struts-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Martin Cooper <mart...@apache.org>
Subject Re: xhtml javascript hiding methods
Date Sun, 19 Jan 2003 03:48:11 GMT


On Fri, 17 Jan 2003, David Graham wrote:

> Martin,
> You should realize that without this attribute the xhtml outputting feature
> of Struts will be *completely* useless if you use the validator.  The
> javascript will not work surrounded by a CDATA section.  An alternative is
> to hide the js in an html comment.  This attribute gives people the choice
> of how they want to hide the javascript.
>
> Which section of the spec does this violate?  I've only found compatibility
> suggestions.  As I understand it a -1 vote means the change can't take
> place.  Is this really your vote on this?

My original intent was indeed to veto the change. Since then, I have
re-read the spec and realised that the normative section does not mandate
CDATA, while the section that appears to mandate it is in fact informative
rather than normative.

However, I'm not quite ready to +1 the change as is. As Craig pointed out,
the use of comments to hide script in XHTML will cause it to be dropped
entirely if the page is processed using an XML processor. Further, any
browser that's going to be useful with XHTML will support scripts, so
there is no reason to enclose the scripts in comments. (Recall that the
original reason for using comments was to prevent the script from being
rendered as content by browsers that didn't know what that was.) Finally,
note that using the script directly, with neither comments nor CDATA works
fine with current browsers.

My proposal, therefore, is to leave the 'cdata' attribute, but to change
the tag behaviours slightly. If XHTML is being generated, then the 'cdata'
attribute determines whether or not script is wrapped in a CDATA block.
Ostensibly, this allows the user to specify when prevention of markup
processing within the script block is necessary. The default should be
'true', as it is currently. In addition, if XHTML is being generated, then
the 'htmlComment' attribute should be ignored, and script blocks never
wrapped in HTML comments.

With the above change, I will remove my -1 on the original changes.

--
Martin Cooper


>
> David
>
>
>
>
>
>
> >From: Martin Cooper <martinc@apache.org>
> >Reply-To: "Struts Developers List" <struts-dev@jakarta.apache.org>
> >To: Struts Developers List <struts-dev@jakarta.apache.org>
> >Subject: Re: xhtml javascript hiding methods
> >Date: Thu, 16 Jan 2003 20:51:27 -0800 (PST)
> >
> >
> >
> >On Thu, 16 Jan 2003, David Graham wrote:
> >
> > > Ok, what do you think about a boolean "cdata" attribute for the
> >javascript
> > > tag?  I think this will accomodate everyone's needs.
> >
> >You mean add a "violate the spec to keep some browsers happy" attribute?
> >-1.
> >
> >--
> >Martin Cooper
> >
> >
> > >
> > > What do you mean that no browsers implement xhtml correctly?  What are
> >they
> > > missing?  Opera writes its own pages in strict xhtml and their browser
> > > supports it.
> > >
> > > Dave
> > >
> > >
> > >
> > >
> > >
> > >
> > > >From: "Craig R. McClanahan" <craigmcc@apache.org>
> > > >Reply-To: "Struts Developers List" <struts-dev@jakarta.apache.org>
> > > >To: Struts Developers List <struts-dev@jakarta.apache.org>
> > > >Subject: Re: xhtml javascript hiding methods
> > > >Date: Wed, 15 Jan 2003 23:04:45 -0800 (PST)
> > > >
> > > >
> > > >
> > > >On Wed, 15 Jan 2003, David Graham wrote:
> > > >
> > > > > Date: Wed, 15 Jan 2003 23:35:27 -0700
> > > > > From: David Graham <dgraham1980@hotmail.com>
> > > > > Reply-To: Struts Developers List <struts-dev@jakarta.apache.org>
> > > > > To: struts-dev@jakarta.apache.org
> > > > > Subject: xhtml javascript hiding methods
> > > > >
> > > > > Well, here are the choices as I understand them:
> > > > >
> > > > > 1.  Use CDATA to hide the javascript and make it completely useless
> >in
> > > > > current browsers.
> > > > >
> > > > > 2.  Use a comment to hide the javascript which allows current
> >browsers
> > > >to
> > > > > work and xml parsers.
> > > > >
> > > > > The xhtml spec does suggest using CDATA but I don't see a reason
the
> > > >comment
> > > > > method won't work.
> > > > >
> > > >
> > > >If you are using XML-based technologies like XSLT to transform things
> >to
> > > >create your output pages, the "commented out" text inside a <script>
> > > >element is going to get dropped on the floor.
> > > >
> > > >What I also don't understand is why anybody is worried about generating
> > > >XHTML markup for the current generation of popular browsers, none of
> >which
> > > >implement it correctly ... but that's a different issue.
> > > >
> > > > > Dave
> > > >
> > > >Craig
> > > >
> > > >
> > > >--
> > > >To unsubscribe, e-mail:
> > > ><mailto:struts-dev-unsubscribe@jakarta.apache.org>
> > > >For additional commands, e-mail:
> > > ><mailto:struts-dev-help@jakarta.apache.org>
> > >
> > >
> > > _________________________________________________________________
> > > MSN 8: advanced junk mail protection and 2 months FREE*.
> > > http://join.msn.com/?page=features/junkmail
> > >
> > >
> > > --
> > > To unsubscribe, e-mail:
> ><mailto:struts-dev-unsubscribe@jakarta.apache.org>
> > > For additional commands, e-mail:
> ><mailto:struts-dev-help@jakarta.apache.org>
> > >
> > >
> >
> >
> >--
> >To unsubscribe, e-mail:
> ><mailto:struts-dev-unsubscribe@jakarta.apache.org>
> >For additional commands, e-mail:
> ><mailto:struts-dev-help@jakarta.apache.org>
>
>
> _________________________________________________________________
> MSN 8 with e-mail virus protection service: 2 months FREE*
> http://join.msn.com/?page=features/virus
>
>
> --
> To unsubscribe, e-mail:   <mailto:struts-dev-unsubscribe@jakarta.apache.org>
> For additional commands, e-mail: <mailto:struts-dev-help@jakarta.apache.org>
>
>


--
To unsubscribe, e-mail:   <mailto:struts-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:struts-dev-help@jakarta.apache.org>


Mime
View raw message