index.html

《Big C++ 》Third Edition电子书和代码全集-Part1

			<li>Supply name as an <tt>XMLCh</tt> array</li>
			<li><tt>XMLCh</tt> array must be deleted</li>
		</ul>
	</li>
	<li>Wrap <tt>getAttribute</tt> method in a help function:</li>
</ul>

</font>

<hr><h2><font color="#009999">29.2 Parsing XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<blockquote>
<pre>string get_attribute(DOMElement* element, string name)
{
   XMLCh* xname = new XMLCh[name.length() + 1];
   XMLString::transcode(name.c_str(),
      xname, name.length());
   XMLCh* xvalue = element-&gt;getAttribute(xname);
   delete xname;
   return XMLCh_to_string(xvalue);
}</pre>
	</blockquote>
	<li>E.g., get the <tt>currency</tt> attribute</li>
		<blockquote><tt>
				string attribute_value = get_attribute(<br>
			&nbsp;&nbsp;&nbsp;price_element, "currency");
		</tt></blockquote>
	</li>
</ul>

</font>

<hr><h2><font color="#009999">29.2 Parsing XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<li>Use <tt>DOMNamedNodeMap</tt> to iterate over attributes</li>
	<li>Returned by <tt>getAttributes</tt>:
		<blockquote><tt>
			DOMNamedNodeMap* attributes = price_element-&gt;getAttributes();
		</tt></blockquote>
	</li>
	<li>Use <tt>item</tt> method to get a particular attribute:
		<blockquote><tt>
			int i = . . .;<br>
			<font color="#0000cc">
				// # between 0 and attributes-&gt;getLength() - 1</font><br>
			DOMNode* attribute_node = attributes-&gt;item(i);<br>
			XMLCh* attribute_name = attribute_node-&gt;getNodeName();<br>
			<font color="#0000cc">// For example, "currency"</font><br>
			XMLCh* attribute_value = attribute_node.getNodeValue();<br>
			<font color="#0000cc">// For example, "USD"</font>
		</tt></blockquote>
	</li>
</ul>

</font>

<hr><h2><font color="#009999">29.2 Parsing XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<li>Text node stored in <tt>DOMText</tt></li>
	<li>If no mixed content, should be only child</li>
	<li>Use <tt>getFirstChild</tt> to retrieve</li>
	<li>Use <tt>getData</tt> to read the text</li>

	<blockquote><tt>
		DOMNode* price_element = . . .;<br>
		DOMText* price_data = <nobr>dynamic_cast&lt;DOMText*&gt;</nobr>(<br>
		&nbsp;&nbsp;&nbsp;price_element-&gt;getFirstChild());<br>
		string price_string = XMLCh_to_string(price_data-&gt;getData());<br>
		&nbsp;&nbsp;&nbsp;<font color="#0000cc">// For example, "24.95"</font><br>
		double price = string_to_double(price_string);<br>
		&nbsp;&nbsp;&nbsp;<font color="#0000cc">// For example, 24.95</font>
	</tt></blockquote>

</ul>

</font>

<hr><h2><font color="#009999">29.2 Parsing XML Documents - Parser example
	</font></h2>
<font size="+1">

<ul>
	<li>Create a helper function for each XML element</li>
	<li><tt>description</tt>, <tt>price</tt>, and <tt>quantity</tt> simple,
		so parsed inline</li>
	<li><tt>main</tt> calls <tt>XMLPlatformUtils::Initialize</tt>; necessary
		to use Xerces</li>
	<li>Program:
	 	<ul>
			<li>Parses
				<a target='bigc' href='Code/parser/items.xml'><tt>items.xml</tt></a>,
				description of a list of product items</li>
			<li>Produces a vector of <tt>Item</tt> objects</li>
		</ul>
	</li>
	<li>Uses <a target='bigc' href='Code/parser/item.h'><tt>item.h</tt></a>,
		<a target='bigc' href='Code/parser/item.cpp'><tt>item.cpp</tt></a>,
		<a target='bigc' href='Code/parser/product.h'><tt>product.h</tt></a>, and
		<a target='bigc' href='Code/parser/product.cpp'><tt>product.cpp</tt></a>
		from (chptr 13)</li>
</ul>

</font>

<hr><h2><font color="#009999">29.2 Parsing XML Documents (parser.cpp)
	</font></h2>
<font size="+1">

<script><!-- 
	iframeWrapCode( "parser/parser.cpp", "95%", "80%" )
//--></script>

</font>

<hr><h2><font color="#009999">Productivity Hint 29.1</font></h2>
<font size="+1">
<hr color="#00ffff" size="6">

<p><font color="#009999">Helper Functions in an XML Parser</font></p>

<ul>
	<li>Given an XML document of a particular structure</li>
	<li>For each element:
		<ul>
			<li>Assign a (new or existing) class or vector (for sequences of the
				same type)</li>
			<li>Provide a helper function of this form:</li>
		</ul>
	</li>
</ul>

</font>

<hr><h2><font color="#009999">Productivity Hint 29.1 (cont.)</font></h2>
<font size="+1">
<hr color="#00ffff" size="6">

<blockquote>
<pre><i>ClassForElement</i> get<i>ElementName</i>(DOMNode* e)
{
   DOMNodeList* children = e-&gt;getChildNodes();
   for (int i = 0; i &lt; children-&gt;getLength(); i++)
   {
      DOMNode* child_node = children-&gt;item(i);
      DOMElement* child_element
         = dynamic_cast&lt;DOMElement*&gt;(child_node);
      if (child_element != NULL)
      {
         Get value of child element
      }
   }
   <i>Use the child element values to construct and
   return a ClassForElement object</i>
}</pre>
</blockquote>

</font>

<hr><h2><font color="#009999">Productivity Hint 29.1 (cont.)</font></h2>
<font size="+1">
<hr color="#00ffff" size="6">

<ul>
	<li>If a child element doesn't contain text, call its
		<tt>get<i>ChildElementName</i></tt> function</li>
	<li>If it does, then read the element data:
		<blockquote>
<pre>DOMText* text_node = dynamic_cast&lt;DOMText*&gt;(
   child_element-&gt;getFirstChild());
string data = XMLCh_to_string(text_node-&gt;getData());</pre>
		</blockquote>
	</li>
	<li>Convert strings to numbers, as necessary</li>
	<li>Implement functions bottom-up</li>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents</font></h2>
<font size="+1">

<ul>
	<li>Avoid simple print statements - easy to introduce an error</li>
	<li>Build a <tt>DOMDocument</tt> object</li>
	<li>Still use <tt>DOMImplementation</tt> object to create parser:
		<blockquote><tt>
			DOMImplementation* implementation<br>
			&nbsp;&nbsp;&nbsp;= DOMImplementation::getImplementation();<br>
			DOMDocument* doc = implementation-&gt;createDocument();<br>
			doc-&gt;setStandalone(true);<br>
			&nbsp;&nbsp;&nbsp;<font color="#0000cc">// An empty document</font>
		</tt></blockquote>
	</li>
	<li>"standalone" flag indicates lack of document type definition
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<li>Use <tt>createElement</tt> and <tt>createTextNode</tt> methods
		of <tt>DOMDocument</tt></li>
	<li>Inputs are an <tt>XMLCh</tt> array</li>
	<li>So, wrap above methods, to take <tt>string</tt>s</li>
	<li>Wrapper for <tt>DOMDocument::createElement</tt>:
		<blockquote>
<pre>DOMElement* create_element(DOMDocument* doc, string name)
{
   XMLCh* xname = new XMLCh[name.length() + 1];
   XMLString::transcode(name.c_str(), xname, name.length());
   DOMElement* r = doc-&gt;createElement(xname);
   delete xname;
   return r;
}</pre>
		</blockquote>
	</li>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<li>To insert text node:
		<ul>
			<li>Convert text to <tt>XMLCh</tt> array</li>
			<li>Wrap in a text node</li>
			<li>Create an element node</li>
			<li>Set text node as child</li>
		</ul>
	</li>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<blockquote>
<pre>DOMElement* create_text_element(DOMDocument* doc,
   string name, string text)
{
   XMLCh* xtext = new XMLCh[text.length() + 1];
   XMLString::transcode(text.c_str(), xtext, text.length());
   DOMText* textNode = doc-&gt;createTextNode(xtext);
   delete xtext;
   DOMElement* r = create_element(doc, name);
   r-&gt;appendChild(textNode);
   return r;
}</pre>
	</blockquote>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<li>Build DOM tree to save to an XML document, for persistence or
		transmission</li>
	<li>Set attributes w/the <tt>setAttribute</tt> method</li>
	<li>E.g.,
		<blockquote><tt>
			XMLCh xname = . . .;<br>
			XMLCh xvalue = . . .;<br>
			price_element-&gt;setAttribute(xname, xvalue);
		</tt></blockquote>
	</li>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<li>Start with root, then add children</li>
	<li>Consider the simple tree:</li>

	<script><!--
		image( "fig08.png" )
	//--></script>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<li>First, create elements:
		<blockquote>
<pre>DOMElement* itemElement = create_element(doc, "item");
DOMElement* product_element = create_element(doc, "product");
DOMElement* description_element = create_text_element(doc,
   "description", "Ink Jet Refill Kit");
DOMElement* price_element = create_text_element(doc,
   "price", "29.95");
DOMElement* quantity_element
	 = create_text_element(doc, "quantity", "8");</pre>
		</blockquote>
	</li>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<li>Then add elements to the document:
		<blockquote>
<pre>doc-&gt;appendChild(item_element);
item_element-&gt;appendChild(product_element);
item_element-&gt;appendChild(quantity_element);
product_element-&gt;appendChild(description_element);
product_element-&gt;appendChild(price_element);</pre>
		</blockquote>
	</li>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<li>Organize these steps into helper functions</li>
	<li>Sample program (later) uses:
		<blockquote>
<pre>DOMElement* create_product(DOMDocument* doc, const Product&amp; p)
DOMElement* create_item(DOMDocument* doc, const Item&amp; anItem)
DOMElement* create_item_list(DOMDocument* doc,
	 const vector&lt;Item&gt;&amp; items)</pre>
		</blockquote>
	</li>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<li>Obtain <tt>DOMWriter</tt> object from the <tt>DOMImplementation</tt>
		object:
		<blockquote><tt>
			DOMWriter* writer = implementation-&gt;createDOMWriter();<br>
			writer-&gt;setFeature(XMLUni::fgDOMWRTFormatPrettyPrint, true);
		</tt></blockquote>
	</li>
	<li>"Pretty print" aligns elements</li>
	<li>Default is w/out whitespace</li>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents (cont.)</font></h2>
<font size="+1">

<ul>
	<li>A "format target" is needed</li>
	<li>To print to <tt>stdout</tt>:
		<blockquote><tt>
			XMLFormatTarget* out = new StdOutFormatTarget();<br>
			writer-&gt;writeNode(out, *doc);
		</tt></blockquote>
	</li>
	<li>To write to a local file:
		<blockquote><tt>
			XMLCh* filename = . . .;
			XMLFormatTarget* out = new LocalFileFormatTarget(filename);
			writer-&gt;writeNode(out, *doc);
		</tt></blockquote>
	</li>
	<li><tt>writeNode</tt> takes a <tt>DOMDocument</tt>, not a ptr</li>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents - Sample</font></h2>
<font size="+1">

<ul>
	<li>Sample program uses
		<a target='bigc' href='Code/builder/item.h'><tt>item.h</tt></a>,
		<a target='bigc' href='Code/builder/item.cpp'><tt>item.cpp</tt></a>,
		<a target='bigc' href='Code/builder/product.h'><tt>product.h</tt></a>, and
		<a target='bigc' href='Code/builder/product.cpp'><tt>product.cpp</tt></a>
		from (chptr 13)</li>
	<li>Added <tt>get_product</tt> and <tt>get_quantity</tt> to <tt>Item</tt></li>
</ul>

</font>

<hr><h2><font color="#009999">29.3 Creating XML Documents (<tt>builder.cpp</tt>)
	</font></h2>
<font size="+1">

<script><!-- 
	iframeWrapCode( "builder/builder.cpp", "90%", "80%" )
//--></script>

</font>

<hr><h2><font color="#009999">29.4 Document Type Definitions</font></h2>
<font size="+1">

<ul>
	<li>DTD - set of rules for correctly formed documents of a particular
		type</li>
	<li>E.g., consider the <tt>items</tt> document:
		<ul>
			<li><tt>items</tt> denotes a sequence of <tt>item</tt> elements</li>
			<li><tt>item</tt> contains a <tt>product</tt> and a <tt>quantity</tt></li>
			<li><tt>product</tt> contains a <tt>description</tt> and a
				<tt>price</tt></li>
			<li>Each of these elements contains descriptive text</li>
		</ul>
	</li>
	<li>DTD formalizes this description</li>
</ul>

</font>

<hr><h2><font color="#009999">29.4 Document Type Definitions (cont.)</font></h2>
<font size="+1">

<ul>
	<li>Sequence of rules that describe:
		<ul>
			<li>Legal attributes for each element type</li>
			<li>Legal child elements for each element type</li>
		</ul>
	</li>
	<li>Legal child elements are described by an ELEMENT rule:
		<blockquote><tt>
			&lt;!ELEMENT items (item*)&gt;
		</tt></blockquote>
		<tt>items</tt> must contain 0 or more <tt>item</tt> elements
	</li>
	<li>Delimited by <tt>&lt;!. . .&gt;</tt></li>
	<li>Contains element whose children are to be constrained</li>
	<li>Followed by description of children</li>
</ul>

</font>

<hr><h2><font color="#009999">29.4 Document Type Definitions (cont.)</font></h2>
<font size="+1">

<ul>
	<li>Definition of <tt>item</tt> node:
		<blockquote><tt>
			&lt;!ELEMENT item (product, quantity)&gt;
		</tt></blockquote>
	</li>
	<li>Children of <tt>item</tt> must be a <tt>product</tt> node, followed
		by a <tt>quantity</tt> node</li>
	<li><tt>product</tt> is similar:
		<blockquote><tt>
			&lt;!ELEMENT product (description, price)&gt;
		</tt></blockquote>
	</li>
</ul>

</font>

<hr><h2><font color="#009999">29.4 Document Type Definitions (cont.)</font></h2>
<font size="+1">

<ul>
	<li>The three remaining node types:
		<blockquote><tt>
			&lt;!ELEMENT quantity (#PCDATA)&gt;<br>
			&lt;!ELEMENT description (#PCDATA)&gt;<br>
			&lt;!ELEMENT price (#PCDATA)&gt;
		</tt></blockquote>
	</li>
	<li><tt>#PCDATA</tt> refers to text; <i>parsable character data</i></li>
</ul>

</font>

<hr><h2><font color="#009999">29.4 Document Type Definitions (cont.)</font></h2>
<font size="+1">

<ul>
	<li>Certain characters need to be encoded:
		<br><br>
		<table border='1'>
			<tr bgcolor='#00ffff'>
				<th>Character</th>
				<th>Encoding</th>
				<th>Name</th>
			</tr>
			<tr>
				<td align='center'>&lt;</td>
				<td align='center'>&amp;lt;</td>
				<td align='center'>Less than (left angle bracket)</td>
			</tr>
			<tr>
				<td align='center'>&gt;</td>
				<td align='center'>&amp;gt;</td>
				<td align='center'>Greater than (right angle bracket)</td>
			</tr>
			<tr>
				<td align='center'>&amp;</td>
				<td align='center'>&amp;amp;</td>