void FromXmlStream(const LPBYTE lpData, DWORD dwLen)
void FromXmlStream(const LPBYTE lpData, DWORD dwLen, const CXmlSerializationContext& context)
void FromXmlStream(const CBinaryData& data)
void FromXmlStream(const CBinaryData& data, const CXmlSerializationContext& context)
  Property Description  
    Method Name FromXMLStream  
    Argument - lpData A pointer to a buffer containing the XML data be de-serialized. The data can be in any valid (and supported) encoding, UTF-8, UNICODE etc. This can be an Xml document or Fast Infoset document and can be a compressed gzip or zlib file  
    Argument - dwLen The size of the data, this is the number of bytes pointed to by lpData  
    Argument - data The XML data to be de-serialized (See CBinaryData for more information)  
    Argument - context

The CXmlSerializationContext object controls the way in which XML is serialized/de-serialized. Its main role is to control the way in which validation is performed and which namespaces are output.
If this is not specified, the a default (CXmlSerializationContext::Default global static) instance of the class is used. If you are using several libraries generated from different schemas, or you want to change the way validation is performed for during the lifetime of the application or you are writing multithreaded code, then you should consider creating your own instance(s) of the CXmlSerializationContext.

Note: If you are writing a multithreaded app it is highly recommended that you use a different instance of this class on each thread, as access to the static instance is not synchronized. Although read only operations to the static instance (CXmlSerializationContext::Default) of the class are thread safe, if the global instance CXmlSerializationContext::Default is modified, then this could potentially cause threading problems.

 
    Description De-Serializes an XML file into the current object.  
    Remarks

This method will only load XML that complies with the XSD schema.
It will raise an exception if the XML is invalid.

This method provides a way to treat your XML as if it were binary, there are a number of gotchas when manipulating encoded (ie UTF-8) data, if you treat it to much like a string (especially attempting to convert it to UNICODE as if it where a n ASCII string) then you can damage it (see below & Multi-Language Support). This method simply allows you to treat it as a chunk of binary data, avoiding any potential encoding issues.

The root element in the XML held within lpszXMLIn must correspond to the class which this method is being called on. So you have a simple schema.

<?xml version="1.0" encoding="UTF-8"?>
<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"
           elementFormDefault="qualified"
           attributeFormDefault="unqualified">
  <xs:element name="Person">
    <xs:complexType>
      <xs:sequence>
        <xs:element name="name" type="xs:string"/>
        <xs:element name="Address">
          <xs:complexType>
            <xs:sequence>
              <xs:element name="HouseNumber" type="xs:string"/>
              <xs:element name="PostCode" type="xs:string"/>
            </xs:sequence>
          </xs:complexType>
        </xs:element>
      </xs:sequence>
    </xs:complexType>
  </xs:element>
</xs:schema>


This would cause 2 classes to be generated CPerson & CAddress.
So if you had the XML file test.xml, which contained.

<?xml version="1.0" encoding="UTF-8"?>
<Person>
  <name>Joe Blogs</name>
  <Address>
    <HouseNumber>15</HouseNumber>
    <PostCode>LS5 9PQ</PostCode>
  </Address>
</Person>


You would call FromXML on a CPerson object ie

CPersonPtr spPerson = CPerson::CreateInstance();
spPerson->FromXMLFile("c:\\test.xml")
;

Character encoding.
The underlying parser is expat version 1.95.6, as supports the following character encodings

  • UTF-8
  • UTF-16
  • UNICODE

Note - If you are working with a UNICODE build and UTF-8 encoded data.
This issue does not relate directly to our product, but has come up a number of times and is worth noting. UTF-8 (or other similar) encoded data should not be held in a UNICODE string. Attempting to convert a UTF-8 encoded string to UNICODE can damage the data (UTF-8 uses all the values 0-255 to encode the data, some of this range is not valid in UNICODE, so attempting the conversion results in '?' being substituted for invalid chars).
If you are not familiar with char encoding's, and this is possibly going to be an issue, then we recommend that within your application you consider XML data to be binary in nature, and use the FromXmlStream method.

See Multi-Language Support and Global Functions for more information.