[
Lists Home |
Date Index |
Thread Index
]
The world badly needs a "standard" pull API. Nothing is driving
a wedge between the data-heads and the doc-heads as
forcefully as the lack of a pull API.
I took a shot at something similar last year called RAX:
http://www.xml.com/pub/a/2000/04/26/rax
Sean
At 17:11 17/12/2001 -0500, John Cowan wrote:
>We currently have two base-level APIs for XML processing in
>the Java world: DOM and its variant JDOM, which build in-memory
>trees, and SAX, which pushes a stream of events to application
>event-handling methods.
>
>This is a first design for XMLIterator, a third base-level API
>which allows an application to pull content from XML. This
>avoids the memory demand and navigation issues of DOM, and
>is a more straightforward programming model than SAX, which
>requires magic data connections between the event handlers in
>order to maintain application state. XMLIterator extends
>the familiar Iterator interface, so it models an XML document
>as a linear collection of partially specified nodes.
>
>I am asking this list to help me refine the design of XMLIterator,
>and then build two implementations: SAXAdapter and DOMAdapter,
>for layering over SAX and DOM parsers respectively.
>
>
>// This is version 0.1 of XMLIterator
>// It supports SAX2 events only, and does not handle
>// prefix-mapping events (because I haven't figured out
>// what the Right Thing is). Stuff provided only by
>// DOM should be factored in too.
>
>package org.ccil.cowan.iter;
>public interface XMLIterator
> extends java.util.Iterator {
>
>// Processing model: Iteration starts with the
>// ELEMENT node of the document element, and on successive
>// calls to next(), proceeds through the document,
>// returning ELEMENT, ATTRIBUTE, SKIPPED_ENTITY,
>// END_ELEMENT, and PI nodes in document order
>// (except that ATTRIBUTE nodes appear just after
>// their owner ELEMENT node in arbitrary order),
>// followed by the END node.
>
>// hasNext is inherited from Iterator, and returns true
>// if the current node is not an END node
>
>// next is inherited from Iterator, and returns an XMLNode
>// object, which may be self; XMLIterators are encouraged
>// to play the role of both the iterator and the component
>// object, to avoid excessive object creation (as such,
>// the XMLNode is considered invalid after the following
>// invocation of next).
>
>// remove is inherited from Iterator, and throws an
>// exception, since XMLIterator is read-only
>
>// Node types: legal return values of XMLNode.getType method
>public static int END = 0;
>public static int ELEMENT = 1;
>public static int ATTRIBUTE = 2;
>public static int SKIPPED_ENTITY = 3;
>public static int END_ELEMENT = 4;
>public static int PI = 5;
>
>// Attribute types: legal return values of getAttributeType method
>public static int CDATA = 0;
>public static int ID = 1;
>public static int IDREF = 2;
>public static int IDREFS = 3;
>public static int NMTOKEN = 4; // also used for enumerations
>public static int NMTOKENS = 5;
>public static int ENTITY = 6;
>public static int ENTITIES = 7;
>public static int NOTATION = 8;
>
>
>// Convenience methods
>
>// If the current node is an ELEMENT or ATTRIBUTE node,
>// skip all nodes to the next non-ATTRIBUTE node.
>// This allows us to ignore attributes if we do not care
>// about any more.
>public void skipAttributes();
>
>// Skip all nodes up to and including the END_ELEMENT node
>// corresponding to the most recently seen ELEMENT node
>// (the current node, if that is an ELEMENT node)
>public void skipElement();
>
>}
>
>
>package org.ccil.cowan.iter;
>public interface XMLNode {
>
>// Read-only properties of the current node
>// Lazy implementation is encouraged
>
>// Returns the current Locator object
>public org.xml.sax.Locator getLocator();
>
>// Returns a node type code
>public int getType();
>
>// If the current node is an ATTRIBUTE node, return
>// an attribute type code
>public int getAttributeType();
>
>// If the current node is an ELEMENT or ATTRIBUTE node,
>// and namespace URI information is available,
>// return it
>public string getNSURI();
>
>// If the current node is an ELEMENT or ATTRIBUTE node,
>// and QName information is available, return it
>public string getQName();
>
>// If the current node is an ELEMENT or ATTRIBUTE node, and
>// local name information is available, return it;
>// if the current node is a PI node, return the target;
>// if the current node is a SKIPPED_ENTITY node,
>// return the entity name
>public string getLocalName();
>
>// If the current node is an ELEMENT node,
>// return an Attributes object containing the attributes
>public org.xml.sax.Attributes getAttributes();
>
>// If the current node is an ELEMENT node,
>// return all text content up to the next tag;
>// if the current node is an ATTRIBUTE node,
>// return the normalized attribute value;
>// if the current node is an END_ELEMENT node,
>// return all text content up to the next tag;
>// if the current node is a PI node,
>// return the content of the PI
>public string getValue();
>
>// If the current node is an ELEMENT or END_ELEMENT node,
>// return true if the value property consists of ignorable whitespace
>public bool isIgnorableWhitespace();
>
>}
>
>
>--
>Not to perambulate || John Cowan <jcowan@reutershealth.com>
> the corridors || http://www.reutershealth.com
>during the hours of repose || http://www.ccil.org/~cowan
> in the boots of ascension. \\ Sign in Austrian ski-resort hotel
>
>
>-----------------------------------------------------------------
>The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
>initiative of OASIS <http://www.oasis-open.org>
>
>The list archives are at http://lists.xml.org/archives/xml-dev/
>
>To subscribe or unsubscribe from this list use the subscription
>manager: <http://lists.xml.org/ob/adm.pl>
>
http://www.propylon.com
|
