XML SQL = future?
The appearance of XML standards and documents is a relational database, and the combination of SQL and XML-based SQL and XML is a neutralization path for accessing two-dimensional table data. So, the long SQL / XML binding trip started. With the advent of new XML documentation specifications, manufacturers are increasing the intensity of XML support in RDBMS (relational database management system). When XML is launched five years ago, the prospects of its rewriting data management rules have attracted the attention of relational database vendors, but they have no panic. During ten years ago, I have experienced this scene, when the object database was given the role of the example change. This new software specification does appear, and popularize the persistence concept: the ability to save and retrieve the programming language object without the need to convert the relational form. However, the result is that RDBMS learns a new "trick", which is to find how to save complex data types using SQL: 1999 object model. There is now a JDO (Java data object) application for relational databases and object databases. Microsoft said that the upcoming YUKON version SQL Server will be able to keep the .NET object. After absorbing the object, RDBMS manufacturers are now working hard to absorb XML documents. However, don't expect history to be simply repeated. We all know that most of the information of operating companies is stored in our creation and exchange documents, which are rarely saved in the enterprise database. Since XML can represent both we have seen and exposed documents (such as purchasing orders), you can represent information about these documents on the web service network, so our database can save and manage XML documents more critical than ever. . A real heavy bomb is in manufacturing, no one accurately knows what impact it will produce, but it can analyze it, and make some speculation. SQL / XML combined trip The first step in the first step is the first step in the first step in the first step, it is published as an XML format. XML publishing is a logical starting point because it can easily represent the SQL result collection in XML, because so many dynamic web pages are provided by SQL queries. The traditional method requires the program to build a web page with the program. The new method makes dynamic web pages in a fully published manner, using the XML representation of the SQL-to-XML query, and dissolves XML into the HTML using XSLT (Extensible Style Table Language Conversion). Initially, these virtual documents are created with proprietary SQL extensions. Now there is a new ISO / ANSI standard called SQL / XML, which defines a generic method. Currently, SQL / XML has been supported by Oracle and DB2. It defines an XML operator-oriented operator for this unit's XML data type in these products. SQL Server now does not support XML data types or SQL / XML extensions, Microsoft is scheduled for Yukon, which is launched in 2004, will support them. Most of the information stored in the store is saved in the storage document, not the relational database. The reason why these documents entered into the database always exist, that is, centralized management and full text search, but in the absence of a relationship method in which data in the document is established in the database, these reasons do not have persuasiveness . And XML provides people with argument. When the corporate document is mapped from the existing format to XML (this is a long distance that has just started), the relationship between two styles of data is possible. For example, there is an insurance application that saves claim data in a relational form and saving the claim document in XML format. Hybrid SQL / XML database allows this application to extract XML paragraphs from the document subset. This subset can be created by combining XML elements in the document with column values in the relationship table. Using some different types of storage and query strategies, it has achieved great progress. In storage, there are two generic methods. One is to enter the entire document into the column of the database, or you can put the document "tear" and put it in multiple relational forms. The latter method takes full advantage of the database's query engine and powerful update feature, but from the irregular XML data to SQL mapping than mapping from SQL to XML.
If your XML document is controlled by the XML mode, the map will be helpful. The XML mode is described for the XML-to-SQL mapper to provide a clue, and the user can add comments to such descriptions to more accurately control mapping. If the database supports how to receive an irregular XML data object, it will help the map. Oracle expands relational database technology to include objects as part of SQL: 1999. It is already mature on its 8i and 9i, which is the type system that can represent the XML mode in the object / relationship type. The query strategy XPath is the basis for all XML query strategies, which is a syntax for generating a tree structure and deleting branches down. When a XSLT style table converts an XML document, it uses XPath to isolate the segmentation of the document. A relational database that supports XML queries (including old products Oracle, DB2, and SQL Server, New Army like OpenLink Software, but not including mysql) use XPath in the same way. At first, this support for XPath is provided in a proprietary extension. Recently, the SQL / XML standard defines a general collection of SQL extensions with XPath consciousness. XPath also applies in the XQuery standard that W3C is about to be released. The XQuery standard is committed to adapting the SQL data connection function to the world of semi-structured XML data. IBM said that it is working hard to develop XQuery to handle XML content in a manner that SQL developers can be familiar. Although the manufacturer is in a hurry to wait for the ultimate completion of XQuery 1.0, their XQuery app will be powerful as the current SQL / XML application in some respects. The most obvious is that XQuery does not define the syntax of the elements used to update the elements in the XML document. Although SQL / XML update mechanism has not been approved, it has been defined and has been applied in Oracle and DB2. SQL / XML takes away the wind of XQuery? In short, XQuery looks only a alternative to completing the work that can be done with SQL and XPath. However, from a long time, developers may wish to keep XML abstraction on all of their data sources. In this case, as a rich and comprehensive programming language developed to handle complex data, XQuery may become an important example. The future of the document allows us to assume a moment of 2005, there is a purchase order that is flowing through the business process. This is an XML document created with tools like InfoPath, which is mixed on top of the core data and context data. The core data including the product number and department code will be entered into a column of a relational table. The context data that may be included by the requestor, the reviewer, and approvers will still be saved in the form of documentation. Currently, this context content has never been saved in RDBMS. It is important to understand how the data arrives there and its meaning. Once fill in, this purchase order is entered into a workflow that flows on a Web service network. Security services can perform authorization policies by updating the SOAP header, and the searches can search for document collection with the SOAP header with the same associated ID. The intermediate stage of these activities will require some database technology to manage XMLs that transparently appear in the query, but this may not be Oracle or DB2 tasks. At this time, a special XML database, such as the Tamino or Sleepycat Software of Software AG, may be more suitable for completing this task. They are very fast, which can be used very well for dynamic XML documents, even when these documents lack the RDBMS SQL / XML mapping mode. In the workflow process and after completion, the document will be accessible to the parties through a URL.
