Homework #5. Customize JSP Tag and XML/DTD

• Modify the customized JSP database tag <jspx:database> implementation in cs526/jsp/dbjsp to define a new <db:select> tag which takes three attributes: column, table, where, similar to that of the select SQL statement.
• For example, <db:select column="item, sales" table="jun" where="sales >2000"> will retrieve the corresponding record in the table called "jun" in cs526db.
• Note that the column and where attributes are optional, if they are missing, use * for column value (select all columns) and omit the where clause in the SQL statement.
• cp ~cs526/jsp/dbjsp to you home directory. Reuse as much the web page and source code in dbjsp as you like.
• Modify the db.jsp web page to include several <db:select> tags.
• Modify DataBase.java or create your own servlet to implement the new customized tag.
• Create a db-tag.tld which specifies the mapping between the new tag and the Java Servlet that realizes its semantics.
• Email me the url of the screendump with web access to tomcat web server with your db.jsp web page.

• Create a xml subdirectory under your public_html directory. Change its access right to allow web server access.
• Use grade.xml and grade.xsl in ~cs301/public_html/xml/example as an example.
  Use purchase.xml in hw3/4 as a template to specify the Document Type Data file, called it purchase.dtd, for checking the other purchase documents following the same document structure. Save purchase.xml and purchase.dtd in the xml directory.
  Add the following line as the third line in purchase.xml, right after the two processing instruction line.
  <!DOCTYPE purchase SYSTEM "purchase.dtd">
• Create an XSL file that take the purchase.xml and generate a HTML file with table indicating the purchase items and their related info.
• Email me the url to your purchase.xml with .xsl and .dtd.
  

Hint. Check validity of an XML document.

• cp -r ~cs301/public_html/xml/xml4j_2_0_6 to your xml directory.
• It contains the IBM's XML for Java parser (XML4J), high performance, modular XML parser written in 100% Java.
• I have include a shell script, checkxml.sh for checking if an xml document is conformed to the dtd specification it references. The web pages of the parser is in docs directory and available as http://cs.uccs.edu/~cs301/xml/xml4j_2_0_6/docs/
• The ValidatingSAXParser will check the xml file according to the referenced dtd file.
• The test subdirectory including a personalbad.xml which violates the data type definition specified in personal.dtd.
• When run with "./checkxml.sh personalbad.xml" you will get the following msg:
  [cs526@sanluis test]$ ./checkxml.sh personalbad.xml
  personalbad.xml:
  [Error] personalbad.xml:13:25: Attribute, "mananger", is not declared in element, "link".
  [Error] personalbad.xml:15:17: Element "<person>" contains incomplete content for the rule, "(name,email*,url*,link?)".
  <?xml version="1.0" encoding="UTF-8"?>
  <personnel>
  <person id="Big.Boss">
  <name><family>Boss</family> <given>Big</given></name>
  <email>chief@foo.com</email>
  <url href="http://www.foo.com/~chief"></url>
  <link subordinates="one.worker"></link>
  </person>
  <person id="one.worker">
  <email>worker1@foo.com</email>
  <name><family>Workman</family> <given>John</given></name>
  <link mananger="Big.Boss"></link>
  <url href="http://www.foo.com/~workman"></url>
  </person>
  </personnel>
  
  The personal.dtd contains the following:
  <?xml version="1.0" encoding="iso-8859-1"?>
  <!ELEMENT personnel (person)+>
  <!ELEMENT person (name,email*,url*,link?)>
  <!ATTLIST person id ID #REQUIRED>
  <!ELEMENT family (#PCDATA)>
  <!ELEMENT given (#PCDATA)>
  <!ELEMENT name (#PCDATA|family|given)*>
  <!ELEMENT email (#PCDATA)>
  <!ELEMENT url EMPTY>
  <!ATTLIST url href CDATA #REQUIRED>
  <!ELEMENT link EMPTY>
  <!ATTLIST link
  manager IDREF #IMPLIED
  subordinates IDREFS #IMPLIED>
  
  
  
• Just copy the corresponding purchase.xml and purchase.dtd files into the xml4j_2_0_6 directory. Then run "./checkxml.sh purchase.xml" to validate the xml documents.