Re: [xsl] service marks and other special symbols

Subject: Re: [xsl] service marks and other special symbols
From: "António Mota" <amsmota@xxxxxxxxx>
Date: Mon, 6 Mar 2006 14:33:42 +0000
Maybe this reading helps... At least the author says it's "The
Absolute Minimum Every Software Developer Absolutely, Positively Must
Know About Unicode and Character Sets (No Excuses!)"...

http://www.joelonsoftware.com/articles/Unicode.html

On 03/03/06, Michael Kay <mike@xxxxxxxxxxxx> wrote:
>
> > I have a perplexing issue that I just can't figure out.  I
> > want to have my xsl document output the code for the service
> > mark symbol (&#8480;) but it looks like the xsl parser
> > replaces the code with a ? or square symbol in the html
> > stream.
>
> Let's try to get the terminology right first: an XSLT processor takes a
> source tree as input and produces a result tree as output. The XML parser
> converts the source XML into a source tree, and the XSLT serializer
converts
> the result tree into serialized XML or HTML. The HTML is then typically
> displayed on the screen by some software, such as an editor or a browser.
> The thing that's putting the ? or square symbol on your screen is this last
> component, the display software, which is about as far from the parser in
> this sequence of events as you could get.
>
> If the display software is failing to display the character correctly this
> could be because the character isn't present in the font, or it could be
> because it's misconfigured to think that the document is in a different
> encoding from its actual encoding.
>
> The serializer has almost certainly put the correct character in the
output,
> but you're having difficulty seeing it there because of the limitations of
> your display software.
>
> It can be helpful in such cases to force the serializer to output the
> character as a symbolic character reference, and the simplest way to do
this
> is to set <xsl:output encoding="us-ascii"/>. There's no need to use
> character maps or disable-output-escaping for this. A better solution,
> however, would be to work out what's actually going wrong downstream on the
> display side.
>
> Michael Kay
> http://www.saxonica.com/

Current Thread