"value substitution/population" in template XML using XPath

B

Bilal

Hello
I'm stuck on this problem for quite some time and hope somebody woul
be able to guide me

Basically, I need to populate a large number of "template" XML file
which have all elements/attributes etc. defined but the values in thes
elements/attributes might be blank or place holders, as two example
below

Example - File type
<?xml version="1.0" encoding="UTF-8"?
<fileType1
<element1
<element2 attr2="xxx"
<element3
<element4 attr4="yyy"
</element4
</element3
</element2
</element1
</fileType1

o

Example - File type
<?xml version="1.0" encoding="UTF-8"?
<fileType2
<element1
<element2 attr2="xxx"
<element13
<element14
<element15>ZZZ</element15
</element14
</element13
</element2
<element3
</element3
</element1
</fileType2

and the new values to be populated (or atleast their XPath) ar
indicated in a seperate XML data file as, and easily obtained a
xpath/value pair using Java or XSLT etc.

<?xml version="1.0" encoding="UTF-8"?
<newValues
<file type="fileType2"
<element
<x-path>/fileType2/element1/element3</x-path
<value>newValue3</value
</element
<element
<x-path>/fileType2/element1/element2/@attr2</x-path
<value>ABC</value
</element
<element
<x-path>/fileType2/element1/element2/element13/element14/element15</x-path
<value>newValue15</value
</element
</file
<file type="fileType1"
<element
<x-path>/fileType1/element1/element2/@attr2</x-path
<value>123</value
</element
<element
<x-path>/fileType1/element1/element2/element3/element4/@attr4</x-path
<value>999</value
</element
</file
</newValues


So what is the best approach to read the 'template' example file(s) an
output it with new values for xpath indicated in the data file

I have used DOM/JDOM to perform this task at a limited scale, but th
complicating factors are

- the number of unique "template" files is rather large (al
generated from relevent XSD
- the structure of each template is quite deep/comple
- the number of xpath/value would be arbitrary (say min of 5 "fields
to be changed, max all of them!

Essentially this task is a 'value substitution' or 'data population
problem for an XML file, where the xpath/value pair are indicated i
another XML file. IMHO (re)constructing the output XML from scratc
appears would be overkill, and not feasbile to do it programatically i
Java code etc. Hence my search for a more efficient way to do it, sa
with XSLT or some other technique, which 'processes' the original XM
input file WITHOUT altering the structure but substituting the desire
values as required.

Would appreciate any and all help in this regards

Regards

Bilal B
 
J

Joe Kesselman

Y'know, I'd consider (ab)using XSLT for this.

Start with the standard identity transform, to handle most of the document.

Then as the equivalent of your
<element>
<x-path>/fileType2/element1/element3</x-path>
<value>newValue3</value>
</element>

write a template something like:

<xsl:template match="/fileType2/element1/element3">
<xsl:copy>
<xsl:apply-templates select="@*"/>
<xsl:choose>
<xsl:when test="node()"> <!-- If not empty -->
<xsl:apply-templates select="node()"/>
<xsl:when>
<xsl:eek:therwise>newValue3</xsl:eek:therwise>
</xsl:choose>
</xsl:copy>
</xsl:template>

Obviously this could handle more complicated conditions/operations as well.

Note that if you don't want to write this code by hand it would be
straightforward to write a stylesheet which generates this stylesheet
from the <newValues> sketch you showed us.

XPath-match-and-transform is exactly what XSLT is designed for. This
isn't a typical application for it, but it's can do the job, and
probably beats hand-coding a new application for the purpose.



Having said all that: You said you're using schemas. I believe schemas
can specify default values. Is there a reason you aren't using that
mechanism rather than inventing a new one?
 
B

Bilal

Hi Joe,
Thanks for your reply; I'll explore your suggestions further to fully
understand them as my XSLT experience is, to be honest, between advanced
beginner & intermediate :)
Note that if you don't want to write this code by hand it
would be straightforward to write a stylesheet which
generates this stylesheet from the <newValues> sketch you
showed us.

Having said all that: You said you're using schemas. I
believe schemas can specify default values. Is there a
reason you aren't using that mechanism rather than
inventing a new one?

Yes, I do have the schema so if the task can be done by using the
schema, I would try that approach as well. Indeed default values for
elements/attributes can be specified in the schema as below:

<xs:element name="color" type="xs:string" default="red"/>

The values for elements/attributes as exampled in the <newValues>
sketch will differ from case to case, so unsure how that would be
useful? Wouldn't that necessitate changing the schema file to generate a
new instance xml? Also, how would one use the xpath to navigate the
schama inself? Pardon my ignorance if that is a something obvious but my
XSLT experience has been limited to plain XML files rather than XSD
files, though I recognize that the principle remains the same for the
two. So could you elaborate please?

Thanks again!

Regards,

Bilal B.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

Forum statistics

Threads
473,995
Messages
2,570,226
Members
46,815
Latest member
treekmostly22

Latest Threads

Top