-------------------------------------------------------------------------------
XML -- eXtensibe Markup Language

Essentually based on SGML on which HTML is based.  As such it is very
commonly used on the WWW.  Basically an extended and generalized form.

Because of this most browsers can directly pretty print XML data.

NOTE: Unlike HTML, all tags must be closed.


Example...  Recent Documents List...
   ~/.local/share/recently-used.xbel


A DTD at the start (or included) gives a hint to how the data structure is
arranged before the data begins.  (EG: syntax)

-----
Examples...
These two files are equivelent

=======8<--------
<!-- Comment -->
<data>
  <users>
    <user login="grep" fullname="Gary R Epstein" />
    <user login="stty" fullname="Simon T Tyson" />
  </users>
  <timeout>50</timeout>
</data>
=======8<--------

=======8<--------
<data>
  <users>
    <user> <login>grep</login> <fullname>Gary R Epstein</fullname></user>
    <user> <login>stty</login> <fullname>Simon T Tyson</fullname></user>
  </users>
  <timeout>50</timeout>
</data>
=======8<--------

White space is however preserved...
These are not the same

=======8<--------
<child>
  Text
</child>
=======8<--------

=======8<--------CUT HERE----------
<child>Text</child>
=======8<--------CUT HERE----------

-------------------------------------------------------------------------------
Inclusion

The top level <tag> can include
     xmlns:xi="http://www.w3.org/2001/XInclude"
Which specifies a scheme to allow XML files to include the contents from
other files...   If the XML reader follows that syntax!

  https://en.wikipedia.org/wiki/XInclude

=======8<--------
<main_config xmlns:xi="http://www.w3.org/2001/XInclude">

  <xi:include href="included_config.xml"/>

  <comment><xi:include href="included_text.txt" parse="text"/></comment>
</main_config>
=======8<--------

Openbox is known to follow this syntax, not certain of other XML readers.

-------------------------------------------------------------------------------
Document Type Definitions (DTDs)

DTD can be declared externally...

=======8<--------
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE bookstore SYSTEM "Bookstore.dtd">
...
=======8<--------

Or internally

=======8<--------
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE bookstore [
<!ELEMENT bookstore (book+)>
<!ELEMENT book (title, price)>
<!ATTLIST book category CDATA "Literature">
<!ELEMENT title (#PCDATA)>
<!ELEMENT price (#PCDATA)>
]>

<bookstore>
  <book category="COOKING">
    <title>Everyday Italian</title>
    <price>30.00</price>
  </book>
</bookstore>
=======8<--------

-------------------------------------------------------------------------------
Shell

   xmlstarlet   Query XML documents

-------------------------------------------------------------------------------
Perl API

XML Project   http://perl-xml.sourceforge.net/
  Summery FAQ on different XML perl modules

XML::Parser
  First parser for XML started by Larry Wall.
XML::Simple
  Good for simpler forms of XML (depreciated)
XML::SAX
  Simple API for XML
XML::DOM
  Loads document completely into memory
XML::XSLT
  Hard to use?

For a basic summary see articals
  Pasring XML into a simple hash
    http://www.perlmonks.org/?node_id=90287
  Death to Dot Star!  (using regexp with XML is not good)
    http://www.perlmonks.org/?node_id=24640

-------------------------------------------------------------------------------
Perl XML::Simple

Read the XML into a single data structure, can use SAX to add handlers
Strict mode ensure things work as expected!

You can provide infomation on how the data is to be stored.
EG : Into arrays, hashed (indexed), or single values

  use XML::Simple qw(:strict);

  my $xml_data = XMLin("file.xml",
        ForceArray => [      # force array even with one element
          qw( HASH ),        # these are array/hash elements
          qr/_list$/,        # you can use regexp to select
        ],
        KeyAttr => {         # Array to Hash, specifing the key
          HASH => 'KEY',
        },
        GroupTags => {       # simply a tags that only has one sub-tag type
          TAG => 'SUBTAG',   # ->{TAG}->{SUB}  becomes just ->{TAG}
                             # SUBTAG may be a value, array, or hash
        },
        Handler => {         # use SAX to process sections
          ARRAY => &function,
        }
      );

  use Data::Dumper;
  print Dumper($xml_data);


-------------------------------------------------------------------------------
Perl XML::SAX     Simple API for XML

Uses handlers and filters to avoid reading in LARGE data structures

-------------------------------------------------------------------------------
Perl XML::Twig

Efficentally process XML files.
You can process the file while it is still being read in.

http://www.xmltwig.org/

Process can use callbacks, or just process smaller parts of the data tree.

See tutorial for the various ways of process the tree.
Especially: 4.3 The flush and purge methods

-------------------------------------------------------------------------------