc>group 1 to decode</desc> <rules> <rule><![CDATA[['^\d(.)','14']]]></rule> <rule><![CDATA[['^(<xyz>)']]]></rule> <rule><![CDATA[['^(</xyz>)']]]></rule> </rules> </sourcetype>` [download] That's still ugly, but a little easier to read than the version with `<`, and at least you can cut and paste code in the rule elements. The `<![CDATA[...]]>` construct prevents everything in the section from being parsed as XML. You still need it to be valid text (unicode by default) and not to include `]]>`, but anything else is fine.	[reply] [d/l] [select]
Re: XML:Simple read tag value with regex by kcott (Archbishop) on May 07, 2013 at 13:34 UTC
G'day filipebean, There's a number of special characters that may need to be escaped in XML content: "`<`" to "`<`"; "`>`" to "`>`"; and "`&`" to "`&`" (see XML Predefined Entities). So, `<rule>['^(<xyz>)']</rule>` [download] would become `<rule>['^(<xyz>)']</rule>` [download] Given the readability issues, it might be better to use CDATA blocks (see XML CDATA Sections) to escape the entire regex, e.g. `<rule><![CDATA[['^(<xyz>)']]]></rule>` [download] [Note: I haven't investigated how XML::Simple interacts with these constructs.] -- Ken	[reply] [d/l] [select]
Re^2: XML:Simple read tag value with regex by filipebean (Novice) on May 07, 2013 at 15:06 UTC
Hi all, thank for your answers. I tried the CDATA as you suggest and it works perfectly :) best regards.	[reply]