XML::Smart::Tutorial - Tutorial and examples for XML::Smart.

This document is a tutorial for XML::Smart and shows some examples of usual things.

In XML::Smart the key CONTENT is reserved and shouldn't be used directly, since XML::Smart will deal with the convertion of arguments to node contents, including multiple node contents autimatically.

  $xml->{root}{foo} = 'simple value' ;

Here foo will be a normal argument/attribute value, and will generate this XML data:

  <root foo="simple value"/>

But if you insert some tag or lines in the values by default XML::Smart will convert it to a node content:

  $xml->{root}{foo} = "line0\nlien1\nline2\n" ;

And will generate that XML data:

  <root>
  <foo>line0
  lien1
  line2
  </foo>
  </root>

But what you can do if you want to force some type, let's say, have a node content with a simple value:

  $xml->{root}{foo} = 'simple value' ;
  $xml->{root}{foo}->set_node(1) ;

And will generate that XML data:

  <root>
    <foo>simple value</foo>
  </root>

When you have interpolated content/data you need to work in a different. Let's say that you load this XML data:

  <root>
  content0
  <tag1 arg="1"/>
  content1
  </root>

If you access directly the root key as string you will get all the content parts grouped. So, this code:

  my $xml = new XML::Smart(q`
  <root>
  content0
  <tag1 arg="1"/>
  content1
  </root>
  `,'smart') ;

  print "#$xml->{root}#" ;

Will print that:

  #
  content0

  content1
  #

To access each part of the content independently you should use an array that receive the method BIBIcontent()BI:

  my @content = $xml->{root}->content ;

  print "#$content[0]#\n" ;

And this will print that:

  #
  content0
  #

Now to set the multiple content values you should use the method BIBIcontent()BI with 2 arguments:

  $xml->{root}->content(0,'new content') ;

And now the XML data produced will be:

  <root>new content<tag1 arg="1"/>
  content1
  </root>

If you use the method content() with only one argument it will remove all the multiple contents and will set the new value in the place of the 1st content.

By defaul XML::Smart will use XML::Parser or XML::Smart::Parser (in this order of preference) to load a XML data.

To force or define by your self the parser you can use the 2nd argument option when creating a XML::Smart object:

  my $xml = new XML::Smart( 'some.xml' , 'XML::Parser' ) ;

  ## and

  my $xml = new XML::Smart( 'some.xml' , 'XML::Smart::Parser' ) ;

XML::Smart also has an extra parser, XML::Smart::HTMLParser, that can be used to load HTML as XML, or to load wild XML data:

  my $xml = new XML::Smart( 'some.xml' , 'XML::Smart::HTMLParser' ) ;

Aliases for the parser options:

  SMART|REGEXP   => XML::Smart::Parser
  HTML           => XML::Smart::HTMLParser

So, you can use as:

  my $xml = new XML::Smart( 'some.xml' , 'smart' ) ;
  my $xml = new XML::Smart( 'some.xml' , 'html' ) ;

You can customize the way that the parser will treat the XML data:

  ## For lower case:

  my $xml = new XML::Smart( 'some.xml' ,
  lowtag => 1 ,
  lowarg => 1 ,
  ) ;

  ## For uper case:

  my $xml = new XML::Smart( 'some.xml' ,
  upertag => 1 ,
  uperarg => 1 ,
  ) ;

** Note, this option will work only when the XML is parsed by XML::Smart::HTMLParser, since only it accept arguments without values!

  my $xml = new XML::Smart(
  '<root><foo arg1="" flag></root>' ,
  'XML::Smart::HTMLParser' ,
  arg_single => 1 ,
  ) ;

Here's the tree of the example above:

  'root' => {
              'foo' => {
                         'flag' => 1,
                         'arg1' => ''
                       },
            },

XML::Smart can redirect the parsing process to personalized functions:

  my $xml = XML::Smart->new( 'some.xml' ,
  on_start => \&on_start ,
  on_char  => \&on_char ,
  on_end   => \&on_end ,
  ) ;

  sub on_start {
    my ( $tag , $pointer , $pointer_back ) = @_ ;
    $pointer->{$tag}{type_user} = 1 if $tag =~ /(?:name|age)/ ;
  }

  sub on_char {
    my ( $tag , $pointer , $pointer_back , $content) = @_ ;
    $$content =~ s/\s+/ /gs ;
  }

  sub on_end {
    my ( $tag , $pointer , $pointer_back ) = @_ ;
    $pointer->{$tag}{type_extra} = 1 if $tag =~ /(?:more|tel|address)/ ;
  }

Graciliano M. P. <gm@virtuasites.com.br>

I will appreciate any type of feedback (include your opinions and/or suggestions). ;-P

Enjoy and thanks for who are enjoying this tool and have sent e-mails! ;-P

This document was written in ePod (easy-POD), than converted to POD, and from here you know the way.