EVOLUTION-NINJA

Edit File: xmlreader.html

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
 "http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
 <meta http-equiv="Content-Type" content="text/html">
 <style type="text/css"></style>

 <title>Libxml2 XmlTextReader Interface tutorial</title>
</head>

<body bgcolor="#fffacd" text="#000000">
<h1 align="center">Libxml2 XmlTextReader Interface tutorial</h1>

This document describes the use of the XmlTextReader streaming API added
to libxml2 in version 2.5.0 . This API is closely modeled after the <a
href="http://dotgnu.org/pnetlib-doc/System/Xml/XmlTextReader.html">XmlTextReader</a>
and <a
href="http://dotgnu.org/pnetlib-doc/System/Xml/XmlReader.html">XmlReader</a>
classes of the C# language.

This tutorial will present the key points of this API, and working
examples using both C and the Python bindings:

Table of content:
<ul>
 <li><a href="#Introducti">Introduction: why a new API</a></li>
 <li><a href="#Walking">Walking a simple tree</a></li>
 <li><a href="#Extracting">Extracting informations for the current
 node</a></li>
 <li><a href="#Extracting1">Extracting informations for the
 attributes</a></li>
 <li><a href="#Validating">Validating a document</a></li>
 <li><a href="#Entities">Entities substitution</a></li>
 <li><a href="#L1142">Relax-NG Validation</a></li>
 <li><a href="#Mixing">Mixing the reader and tree or XPath
 operations</a></li>
</ul>

<h2><a name="Introducti">Introduction: why a new API</a></h2>

Libxml2 <a href="http://xmlsoft.org/html/libxml-tree.html">main API is
tree based</a>, where the parsing operation results in a document loaded
completely in memory, and expose it as a tree of nodes all availble at the
same time. This is very simple and quite powerful, but has the major
limitation that the size of the document that can be hamdled is limited by
the size of the memory available. Libxml2 also provide a <a
href="http://www.saxproject.org/">SAX</a> based API, but that version was
designed upon one of the early <a
href="http://www.jclark.com/xml/expat.html">expat</a> version of SAX, SAX is
also not formally defined for C. SAX basically work by registering callbacks
which are called directly by the parser as it progresses through the document
streams. The problem is that this programming model is relatively complex,
not well standardized, cannot provide validation directly, makes entity,
namespace and base processing relatively hard.

The <a
href="http://dotgnu.org/pnetlib-doc/System/Xml/XmlTextReader.html">XmlTextReader
API from C#</a> provides a far simpler programming model. The API acts as a
cursor going forward on the document stream and stopping at each node in the
way. The user's code keeps control of the progress and simply calls a
Read() function repeatedly to progress to each node in sequence in document
order. There is direct support for namespaces, xml:base, entity handling and
adding DTD validation on top of it was relatively simple. This API is really
close to the <a href="http://www.w3.org/TR/DOM-Level-2-Core/">DOM Core
specification</a> This provides a far more standard, easy to use and powerful
API than the existing SAX. Moreover integrating extension features based on
the tree seems relatively easy.

In a nutshell the XmlTextReader API provides a simpler, more standard and
more extensible interface to handle large documents than the existing SAX
version.

<h2><a name="Walking">Walking a simple tree</a></h2>

Basically the XmlTextReader API is a forward only tree walking interface.
The basic steps are:
<ol>
 <li>prepare a reader context operating on some input</li>
 <li>run a loop iterating over all nodes in the document</li>
 <li>free up the reader context</li>
</ol>

Here is a basic C sample doing this:
<pre>#include &lt;libxml/xmlreader.h&gt;

void processNode(xmlTextReaderPtr reader) {
    /* handling of a node in the tree */
}

int streamFile(char *filename) {
    xmlTextReaderPtr reader;
    int ret;

reader = xmlNewTextReaderFilename(filename);
 if (reader != NULL) {
 ret = xmlTextReaderRead(reader);
 while (ret == 1) {
 processNode(reader);
 ret = xmlTextReaderRead(reader);
 }
 xmlFreeTextReader(reader);
 if (ret != 0) {
 printf("%s : failed to parse\n", filename);
 }
 } else {
 printf("Unable to open %s\n", filename);
 }
}</pre>

A few things to notice:
<ul>
 <li>the include file needed : <code>libxml/xmlreader.h</code></li>
 <li>the creation of the reader using a filename</li>
 <li>the repeated call to xmlTextReaderRead() and how any return value
 different from 1 should stop the loop</li>
 <li>that a negative return means a parsing error</li>
 <li>how xmlFreeTextReader() should be used to free up the resources used by
 the reader.</li>
</ul>

Here is similar code in python for exactly the same processing:
<pre>import libxml2

def processNode(reader):
    pass

def streamFile(filename):
    try:
        reader = libxml2.newTextReaderFilename(filename)
    except:
        print "unable to open %s" % (filename)
        return

ret = reader.Read()
    while ret == 1:
        processNode(reader)
        ret = reader.Read()

if ret != 0:
 print "%s : failed to parse" % (filename)</pre>

The only things worth adding are that the <a
href="http://dotgnu.org/pnetlib-doc/System/Xml/XmlTextReader.html">xmlTextReader
is abstracted as a class like in C#</a> with the same method names (but the
properties are currently accessed with methods) and that one doesn't need to
free the reader at the end of the processing. It will get garbage collected
once all references have disapeared.

<h2><a name="Extracting">Extracting information for the current node</a></h2>

So far the example code did not indicate how information was extracted
from the reader. It was abstrated as a call to the processNode() routine,
with the reader as the argument. At each invocation, the parser is stopped on
a given node and the reader can be used to query those node properties. Each
Property is available at the C level as a function taking a single
xmlTextReaderPtr argument whose name is
<code>xmlTextReader</code>Property , if the return type is an
<code>xmlChar *</code> string then it must be deallocated with
<code>xmlFree()</code> to avoid leaks. For the Python interface, there is a
Property method to the reader class that can be called on the
instance. The list of the properties is based on the <a
href="http://dotgnu.org/pnetlib-doc/System/Xml/XmlTextReader.html">C#
XmlTextReader class</a> set of properties and methods:
<ul>
 <li>NodeType: The node type, 1 for start element, 15 for end of
 element, 2 for attributes, 3 for text nodes, 4 for CData sections, 5 for
 entity references, 6 for entity declarations, 7 for PIs, 8 for comments,
 9 for the document nodes, 10 for DTD/Doctype nodes, 11 for document
 fragment and 12 for notation nodes.</li>
 <li>Name: the <a
 href="http://www.w3.org/TR/REC-xml-names/#ns-qualnames">qualified
 name</a> of the node, equal to (Prefix:)LocalName.</li>
 <li>LocalName: the <a
 href="http://www.w3.org/TR/REC-xml-names/#NT-LocalPart">local name</a> of
 the node.</li>
 <li>Prefix: a shorthand reference to the <a
 href="http://www.w3.org/TR/REC-xml-names/">namespace</a> associated with
 the node.</li>
 <li>NamespaceUri: the URI defining the <a
 href="http://www.w3.org/TR/REC-xml-names/">namespace</a> associated with
 the node.</li>
 <li>BaseUri: the base URI of the node. See the <a
 href="http://www.w3.org/TR/xmlbase/">XML Base W3C specification</a>.</li>
 <li>Depth: the depth of the node in the tree, starts at 0 for the
 root node.</li>
 <li>HasAttributes: whether the node has attributes.</li>
 <li>HasValue: whether the node can have a text value.</li>
 <li>Value: provides the text value of the node if present.</li>
 <li>IsDefault: whether an Attribute node was generated from the
 default value defined in the DTD or schema (unsupported
 yet).</li>
 <li>XmlLang: the <a
 href="http://www.w3.org/TR/REC-xml#sec-lang-tag">xml:lang</a> scope
 within which the node resides.</li>
 <li>IsEmptyElement: check if the current node is empty, this is a
 bit bizarre in the sense that <code>&lt;a/&gt;</code> will be considered
 empty while <code>&lt;a&gt;&lt;/a&gt;</code> will not.</li>
 <li>AttributeCount: provides the number of attributes of the
 current node.</li>
</ul>

Let's look first at a small example to get this in practice by redefining
the processNode() function in the Python example:
<pre>def processNode(reader):
 print "%d %d %s %d" % (reader.Depth(), reader.NodeType(),
 reader.Name(), reader.IsEmptyElement())</pre>

and look at the result of calling streamFile("tst.xml") for various
content of the XML test file.

For the minimal document "<code>&lt;doc/&gt;</code>" we get:
<pre>0 1 doc 1</pre>

Only one node is found, its depth is 0, type 1 indicate an element start,
of name "doc" and it is empty. Trying now with
"<code>&lt;doc&gt;&lt;/doc&gt;</code>" instead leads to:
<pre>0 1 doc 0
0 15 doc 0</pre>

The document root node is not flagged as empty anymore and both a start
and an end of element are detected. The following document shows how
character data are reported:
<pre>&lt;doc&gt;&lt;a/&gt;&lt;b&gt;some text&lt;/b&gt;
&lt;c/&gt;&lt;/doc&gt;</pre>

We modifying the processNode() function to also report the node Value:
<pre>def processNode(reader):
 print "%d %d %s %d %s" % (reader.Depth(), reader.NodeType(),
 reader.Name(), reader.IsEmptyElement(),
 reader.Value())</pre>

The result of the test is:
<pre>0 1 doc 0 None
1 1 a 1 None
1 1 b 0 None
2 3 #text 0 some text
1 15 b 0 None
1 3 #text 0

1 1 c 1 None
0 15 doc 0 None</pre>

There are a few things to note:
<ul>
 <li>the increase of the depth value (first row) as children nodes are
 explored</li>
 <li>the text node child of the b element, of type 3 and its content</li>
 <li>the text node containing the line return between elements b and c</li>
 <li>that elements have the Value None (or NULL in C)</li>
</ul>

The equivalent routine for <code>processNode()</code> as used by
<code>xmllint --stream --debug</code> is the following and can be found in
the xmllint.c module in the source distribution:
<pre>static void processNode(xmlTextReaderPtr reader) {
 xmlChar *name, *value;

name = xmlTextReaderName(reader);
    if (name == NULL)
        name = xmlStrdup(BAD_CAST "--");
    value = xmlTextReaderValue(reader);

printf("%d %d %s %d",
 xmlTextReaderDepth(reader),
 xmlTextReaderNodeType(reader),
 name,
 xmlTextReaderIsEmptyElement(reader));
 xmlFree(name);
 if (value == NULL)
 printf("\n");
 else {
 printf(" %s\n", value);
 xmlFree(value);
 }
}</pre>

<h2><a name="Extracting1">Extracting information for the attributes</a></h2>

The previous examples don't indicate how attributes are processed. The
simple test "<code>&lt;doc a="b"/&gt;</code>" provides the following
result:
<pre>0 1 doc 1 None</pre>

This proves that attribute nodes are not traversed by default. The
HasAttributes property allow to detect their presence. To check
their content the API has special instructions. Basically two kinds of operations
are possible:
<ol>
 <li>to move the reader to the attribute nodes of the current element, in
 that case the cursor is positionned on the attribute node</li>
 <li>to directly query the element node for the attribute value</li>
</ol>

In both case the attribute can be designed either by its position in the
list of attribute (MoveToAttributeNo or GetAttributeNo) or
by their name (and namespace):
<ul>
 <li>GetAttributeNo(no): provides the value of the attribute with
 the specified index no relative to the containing element.</li>
 <li>GetAttribute(name): provides the value of the attribute with
 the specified qualified name.</li>
 <li>GetAttributeNs(localName, namespaceURI): provides the value of the
 attribute with the specified local name and namespace URI.</li>
 <li>MoveToAttributeNo(no): moves the position of the current
 instance to the attribute with the specified index relative to the
 containing element.</li>
 <li>MoveToAttribute(name): moves the position of the current
 instance to the attribute with the specified qualified name.</li>
 <li>MoveToAttributeNs(localName, namespaceURI): moves the position
 of the current instance to the attribute with the specified local name
 and namespace URI.</li>
 <li>MoveToFirstAttribute: moves the position of the current
 instance to the first attribute associated with the current node.</li>
 <li>MoveToNextAttribute: moves the position of the current
 instance to the next attribute associated with the current node.</li>
 <li>MoveToElement: moves the position of the current instance to
 the node that contains the current Attribute node.</li>
</ul>

After modifying the processNode() function to show attributes:
<pre>def processNode(reader):
 print "%d %d %s %d %s" % (reader.Depth(), reader.NodeType(),
 reader.Name(), reader.IsEmptyElement(),
 reader.Value())
 if reader.NodeType() == 1: # Element
 while reader.MoveToNextAttribute():
 print "-- %d %d (%s) [%s]" % (reader.Depth(), reader.NodeType(),
 reader.Name(),reader.Value())</pre>

The output for the same input document reflects the attribute:
<pre>0 1 doc 1 None
-- 1 2 (a) [b]</pre>

There are a couple of things to note on the attribute processing:
<ul>
 <li>Their depth is the one of the carrying element plus one.</li>
 <li>Namespace declarations are seen as attributes, as in DOM.</li>
</ul>

<h2><a name="Validating">Validating a document</a></h2>

Libxml2 implementation adds some extra features on top of the XmlTextReader
API. The main one is the ability to DTD validate the parsed document
progressively. This is simply the activation of the associated feature of the
parser used by the reader structure. There are a few options available
defined as the enum xmlParserProperties in the libxml/xmlreader.h header
file:
<ul>
 <li>XML_PARSER_LOADDTD: force loading the DTD (without validating)</li>
 <li>XML_PARSER_DEFAULTATTRS: force attribute defaulting (this also imply
 loading the DTD)</li>
 <li>XML_PARSER_VALIDATE: activate DTD validation (this also imply loading
 the DTD)</li>
 <li>XML_PARSER_SUBST_ENTITIES: substitute entities on the fly, entity
 reference nodes are not generated and are replaced by their expanded
 content.</li>
 <li>more settings might be added, those were the one available at the 2.5.0
 release...</li>
</ul>

The GetParserProp() and SetParserProp() methods can then be used to get
and set the values of those parser properties of the reader. For example
<pre>def parseAndValidate(file):
 reader = libxml2.newTextReaderFilename(file)
 reader.SetParserProp(libxml2.PARSER_VALIDATE, 1)
 ret = reader.Read()
 while ret == 1:
 ret = reader.Read()
 if ret != 0:
 print "Error parsing and validating %s" % (file)</pre>

This routine will parse and validate the file. Error messages can be
captured by registering an error handler. See python/tests/reader2.py for
more complete Python examples. At the C level the equivalent call to cativate
the validation feature is just:
<pre>ret = xmlTextReaderSetParserProp(reader, XML_PARSER_VALIDATE, 1)</pre>

and a return value of 0 indicates success.

<h2><a name="Entities">Entities substitution</a></h2>

By default the xmlReader will report entities as such and not replace them
with their content. This default behaviour can however be overriden using:

<code>reader.SetParserProp(libxml2.PARSER_SUBST_ENTITIES,1)</code>

<h2><a name="L1142">Relax-NG Validation</a></h2>

Introduced in version 2.5.7

Libxml2 can now validate the document being read using the xmlReader using
Relax-NG schemas. While the Relax NG validator can't always work in a
streamable mode, only subsets which cannot be reduced to regular expressions
need to have their subtree expanded for validation. In practice it means
that, unless the schemas for the top level element content is not expressable
as a regexp, only chunk of the document needs to be parsed while
validating.

The steps to do so are:
<ul>
 <li>create a reader working on a document as usual</li>
 <li>before any call to read associate it to a Relax NG schemas, either the
 preparsed schemas or the URL to the schemas to use</li>
 <li>errors will be reported the usual way, and the validity status can be
 obtained using the IsValid() interface of the reader like for DTDs.</li>
</ul>

Example, assuming the reader has already being created and that the schema
string contains the Relax-NG schemas:
<pre><code>rngp = libxml2.relaxNGNewMemParserCtxt(schema, len(schema)) 
rngs = rngp.relaxNGParse() 
reader.RelaxNGSetSchema(rngs) 
ret = reader.Read() 
while ret == 1: 
 ret = reader.Read() 
if ret != 0: 
 print "Error parsing the document" 
if reader.IsValid() != 1: 
 print "Document failed to validate"</code> 
</pre>

See <code>reader6.py</code> in the sources or documentation for a complete
example.

<h2><a name="Mixing">Mixing the reader and tree or XPath operations</a></h2>

Introduced in version 2.5.7

While the reader is a streaming interface, its underlying implementation
is based on the DOM builder of libxml2. As a result it is relatively simple
to mix operations based on both models under some constraints. To do so the
reader has an Expand() operation allowing to grow the subtree under the
current node. It returns a pointer to a standard node which can be
manipulated in the usual ways. The node will get all its ancestors and the
full subtree available. Usual operations like XPath queries can be used on
that reduced view of the document. Here is an example extracted from
reader5.py in the sources which extract and prints the bibliography for the
"Dragon" compiler book from the XML 1.0 recommendation:
<pre>f = open('../../test/valid/REC-xml-19980210.xml')
input = libxml2.inputBuffer(f)
reader = input.newTextReader("REC")
res=""
while reader.Read():
 while reader.Name() == 'bibl':
 node = reader.Expand() # expand the subtree
 if node.xpathEval("@id = 'Aho'"): # use XPath on it
 res = res + node.serialize()
 if reader.Next() != 1: # skip the subtree
 break;</pre>

Note, however that the node instance returned by the Expand() call is only
valid until the next Read() operation. The Expand() operation does not
affects the Read() ones, however usually once processed the full subtree is
not useful anymore, and the Next() operation allows to skip it completely and
process to the successor or return 0 if the document end is reached.

<a href="mailto:xml@gnome.org">Daniel Veillard</a>

$Id$

</body>
</html>