Parsing for double elements through xsl

[Linda]

Hi,

Sorry if this is the wrong group I'm posting to. We are using a CMS
which saves all files as .xml. We have an xslt file where we parse
all html tags. At the moment, the CMS sometimes puts in < br / <
br / to separate content instead of placing text in < p tags.

I'm trying to find any instance where a < br follows a < br >,
however for some reason I'm not getting the syntax quite correct. I'm
just starting out with xsl, so it could be a silly mistake, but I
can't find out a solution.

We use xsl v 1.0. The code that I've tried is:

<xsl:template match="br">

<xsl:choose>
<xsl:when test="preceding-sibling::br[1]"></xsl:when>
<xsl:otherwise> <xsl:copy-of select="."/></xsl:otherwise>
</xsl:choose>
</xsl:template>

I thought that the [1] is meant to get the immediately preceding
sibling. However it still finds instances where there is a single br.

Can someone help in regards to how to find an immediate preceding
sibling. An example of the code I'm using is:

Text< br />
Text< br />< br />
Text< br />< br />
Text< br />
Text.

It ignores the first < br instance in the first line (as expected),
but it then finds the last single < br instance in the 4th line.
Eventually I'm hoping to then move the text in < p tags (using
preceding-sibling::text()[1]/ following-sibling::text()[1]), but for
the moment I just want to single out the double br.

Regards,
Linda

[dnovatchev@gmail.com]

Re: Parsing for double elements through xsl

<xsl:when test="preceding-sibling::br[1]"></xsl:when>

This would select the nearest preceding-sibling element named "br", if
such exists.

You want to test if the nearest preceding-sibling element has the name
"br", and this is essentially different:

Use:

preceding-sibling::*[1][self::br]


Hope this helped.


Cheers,
Dimitre Novatchev



On Mar 16, 3:32 pm, Linda <madhatter....@ gmail.comwrote:

Hi,
>
Sorry if this is the wrong group I'm posting to.  We are using a CMS
which saves all files as .xml.  We have an xslt file where we parse
all html tags.  At the moment, the CMS sometimes puts in < br / <
br / to separate content instead of placing text in < p tags.
>
I'm trying to find any instance where a < br follows a < br >,
however for some reason I'm not getting the syntax quite correct.  I'm
just starting out with xsl, so it could be a silly mistake, but I
can't find out a solution.
>
We use xsl v 1.0.  The code that I've tried is:
>
<xsl:template match="br">
>
   <xsl:choose>
    <xsl:when test="preceding-sibling::br[1]"></xsl:when>
    <xsl:otherwise> <xsl:copy-of select="."/></xsl:otherwise>
  </xsl:choose>
</xsl:template>
>
I thought that the [1] is meant to get the immediately preceding
sibling.  However it still finds instances where there is a single br.
>
Can someone help in regards to how to find an immediate preceding
sibling.  An example of the code I'm using is:
>
Text< br />
Text< br />< br />
Text< br />< br />
Text< br />
Text.
>
It ignores the first < br instance in the first line (as expected),
but it then finds the last single < br instance in the 4th line.
Eventually I'm hoping to then move the text in < p tags (using
preceding-sibling::text()[1]/ following-sibling::text()[1]), but for
the moment I just want to single out the double br.
>
Regards,
Linda

[Linda]

Re: Parsing for double elements through xsl

Thanks for that, I gave that a try, however in the test, it is still
finding any br, not just the immediate br. Should I be putting this
test within a for-each? or should I be trying to find if the
preceding sibling is text instead?

On Mar 17, 1:04 pm, dnovatc...@gmai l.com wrote:

<xsl:when test="preceding-sibling::br[1]"></xsl:when>

>
This would select the nearest preceding-sibling element named "br", if
such exists.
>
You want to test if the nearest preceding-sibling element has the name
"br", and this is essentially different:
>
Use:
>
preceding-sibling::*[1][self::br]
>
Hope this helped.
>
Cheers,
Dimitre Novatchev
>

[Peter Flynn]

Re: Parsing for double elements through xsl

dnovatchev@gmai l.com wrote:

> <xsl:when test="preceding-sibling::br[1]"></xsl:when>

>
This would select the nearest preceding-sibling element named "br", if
such exists.
>
You want to test if the nearest preceding-sibling element has the name
"br", and this is essentially different:
>
Use:
>
preceding-sibling::*[1][self::br]

An alternative is preceding-sibling::*[1][local-name()='br']

However, when an XPath statement fails when expected to work (IMHE),
there is a namespace in effect somewhere :-) Check...

>I thought that the [1] is meant to get the immediately preceding
>sibling.

Yes, but it's already qualified by the ::br, so your orginal statement
matched the closest preceding element called br (no matter how much
earlier it occurred among the siblings), not the immediately preceding
element. Testing for ::*[1] always finds the immediately preceding
element; you then test for the name.

///Peter
--
XML FAQ: http://xml.silmaril.ie/

[dnovatchev@gmail.com]

Re: Parsing for double elements through xsl

Thanks for that, I gave that a try, however in the test, it is still

finding any br, not just the immediate br. Should I be putting this

Then it's highly likely you might have a default namespace defined and
in
scope.

Search for articles explaining this most FAQ for XPath expressions.

Cheers,
Dimitre Novatchev

"Linda" <madhatter.no1@ gmail.comwrote in message
news:993fdfa5-8024-4c51-96b9-
a28259bc382f@k1 3g2000hse.googl egroups.com...

Thanks for that, I gave that a try, however in the test, it is still
finding any br, not just the immediate br. Should I be putting this
test within a for-each? or should I be trying to find if the
preceding sibling is text instead?
>
On Mar 17, 1:04 pm, dnovatc...@gmai l.com wrote:

<xsl:when test="preceding-sibling::br[1]"></xsl:when>

>>
>This would select the nearest preceding-sibling element named "br", if
>such exists.
>>
>You want to test if the nearest preceding-sibling element has the name
>"br", and this is essentially different:
>>
>Use:
>>
> preceding-sibling::*[1][self::br]
>>
>Hope this helped.
>>
>Cheers,
>Dimitre Novatchev
>>

[Linda]

Re: Parsing for double elements through xsl

Thank you both for the explainations. I didn't think that it had been
defined elsewhere, but I'll let you know if I don't solve the
problem. I've at least got some good directions at what I can search
for now.

Linda

>
Then it's highly likely you might have a default namespace defined and
in
scope.
>
Search for articles explaining this most FAQ for XPath expressions.