Lecture Todays Lecture 28: XSLT
Kenneth M. Anderson Software Methods and Tools CSCI 3308 - Fall Semester, 2003 Introduce XSLT
background concepts examples
XSLT stands for XML Stylesheet Language, Transformations
December 1, 2003
University of Colorado, 2003
2
Transformations
XSLT was developed as part of the XML stylesheet standards effort Whats a stylesheet?
A stylesheet is a device for specifying presentation information independent of content For instance, in Microsoft Word, you can specify that a heading should appear in 36pt Times bold font with double spacing above and below
Then all headings will appear that way, no matter what the heading actually says
Stylesheets in HTML
The Web already has a stylesheet language called cascading stylesheets or CSS This mechanism allows formatting information to be associated with HTML tags, such as <H1> or <P> without using <font> or <b> tags In the last lecture, we asked the question, if CNN switched to using XML in their webpage, how would they associate formatting information with a tag such as <headline>?
December 1, 2003
University of Colorado, 2003
3
December 1, 2003
University of Colorado, 2003
4
XSL and XSLT
The answer, of course, is with the XML Stylesheet language (XSL) Problem: the XSL standards group was having a hard time agreeing on a mechanism for specifying formatting information However, they were making progress with the transformation part of the specification
This part specifies mechanisms for transforming XML to other structures, such as XML->XML, XML->HTML, XML>PDF, etc.
XSLT
Plus, they noticed that a large part of their users
were using the transformation part of the XSL specification to transform XML to HTML and then using CSS to format the resulting document
As such, XSLT was born, the standards group focused on finishing the transformation part of the specification first and published it as XSLT (XSL has since been completed)
December 1, 2003 University of Colorado, 2003 6
December 1, 2003
University of Colorado, 2003
5
Background
To understand XSLT, you must view XML documents as tree structures
XSLT provides rules to transform one tree into another tree It traverses the source tree in an order dictated by the stylesheet and creates the destination tree using the rules of the stylesheet
December 1, 2003 University of Colorado, 2003 7
Example of viewing XML as a tree
<!DOCTYPE gradebook [
<!ELEMENT gradebook (class, student*)> <!ELEMENT class (name, studentsEnrolled)> <!ATTLIST class semester CDATA #REQUIRED> <!ELEMENT name (#PCDATA)> <!ELEMENT studentsEnrolled (#PCDATA)> <!ELEMENT student (name, grade*)> <!ELEMENT grade (#PCDATA)> <!ATTLIST grade name CDATA #REQUIRED>
gradebook class student name student name grade grade grade
name
]>
grade grade studentsEnrolled
December 1, 2003
University of Colorado, 2003
8
Background: XPath
XSLT uses a separate standard, called XPath, to help select nodes in an XML document For instance
gradebook/student/grade is an XPath expression that selects all grade nodes in the example on the previous slide
More XPath examples
//grade
start at the root node and find all grade nodes
gradebook/student[2]
select the second student node under gradebook
For more information on XPath see
< http://www.w3.org/TR/xpath> You will need to know how to create simple XPath expressions (like the ones shown above) to complete lab 10
XPath can even select attributesfor example..
gradebook/student/grade[@name=hw3] will select only those grade nodes that have a value of HW3 for their name attribute
December 1, 2003 University of Colorado, 2003 9
December 1, 2003
University of Colorado, 2003
10
XSLT, the details
XSLT transforms XML documents using stylesheets that are themselves XML documents All XSLT stylesheets have the following form
<?xml version="1.0"?> <xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
templates and transformation rules go here
Stylesheets
Stylesheets consist of templates that match nodes of the source XML tree (i.e. document)
Each template then specifies what should be created in the destination tree (or document) A template looks like this:
<xsl:template match="/"> <html>
<head> <title>Grade Book</title> </head> <xsl:apply-templates/>
The tag is called xsl:template and it has an attribute called match that takes an XPath expression If a node matches this expression (in this case the root note) then the associated text appears in the destination document (except for the xsl:apply-templates part)
12
</xsl:stylesheet> You can use this template when writing your own XSL Stylesheet in Lab 10
</html> </xsl:template>
December 1, 2003
University of Colorado, 2003
11
December 1, 2003
University of Colorado, 2003
XSLT Architecture
HTML XML File XSLT Processor PDF XML XSLT XSLT XSLT Stylesheet Stylesheet Stylesheet same The XML File can be transformed into XML, HTML, PDF, etc. just by using a different stylesheet
University of Colorado, 2003 13
More Details
Stylesheet processing
XSLT processor is handed a document and a stylesheet It starts a (breadth-first) traversal at the root node and checks to see if there is a template match
If so, it applies the template and looks for an xsl:applytemplates element
If such an element exists, it continues the traversal if no such element exists, the traversal stops
If not, it traverses down the tree looking for a template match with some other node of the tree
December 1, 2003
December 1, 2003
University of Colorado, 2003
14
XSL:apply-templates
The apply-templates tag determines if an XSLT processor continues traversing a document once a template match has occurred The apply-templates tag can contain an attribute called select which can specify the specific children to continue traversing using an XPath expression
<xsl:apply-templates>
All children traversed
Processing in XSLT stylesheets
XSLT is very powerful
We cannot cover the entire standard So, the following slides cover only a small subset of the tags that can be placed in an XSLT stylesheet For a good reference on XSLT see:
<http://www.zvon.org/xxl/XSLTreference/Output/index.html>
<xsl:apply-templates select=grade[@name=HW4]>
All grade nodes with a name attribute equal to HW4 traversed (any other nodes skipped during the subsequent traversal)
December 1, 2003
University of Colorado, 2003
15
December 1, 2003
University of Colorado, 2003
16
Repetition
<xsl:for-each select = item>
Do something here ...
Repetition Example
<xsl:template match="/"> <html>
<head>
<title>Grade Book</title>
</xsl:for-each> Again, the select attribute is an XPath expression that selects the nodes to iterate over
</head> <body> <UL> <xsl:for-each select=student/grade>
<LI>Grade: <xsl:value-of select="./></LI>
</xsl:for-each> </UL> </body>
</html> </xsl:template>
December 1, 2003 University of Colorado, 2003 17 December 1, 2003 University of Colorado, 2003 18
Example Explained
This example creates a simple HTML file that contains a list of all the grades received by students in the gradebook
Note: It did not list student names for each set of grades but it could have easily done so. However the student/grade XPath expression in the for-each select attribute skipped past the student nodes and selected only grade nodes The value-of element pulled the value of the grade element (e.g. the grade) into the HTML file The resulting HTML file is shown on the next slide
December 1, 2003 University of Colorado, 2003 19
Generated HTML File
<html>
<head>
<title>Grade Book</title>
In the browser, this file would look like this: Grade Book
Grade: 10 Grade: 7 Grade: 6 Grade: 10
</head> <body> <UL>
<LI>Grade: 10<LI> <LI>...