Consider the schema:
<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema">
<xs:element name="TheList">
<xs:simpleType>
<xs:list itemType="xs:string" />
</xs:simpleType>
</xs:element>
</xs:schema>
And the xml:
<TheList>
This list has 5 values.
</TheList>
How can I iterate over each of the words in the list? To create something like:
<item>This</item>
<item>list</item>
<item>has</item>
<item>5</item>
<item>values.</item>
Based on the answers I've found here and here, I should do something like:
<xsl:stylesheet version="3.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml" version="1.0" encoding="UTF-8" indent="yes"/>
<xsl:template match="TheList">
<xsl:for-each select="tokenize(., ' ')">
<item><xsl:value-of select="." /></item>
</xsl:for-each>
</xsl:template>
</xsl:stylesheet>
However, at least in Altova's XML spy, I am getting this error:
Wrong occurrence to match required sequence type: The supplied sequence ('5' item(s)) has the wrong occurrence to match the sequence type xs:string ('zero or one')
Using the built in debugger, I have been able to determine that the error is thrown when calling tokenize on an element that has been declared as an xs:list. Which makes sense, since the element should already be split according to the rules regarding xs:list. To me, this suggests:
<xsl:stylesheet version="3.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml" version="1.0" encoding="UTF-8" indent="yes"/>
<xsl:template match="TheList">
<xsl:for-each select=".">
<item><xsl:value-of select="." /></item>
</xsl:for-each>
</xsl:template>
</xsl:stylesheet>
However, this treats the list as a single item and does not create a new item element for each word.
The for-each command seems to treat the xs:list as a single element while the tokenize function seems to treat the same xs:list as multiple elements. What am I missing?
If you're using a schema-aware transformation, then you don't need to tokenize the value yourself - the process of atomization does it for you automatically.
<xsl:template match="TheList">
<xsl:for-each select="data(.)">
<item><xsl:value-of select="." /></item>
</xsl:for-each>
</xsl:template>
If you want the code to work in both schema-aware and non-schema-aware environments you can write
<xsl:template match="TheList">
<xsl:for-each select="tokenize(string(.), ' ')">
<item><xsl:value-of select="." /></item>
</xsl:for-each>
</xsl:template>
Related
I'm trying to create a standard-use XSLT that will perform a given task based upon a user-provided XPATH expression as an XSLT parameter.
That is, I need something like this:
<xsl:template match="$paramContainingXPATH">
<!-- perform the task on the node(s) in the given xpath -->
</xsl:template>
For example, suppose I have some XML:
<xml>
<nodeA>whatever</nodeA>
<nodeB>whatever</nodeB>
<nodeC>whatever</nodeC>
<nodeD>whatever</nodeD>
<nodeE>whatever</nodeE>
</xml>
The XSLT needs to transform just a node or nodes matching a provided XPATH expression. So, if the xslt parameter is "/xml/nodeC", it processes nodeC. If the xslt parameter is "*[local-name() = 'nodeC' or local-name() = 'nodeE']", it processes nodeC and nodeE.
This should work for absolutely any XML message. That is, the XSLT cannot have any direct knowledge of the content of the XML. So, it could be a raw XML, or a SOAP Envelope.
I was guessing I might need to grab all the nodes matching the xpath, and then looping over them calling a named template, and using the standard identity template for all other nodes.
All advice is appreciated.
If you really need that feature with XSLT 1.0 or 2.0 then I think you should consider writing one stylesheet that takes that string parameter with the XPath expression and then simply generates the code of a second stylesheet where the XPath expression is used as a match pattern and the other needed templates like the identity template are included statically. Dynamic XPath evaluation is only available in XSLT 3.0 or in earlier versions as a proprietary extension mechanism.
You cannot match a template using a parameter - but you can traverse the tree and compare the path of each node with the given path. Here's a simple example:
XSLT 1.0
<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml" version="1.0" encoding="UTF-8" indent="yes"/>
<xsl:param name="path" select="'/world/America/USA/California'"/>
<xsl:template match="/">
<root>
<xsl:apply-templates select="*"/>
</root>
</xsl:template>
<xsl:template match="*">
<xsl:variable name="path-to-me">
<xsl:for-each select="ancestor-or-self::node()">
<xsl:value-of select="name()" />
<xsl:if test="position()!=last()">
<xsl:text>/</xsl:text>
</xsl:if>
</xsl:for-each>
</xsl:variable>
<xsl:if test="$path=$path-to-me">
<xsl:call-template name="action"/>
</xsl:if>
<xsl:apply-templates select="*"/>
</xsl:template>
<xsl:template name="action">
<return>
<xsl:value-of select="." />
</return>
</xsl:template>
</xsl:stylesheet>
Applied to a slightly more ambitious test input of:
<world>
<Europe>
<Germany>1</Germany>
<France>2</France>
<Italy>3</Italy>
</Europe>
<America>
<USA>
<NewYork>4</NewYork>
<California>5</California>
</USA>
<Canada>6</Canada>
</America>
</world>
the result will be:
<?xml version="1.0" encoding="UTF-8"?>
<root>
<return>5</return>
</root>
This could be made more efficient by passing the accumulated path as a parameter of the recursive template, so that each node needs only to add its own name to the chain.
Note:
The given path must be absolute;
Predicates (including positional predicates) and attributes are not implemented in this. They probably could be, with a bit more effort;
Namespaces are ignored (I don't see how you could pass an XPath as a parameter and include namespaces anyway).
If your processor supports an evaluate() extension function, you could forgo the calculated text path and test for intersection instead.
Edit:
Here's an example using EXSLT dyn:evaluate() and set:intersection():
<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:dyn="http://exslt.org/dynamic"
xmlns:set="http://exslt.org/sets"
extension-element-prefixes="dyn set">
<xsl:output method="xml" version="1.0" encoding="UTF-8" indent="yes"/>
<xsl:param name="path" select="'/world/America/USA/California'"/>
<xsl:variable name="path-set" select="dyn:evaluate($path)" />
<xsl:template match="/">
<root>
<xsl:apply-templates select="*"/>
</root>
</xsl:template>
<xsl:template match="*">
<xsl:if test="set:intersection(. , $path-set)">
<xsl:call-template name="action"/>
</xsl:if>
<xsl:apply-templates select="*"/>
</xsl:template>
<xsl:template name="action">
<return>
<xsl:value-of select="." />
</return>
</xsl:template>
</xsl:stylesheet>
Note that this will also work with with paths like:
/world/America/USA/*[2]
//California
and many others that the text comparison method could not accommodate.
I'm sending the element name as a param to the XSLT
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:xs="http://www.w3.org/2001/XMLSchema" version="2.0">
<xsl:output method="xml"/>
<xsl:param name="user"/>
<xsl:template match="/">
<xsl:call-template name="generic" />
</xsl:template>
<xsl:template name="generic">
<count><xsl:value-of select="count(.//*[local-name()=$user])"/></count>
</xsl:template>
</xsl:stylesheet>
I hope this could help!
Considering this XML,
XML:
<?xml version="1.0" encoding="UTF-8"?>
<items>
<book>
<title>doublebell</title>
<count>available</count>
</book>
<phone>
<brand>nokia</brand>
<model></model>
</phone>
</items>
Mapping Criteria while writing XSLT:
show the newbook/newtitle only if a value is present in input.
show the newbook/newcount only if a value is present in input.
show the newphone/newbrand only if a value is present in input.
show the newphone/newmodel only if a value is present in input.
XSLT:
<?xml version="1.0" encoding="ISO-8859-1"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml" version="1.0" encoding="UTF-8"
indent="yes" />
<xsl:variable name="book" select="items/book" />
<xsl:variable name="phone" select="items/phone" />
<xsl:template match="/">
<items>
<newbook>
<xsl:if test="$book/title!=''">
<newtitle>
<xsl:value-of select="$book/title" />
</newtitle>
</xsl:if>
<xsl:if test="$book/count!=''">
<newcount>
<xsl:value-of select="$book/count" />
</newcount>
</xsl:if>
</newbook>
<xsl:if test="$phone/brand!='' or $phone/model!=''"> <!-- not sure if this condition is required for the above mapping criteria -->
<newphone>
<xsl:if test="$phone/brand!=''">
<newbrand>
<xsl:value-of select="$phone/brand" />
</newbrand>
</xsl:if>
<xsl:if test="$phone/model!=''">
<newmodel>
<xsl:value-of select="$phone/model" />
</newmodel>
</xsl:if>
</newphone>
</xsl:if>
</items>
</xsl:template>
</xsl:stylesheet>
This is my concern:- In my actual XSLT, I have almost 70 conditions like
this, and everytime the XPath search is made twice [or thrice.. ] for
each condition [ for eg: <xsl:if test="$phone/brand!=''"> and <xsl:value-of select="$phone/brand" /> and outer if condition].
Is this much performance overhead? I don't feel it when I ran my application.
I like to hear from experienced people if this is correct way of writing the XSLT. Do I need to save the path in a variable and reuse it as done for $book
and $phone ? In such a case there will be 70+variables just to hold this.
You can approach this quite differently using templates. If you define a template that matches any element whose content is empty and does nothing:
<xsl:template match="*[. = '']" />
or possibly use normalize-space() if you want to consider elements to be empty if they contain only whitespace
<xsl:template match="*[not(normalize-space())]" />
Now with this in place add templates for the elements you are interested in
<xsl:template match="book">
<newbook><xsl:apply-templates /></newbook>
</xsl:template>
<xsl:template match="title">
<newtitle><xsl:apply-templates /></newtitle>
</xsl:template>
and so on. Now the book template will create a newbook element and go on to process its children. When it gets to the title it will have two different templates to choose from and will pick the "most specific" match. If the title is empty then the *[. = ''] template will win and nothing will be output, only if the title is non-empty will it create a newtitle element.
This way you let the template matcher do most of the work for you, you don't need any explicit conditional checks using xsl:if.
<?xml version="1.0" encoding="ISO-8859-1"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml" version="1.0" encoding="UTF-8"
indent="yes" />
<xsl:template match="/">
<items><xsl:apply-templates select="items/*" /></items>
</xsl:template>
<!-- ignore empty elements -->
<xsl:template match="*[not(normalize-space())]" />
<xsl:template match="book">
<newbook><xsl:apply-templates /></newbook>
</xsl:template>
<xsl:template match="title">
<newtitle><xsl:apply-templates /></newtitle>
</xsl:template>
<!-- and so on with similar templates for the other elements -->
</xsl:stylesheet>
Building on Ian's answer, you can also make a generic template that will create the "new" elements for you without having to specify each one individually. That would look like the below:
<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml" indent="yes" />
<xsl:template match="/">
<items><xsl:apply-templates select="items/*" /></items>
</xsl:template>
<xsl:template match="*[not(normalize-space())]" />
<xsl:template match="*">
<xsl:element name="{concat('new',name())}">
<xsl:apply-templates/>
</xsl:element>
</xsl:template>
</xsl:stylesheet>
That last template just rebuilds the element by concatenating the word "new" to the front of it.
I must be missing some fundamental concept of processing an XML document. Here is my source XML:
<?xml version="1.0" encoding="ISO-8859-1"?>
<Root>
<Element>visitorNameAlt</Element>
<Element>visitorScore</Element>
<Element>visitorTimeouts</Element>
<Element>Blank</Element>
<Element>homeNameAlt</Element>
<Element>homeScore</Element>
<Element>homeTimeouts</Element>
<Element>Blank</Element>
<Element>period</Element>
<Element>optionalText</Element>
<Element>flag</Element>
<Element>Blank</Element>
<Element>scoreLogo</Element>
<Element>sponsorLogo</Element>
</Root>
And my XSL stylesheet:
<?xml version="1.0" encoding="ISO-8859-1"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml" version="1.0" indent="yes"/>
<xsl:template match="/">
<xsl:for-each select="/Root">
<xsl:value-of select="position()"/>
<xsl:value-of select="Element"/>
</xsl:for-each>
</xsl:template>
</xsl:stylesheet>
All I want is to pluck the "Element" names from the source XML doc with their relative position in front.
My output is just "1" followed by the first element and nothing more.
I am new to XSLT, but have processed other documents successfully with for-each.
Thanks in advance.
Bill
You're looping over Root tags, not Element tags. Try this:
<xsl:template match="/">
<xsl:for-each select="/Root/Element">
<xsl:value-of select="position()"/>
<xsl:value-of select="."/>
</xsl:for-each>
</xsl:template>
Note that you must change the second value-of select to "." or "text()".
XSLT is not an imperative programming language. The XSLT processor grabs each element in turn and tries to match it to your stylesheet. The idiomatic way to write this is without a for-each:
<xsl:template match="/Root">
<xsl:apply-templates select="Element"/>
</xsl:template>
<xsl:template match="Element">
<xsl:value-of select="position()"/>
<xsl:value-of select="."/>
</xsl:template>
The first template matches the root and tells the processor to apply the stylesheet to all the Element nodes inside the Root. The second template matches those nodes, and outputs the desired information.
I have the following code (eg):
<response>
<parameter>
<cottage>
<cot>
<res>
<hab desc="Lakeside">
<reg cod="OB" prr="600.84>
<lwz>TR#2#AB#200.26#0#QB#OK#20120829#20120830#EU#3-0#</lwz>
<lwz>TR#2#AB#200.26#0#QB#OK#20120830#20120831#EU#3-0#</lwz>
<lwz>TR#2#AB#200.26#0#QB#OK#20120831#20120901#EU#3-0#</lwz>
I need to create a concatenated string that includes the whole of the first 'lwz' line and then the price (200.26, but it can be different in each line) for each corresponding line.
So the output, separating each line with | would be:
TR#2#AB#200.26#0#QB#OK#20120829#20120830#EU#3-0#|200.26|200.26
Thanks
This XSLT 1.0 transformation:
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="text"/>
<xsl:strip-space elements="*"/>
<xsl:template match="lwz[1]">
<xsl:value-of select="."/>
</xsl:template>
<xsl:template match="lwz[position() >1]">
<xsl:value-of select=
"concat('
',
substring-before(substring-after(substring-after(substring-after(.,'#'),'#'),'#'),'#')
)
"/>
</xsl:template>
<xsl:template match="text()"/>
</xsl:stylesheet>
when applied on the provided text (converted to a well-formed XML document !!!):
<response>
<parameter>
<cottage>
<cot>
<res>
<hab desc="Lakeside">
<reg cod="OB" prr="600.84">
<lwz>TR#2#AB#200.26#0#QB#OK#20120829#20120830#EU#3-0#</lwz>
<lwz>TR#2#AB#200.26#0#QB#OK#20120830#20120831#EU#3-0#</lwz>
<lwz>TR#2#AB#200.26#0#QB#OK#20120831#20120901#EU#3-0#</lwz>
</reg>
</hab>
</res>
</cot>
</cottage>
</parameter>
</response>
produces the wanted, correct result:
TR#2#AB#200.26#0#QB#OK#20120829#20120830#EU#3-0#
200.26
200.26
II XSLT 2.0 solution:
This transformation:
<xsl:stylesheet version="2.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="text"/>
<xsl:strip-space elements="*"/>
<xsl:template match="lwz[1]">
<xsl:value-of select="."/>
</xsl:template>
<xsl:template match="lwz[position() >1]">
<xsl:value-of select=
"concat('
', tokenize(.,'#')[4])"/>
</xsl:template>
<xsl:template match="text()"/>
</xsl:stylesheet>
when applied on the above XML document, again produces the wanted, correct result. Note the use of the standard XPath 2.0 function tokenize():
TR#2#AB#200.26#0#QB#OK#20120829#20120830#EU#3-0#
200.26
200.26
You can use the XPath substring function to select substrings from your lwz node data. You don't really give much more detail about your problem, if you want a more detailed answer, perhaps provide the full XML document and your best-guess XSLT
The transformation I am writing must compose a comma separated string value from a given node set. The resulting string must be sorted according to a random (non-alphabetic) mapping for the first character in the input values.
I came up with this:
<?xml version="1.0" encoding="utf-8"?>
<xsl:stylesheet
version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:tmp="http://tempuri.org"
exclude-result-prefixes="tmp"
>
<xsl:output method="xml" indent="yes"/>
<tmp:sorting-criterion>
<code value="A">5</code>
<code value="B">1</code>
<code value="C">3</code>
</tmp:sorting-criterion>
<xsl:template match="/InputValueParentNode">
<xsl:element name="OutputValues">
<xsl:for-each select="InputValue">
<xsl:sort select="document('')/*/tmp:sorting-criterion/code[#value=substring(.,1,1)]" data-type="number"/>
<xsl:value-of select="normalize-space(.)"/>
<xsl:if test="position() != last()">
<xsl:text>,</xsl:text>
</xsl:if>
</xsl:for-each>
</xsl:element>
</xsl:template>
</xsl:stylesheet>
It doesn't work and looks like the XPath document('')/*/tmp:sorting-criterion/code[#value=substring(.,1,1)] does not evaluate as I expect. I've checked to substitute the substring(.,1,1) for a literal and it evaluates to the proper value.
So, am I missing something that makes the sorting XPath expression not to evaluate as I expect or is it simply impossile to do it this way?
If not possible to create a XPath expression that works, is there a work around to achieve my purpose?
Note: I'm constrained to XSLT-1.0
Sample Input:
<?xml version="1.0" encoding="utf-8"?>
<InputValueParentNode>
<InputValue>A input value</InputValue>
<InputValue>B input value</InputValue>
<InputValue>C input value</InputValue>
</InputValueParentNode>
Expected ouput:
<?xml version="1.0" encoding="utf-8"?>
<OutputValues>B input value,C input value,A input value</OutputValues>
Replace the self::node() abbreviation ., with current() function.
A better predicate would be: starts-with(normalize-space(current()),#value)
Besides changing transformation according to Alejandro´s answer, I found it better to use a XSL variable for th mapping data to avoid declaration of a dummy namespace (tmp) as seen in Dimitre´s answer to another related question.
My final implementation:
<?xml version="1.0" encoding="utf-8"?>
<xsl:stylesheet
version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>
<xsl:output method="xml" indent="yes"/>
<xsl:template match="/InputValueParentNode">
<xsl:variable name="sorting-map">
<i code="A" priority="5"/>
<i code="B" priority="1"/>
<i code="C" priority="3"/>
</xsl:variable>
<xsl:variable name="sorting-criterion" select="document('')//xsl:variable[#name='sorting-map']/*"/>
<xsl:element name="OutputValues">
<xsl:for-each select="InputValue">
<xsl:sort select="$sorting-criterion[#code=substring(normalize-space(current()),1,1)]/#priority" data-type="number"/>
<xsl:value-of select="normalize-space(current())"/>
<xsl:if test="position() != last()">
<xsl:text>,</xsl:text>
</xsl:if>
</xsl:for-each>
</xsl:element>
</xsl:template>
</xsl:stylesheet>