hi Greg,
you need a template that matches all text nodes, which tests for an occurance
of '%' and call the recursive named template if there is one:
<xsl:template match="text()">
<xsl:choose>
<xsl:when test="contains(.,'%')">
<xsl:call-template name="convertPercent">
<xsl:with-param name="str" select="."/>
</xsl:call-template>
</xsl:when>
<xsl:otherwise>
<xsl:value-of select="."/>
</xsl:otherwise>
</xsl:choose>
</xsl:template>
Then the named template recursively processes the string converting everything
after a % and before a space into a <field key=..."/> element. Its a good
technique to know as it comes up time and again in xslt 1.0.
<xsl:template name="convertPercent">
<xsl:param name="str" select="'error'"/>
<xsl:choose>
<xsl:when test="contains($str,'%')">
<xsl:value-of select="substring-before($str,'%')"/>
<field key="{substring-before(substring-after($str,'%'),' ')}"/>
<xsl:call-template name="convertPercent">
<xsl:with-param name="str"
select="substring-after(substring-after($str,'%'),' ')"/>
</xsl:call-template>
</xsl:when>
<xsl:otherwise>
<xsl:value-of select="."/>
</xsl:otherwise>
</xsl:choose>
</xsl:template>
You may have to modify the tests slightly to include the full stop in your
output after %name2, as currently that would get dropped as its before the
first occurance of a space, but it isnt so hard. If you get stuck post back,
or checkout Demtire's tokenize templates in fxsl, or I think exslt's tokenise
function can handle multiple delimiters.
cheers
andrew
Hi all,
My input document contains lines that have multiple 'tokens' that need
replacing with xml nodes. Tokens are identified by a percent sign. An example
of a input line with tokens looks like the following:
"Welcome %name1 %name2. You have received a bonus of %bonus points. Your
points expire on %expire365"
It would be converted to :
<line>Welcome <field key="name1"/> <field key="name2"/>. You have received a
bonus of <field key="bonus"/> points. Your points expire on <field
key="expiry" param="365"/></line>
Not being an XSL guru, I can't seem to find any way of successively performing
multiple operation on a string in this fashion. That is, replacing all tokens
of like, say, %name1, and then doing %name2 and so on.
With no 'variables' this seems a little awkward. Can anyone provide an example
or some hints on how I might replace a number of different tokens in a string?
Many thanks.
Greg.
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.467 / Virus Database: 266 - Release Date: 01/04/2003
