XML (originally called Exemel) is a full fledged XML library for Perl 6.
It handles parsing, generating, manipulating and querying XML. It supports element queries, parent element information, namespaces, and an extendable interface.
It supports every major kind of XML Node (XML::Node):
- Document (XML::Document)
- Element (XML::Element)
- Text (XML::Text)
- Comment (XML::Comment)
- PI (XML::PI)
- CDATA (XML::CDATA)
You can easily serialize the objects back to XML text by using any XML::Node object in a string context.
A module that provides a few simple subroutines.
Parse the string as XML, and return an XML::Document object.
Slurp the IO, parse the contents, and return an XML::Document object.
Return an XML::Document object representing the specified file. You will be able to call $xml.save(); to save back to the original file.
See the XML::Element.craft() function for details on how this works.
A role used by the rest of the XML Node classes.
The XML::Element or XML::Document to which this Node belongs. Only an XML::Document will have an undefined $.parent property.
Removes the Node from its parent element.
Removes the Node from its existing parent (if any) and sets the specified node as it's $.parent property.
Returns the Node that exists in the parent just before this one. Returns Nil if there is none.
Returns the Node that exists in the parent just after this one. Returns Nil if there is none.
This is a polymorphic method that exists in all XML::Node objects, and does what is needed to return a clone of the desired Node.
Returns the top-level XML::Document that this Node belongs to.
A Node representing an XML document. You can use array access syntax on it to access children of the root node. You can use hash access syntax on it to access attributes of the root node.
The XML version. Default: '1.0'
The text encoding, if specified in the XML declarator.
Has two keys, 'type' represents the document type, 'value' represents the rest of the DOCTYPE declaration (if applicable.)
The root XML::Element of the document. This also proxies many of the useful XML::Element methods, so that they can be called directly from the XML::Document object.
If an XML::Document represents a file on the file-system, this is the path to that file.
Parse the passed XML and return an XML::Document. If the $filename variable is passed, the $.filename property will be set on the object.
Create a new XML::Document object, with the specified XML::Element as it's root element.
Create a new XML::Document object, representing the specified file. The $.filename property will be set.
Save the XML back into a file. If the $filename parameter is not passed, we use the $.filename property (if it is set, otherwise we return False.)
If the :copy option is true, we don't re-set the $.filename property.
A Node representing an individual XML element. You can use array access syntax to access child nodes, and hash access syntax to access or set attributes.
The tag name of the element.
Any child nodes that may exist. All members of @.nodes MUST be an object that does the XML::Node role.
Unless you are doing something that requires direct access of the @.nodes property, it's probably easier (and less noisy) to use the array access syntax.
XML attributes for the current node. We expect the keys and values to be strings, but you can use numeric values if you want. Remember on emitting or parsing, all values will be strings, even if you set it as a number.
It is recommended that you do not use %.attribs directly to set values. Use the set() method or the hash access syntax to set attribute values, and use the unset() method to delete attributes.
Specifies what attribute will be used as the XML Id when using the getElementById() method. This defaults to 'id' which is used in (X)HTML and thus the most common.
Return a new XML::Element object representing the specified XML string.
Insert an XML::Node at the beginning of our @.nodes list.
Create a new XML::Element with the given name, and insert it to the beginning of our nodes list. Uses craft() to build the element.
Any named parameters will be used as attributes, any positional parameters will be used as child nodes.
Positional parameters can be one of the following:
- An XML::Node object. Will be added as is.
- A String. Will be included as an XML::Text node.
- A Capture. Calls craft() using the Capture as the signature.
- Anything else, will be stringified, and added as an XML::Text node.
Append an XML::Node to the bottom of our @.nodes list.
See insert (Str $name, ...) but at the bottom.
Insert the $new Node before the $existing Node. It only works if the $existing node is actually found in our @.nodes list.
Only works if our $.parent is an XML::Element. Inserts the Node before the current one.
See insert (Str $name, ...) and before(XML::Node $node) and figure it out.
Like before($existing, $new) but put the node after the $existing one.
Like before(XML::Node $node) but put the node after the current one.
As per the others.
Insert to top, a new XML::Element representing the given XML string.
Append to bottom, a new XML::Element representing the given XML string.
Insert a new XML::Element for the XML string, before the current element.
Insert a new XML::Element for the XML string, after the current element.
An alternative to before($existing, $new) using DOM semantics.
An alternative to after($existing, $new) using DOM-like semantics.
If the $existing node is found, replace it with $new, otherwise, we return False.
An alternative to replace() with DOM semantics.
Removes the $node from our child @.nodes if it exists. If it doesn't we return False.
Return our first child node.
Return our last child node.
Pass it a smart match rule, and it will return array index of the first matching node.
Create and return a new XML::Element object with the given name. Named and positional parameters are handled as with insert(Str $name, ...)
Set an attribute with the given $name to the specified $value. If the $value is a Str or Numeric is is added directly. If it is Bool and True, the value will be set to the same as the $name. If it is Bool and False, the attribute will be deleted.
Any other value will be stringified using the .Str method.
A set() call containing no positional paramters, will pass all named parameters to the above set() as key/value pairs.
Each positional parameter passed will be assumed to be the name of an attribute to delete.
We assume the key of each named parameter passed to be the name of an attribute to delete. The value means absolutely nothing and is in fact ignored entirely.
Returns True if the given attribute exists, and has the same value as its name (the definition of an XML boolean.)
Return all child XML::Elements.
Specify a query of named parameters. Special processing parameters are used:
If set, elements must match the given tag name.
If set, elements must match the given namespace prefix.
If set, elements must match the given namespace URI.
If set to a non-zero digit, child elements will also be searched for elements matching the queries. The recursion will traverse a tree depth of the value set to this parameter.
Used with RECURSE if this is set to True, this will recurse even child elements that matched the query.
If this is set to True, we will return the first matched value. If no values match, we will return False. If SINGLE is not specified, or is set to False, we return an Array of all matches (this may be empty if no nodes matched.)
If this is set to True, instead of returning an Array of results, we will return a XML::Element object with the same name as the original input object, with its nodes set to the matched elements.
If set to an Int, the element must be the nth child to match. If set to a Range, the element's position must be within the range. If set to a Whatever match rule (e.g. * > 2) the rule must match.
If this is set to an Int, and RECURSE is not a positive value, then the SINGLE rule will be set to True.
Set to an Int, then we match if the element is not the nth child.
Match only if the element is the first child.
If RECURSE is not a positive value, then SINGLE will be set to True.
Don't include the first child in the results.
Match only if the element is the last child.
If RECURSE is not a positive value, then SINGLE will be set to True.
Don't include the last child in the results.
Match even child nodes. By default this is based on natural position (i.e. the second child element is even) see BYINDEX for details.
Match odd child nodes. By default this is based on natural position (i.e. the first child element is odd) see BYINDEX for details.
If set to True, then the EVEN and ODD rules match against the array index value rather than the natural position. Therefore, the first element will be even, since it is in position 0.
Any other named paramters not in the above list, will be assumed to be attributes that must match. You can match by value, regular expression, or whatever code matches.
my $head = $html.elements(:TAG<head>, :SINGLE); my @stylesheets = $head.elements(:TAG<link>, :rel<stylesheet>); my @middle = $table.elements(:!FIRST, :!LAST); my @not-red = $div.elements(:class(* ne 'red'));
Return the XML::Element with the given id.
Return an array of XML::Elements with the given tag name.
If the boolean $object named parameter is true, then the 'OBJECT' rule will be applied to the query sent to elements().
Return the XML Namespace prefix.
If no prefix is found, it returns an undefined value. If the URI is the default namespace, it returns ''.
Returns the URI associated with a given XML Namespace prefix. If the $prefix is not specified, return the default namespace.
Returns an undefined value if there is no XML Namespace URI assigned.
Associated the given XML Namespace prefix with the given URI. If no $prefix is specified, it sets the default Namespace.
Return an array of all XML::Comment child nodes.
Returns an array of all XML::CDATA child nodes.
Returns an array of all XML::PI child nodes.
Returns an array of all XML::Text child nodes.
A Node representing a portion of plain text.
The raw text, with no whitespace chopped out.
Return the $.text, chomping training newline, replacing multiple whitespace characters with a single space, and trimming off leading and trailing whitespace characters.
Represents an XML Comment
<!-- comment here -->
Contains the string data of the content.
Represents an XML processing instruction.
Contains the string text of the processing instruction.
Represents an XML CDATA structure.
<![CDATA[ random cdata content here ]]>
Contains the string text of the CDATA.
A quick example, for more, see the tests in the 't/' folder.
<test> <greeting en="hello">world</greeting> <for> <item>Yes</item> <item>No</item> <item>Maybe</item> <item>Who cares?</item> </for> </test>
use XML; my $xml = from-xml(:file<test.xml>); say $xml<en> ~ $xml; ## "hello world" say $xml; ## "Maybe" $xml.append('item', 'Never mind'); say $xml; ## <item>Never mind</item>
Timothy Totten, supernovus on #perl6, https://github.com/supernovus/
The XML::Grammar library was originally based on the now defunct XML::Grammar::Document library, but modified to work with Rakudo 'ng' and later 'nom', with other changes specific to this library.