On Feb 6, 7:31=A0pm, Martin Honnen <mahotr...@yahoo.de> wrote:
> Aaron Davies wrote:
> > I'm new to XSLT, and thought I'd start off with (what I hoped was) a
> > fairly simple transform, but I can't make heads or tails of the
> > example code I've found around the web. I'd like to a transform that
> > copies input to output, except that elements of a specific type with
> > attributes with specific values (and all their content) are omitted
> > completely. An example would be processing XHTML and dropping all P
> > paragraphs with class=3D"foo". I assume this is fairly easily derivable
> > from an identity filter, but I'm not sure where to start. Any help
> > would be much appreciated.
>
> <xsl:stylesheet
> =A0 =A0xmlns:xsl=3D"http://www.w3.org/1999/XSL/Transform"
> =A0 =A0xmlns:xhtml=3D"http://www.w3.org/1999/xhtml"
> =A0 =A0version=3D"1.0">
>
> =A0 =A0<xsl:template match=3D"xhtml:p[@class =3D 'foo']"/>
>
> =A0 =A0<xsl:template match=3D"@* | node()">
> =A0 =A0 =A0<xsl:copy>
> =A0 =A0 =A0 =A0<xsl:apply-templates select=3D"@* | node()"/>
> =A0 =A0 =A0</xsl:copy>
> =A0 =A0</xsl:template>
>
> </xsl:stylesheet>

hmm. looks reasonable, but doesn't seem to work on the actual document
i'm trying to change.

i'm trying to hack up an rss feed, and the paragraphs i want to remove
are in the description section. actually, i'm not sure this is doable
with XSLT at all--on closer inspection, the description is a blob of
presumably opaque text, and what ends up as tags in the browser is
ampersand encoded in the actual feed.

should i give up and use sed? :)