Re: BUG?? - please help (JavaScript inside CDATA)

Hi,

On 30 May 2000, at 14:01, Dan Morrison wrote:

> juggy@xxxxxxx wrote:
> > 
> > Hi there,
> > 
> > I just discovered something that appears to be a nasty bug. I am
> > using the MSXMLDOM in a ASP to convert XML + XSl to HTML.
> 
> First, thanks for reminding me what an issue it can be learning a
> third language by way of a second!
> 
> I've noticed the high number of German (and Italian, and...)
> contributors to this group and XML resources in general before.
> Sometimes I've wondered how hard this can be, but decided that it's
> the syntax of the code in question that we're both looking at, and the
> language shouldn't be an issue.
> 
> Your code showed that there's a LOT of context in the variable names
> (and content) used in examples which is very important. By this I mean
> I had to think a lot harder about what you were doing than usual.
> Luckily my German is still good enough to hold up a conversation
> (although I can't spell)...
Sorry about that, but if I start changing my complete file to English 
before I send it, I wouldn't ever get finished. But I'll try to give 
shorter examples and put them down in English.
> 
> 
> SO.. Language issues aside, what you're trying looks OK. 
> We assume you're using the updated MSXML. (RIGHT?!)
> 
> My suspicions would fall upon the & characters that appear once or
> twice in your CDATA. I know CDATA is supposed to be safe from parsing,
> but in many implimentations, it's NOT.
> 
> The attitude I've seen here is that "&" IS, to all intents and
> purposes the same thing as & and (the numerical version which I
> can never remember). This means that parsers are a bit too keen to
> escape it for internal representation. As a keen client-side scripter,
> I disagree, but just work around it. Often I write code that writes
> code that writes code (jumping through three languages on the way). I
> prefer to do my own escaping. There are some output-method settings
> which I believe help.
> 
> SO, see if you get better results if your content strings have no &
> characters in them.
Nope, it doesn't make a difference.
> Also try
>   <xsl:script language="JavaScript">
>     <xsl:comment>
>       <![CDATA[

>         ... ..
Doesn't work: "Object expected"??. 
> As it's 
>   A: more correct. 
>   B: safer from the parser.
s.above.
> If not, what was the view-source output when you did it client-side?
It's perfectly correct - it just has the wrong strings at the wrong 
places.
> Paste that page into a syntax-coulouring editor and it may throw some
> light on things.
> 
> .dan.
> 

Thanks dan, but I haven't got further. I believe that there is 
someting wrong with the MSXML. Maybe find a more up-to-date 
version, though I'm VERY keen on running updates.


Juggy
 
> :=====================:====================:
> : Dan Morrison        : The Web Limited    :
> :  http://here.is/dan :  http://web.co.nz  :
> :  dman@xxxxxxxx      :  danm@xxxxxxxxx    :
> :  04 384 1472        :  04 495 8250       :
> :  025 207 1140       :                    :
> :.....................:....................:
> : If ignorance is bliss, why aren't more people happy?
> :.........................................:
> 
> 
>  XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list
> 



 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list