Nested conditional if else statements in xpath

Question

I have this XML:

<property id="1011">
    <leasehold>No</leasehold>
    <freehold>Yes</freehold>
    <propertyTypes>
        <propertyType>RESIDENTIAL</propertyType>
    </propertyTypes>
</property>

and I want to create an xpath statement that is same as the following nested if-else pseudocode block.

if( propertyTypes/propertyType == 'RESIDENTIAL') {
    if( leasehold == 'Yes' ){
        return 'Rent'
    } else
        return 'Buy'
    }
} else {
    if( leasehold == 'Yes' ){
        return 'Leasehold'
    } else
        return 'Freehold'
    }
}

I've seen something about Becker's method but I couldn't really follow it. XPath isn't my strong point really.

Answer 1

I. In XPath 2.0 one simply translates this to:

   if(/*/propertyTypes/propertyType = 'RESIDENTIAL')
    then
     (if(/*/leasehold='Yes')
       then 'Rent'
       else 'Buy'
     )
     else
       if(/*/leasehold='Yes')
         then 'Leasehold'
         else 'Freehold'

XSLT 2.0 - based verification:

<xsl:stylesheet version="2.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
 xmlns:xs="http://www.w3.org/2001/XMLSchema">
 <xsl:output omit-xml-declaration="yes" indent="yes"/>

 <xsl:template match="/">
  <xsl:sequence select=
   "if(/*/propertyTypes/propertyType = 'RESIDENTIAL')
   then
     (if(/*/leasehold='Yes')
       then 'Rent'
       else 'Buy'
     )
     else
       if(/*/leasehold='Yes')
         then 'Leasehold'
         else 'Freehold'
   "/>
 </xsl:template>
</xsl:stylesheet>

When this transformation is applied on the provided XML document:

<property id="1011">
    <leasehold>No</leasehold>
    <freehold>Yes</freehold>
    <propertyTypes>
        <propertyType>RESIDENTIAL</propertyType>
    </propertyTypes>
</property>

the XPath expression is evaluated and the result of this evaluation is copied to the output:

Buy

---

II. XPath 1.0 solution

In XPath 1.0 there isn't an if operator.

A conditional statement can still be implemented with a single XPath 1.0 expression, but this is more tricky and the expression may not be too readable and understandable.

Here is a generic way (first proposed by Jeni Tennison) to produce $stringA when a condition $cond is true() and otherwise produce $stringB:

concat(substring($stringA, 1 div $cond), substring($stringB, 1 div not($cond)))

One of the main achivements of this formula is that it works for strings of any length and no lengths need to be specified.

Explanation:

Here we use the fact that by definition:

number(true()) = 1

and

number(false()) = 0

and that

1 div 0 = Infinity

So, if $cond is false, the first argument of concat() above is:

 substring($stringA, Infinity)

and this is the empty string, because $stringA has a finite length.

On the other side, if $cond is true() then the first argument of concat() above is:

sibstring($stringA, 1) 

that is just $stringA.

So, depending on the value of $cond only one of the two arguments of concat() above is a nonempty string (respectively $stringA or $stringB).

Applying this generic formula to the specific question, we can translate the first half of the big conditional expression into:

concat(
           substring('rent',
                      1 div boolean(/*[leasehold='Yes'
                                     and
                                       propertyTypes/propertyType = 'RESIDENTIAL'
                                      ]
                                  )
                      ),
           substring('buy',
                      1 div not(/*[leasehold='Yes'
                                     and
                                       propertyTypes/propertyType = 'RESIDENTIAL'
                                      ]
                                  )
                      )
               )

This should give you an idea how to translate the whole conditional expression into a single XPath 1.0 expression.

XSLT 1.0 - based verification:

<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
 <xsl:output omit-xml-declaration="yes" indent="yes"/>

 <xsl:template match="/">
  <xsl:copy-of select=
   "concat(
           substring('rent',
                      1 div boolean(/*[leasehold='Yes'
                                     and
                                       propertyTypes/propertyType = 'RESIDENTIAL'
                                      ]
                                  )
                      ),
           substring('buy',
                      1 div not(/*[leasehold='Yes'
                                     and
                                       propertyTypes/propertyType = 'RESIDENTIAL'
                                      ]
                                  )
                      )
               )
   "/>
 </xsl:template>
</xsl:stylesheet>

When this transformation is applied on the provided XML document (above), the XPath expression is evaluated and the result of this evaluation is copied to the output:

buy

Do note:

If you decide to replace the specific strings with other strings that have different lengths than the original, you simply replace these strings in the above XPath 1.0 expression and you don't have to worry about specifying any lengths.

Answer 2

Becker's method for your data is the following:

concat(substring('Rent',      1 div boolean(propertyTypes/propertyType ="RESIDENTIAL" and leasehold="Yes")),
       substring('Buy',       1 div boolean(propertyTypes/propertyType ="RESIDENTIAL" and leasehold="No")),
       substring('Leasehold', 1 div boolean(propertyTypes/propertyType!="RESIDENTIAL" and leasehold="Yes")),
       substring('Freehold',  1 div boolean(propertyTypes/propertyType!="RESIDENTIAL" and leasehold="No")))

Answer 3

Spent all day today, but works for me this is for Xpath 1.0:

concat(
substring(properties/property[@name="Headline"], 1, string-length(properties/property[@name="Headline"]) * 1), 
substring(properties/property[@name="Name"], 1, not(number(string-length(properties/property[@name="Headline"]))) * string-length(properties/property[@name="Name"]))
)