Regardless, given the statement that xhtml 2.0 will not be html compatible maybe there should be an xhtml taglib. I like the idea of having the xhtml taglib being the html taglib with the xhtml setting defaulted to true. That way, we can easily implement xhtml 1.0. Will JavaServer Faces handle xhtml? Should we even bother with doing this in Struts?
Dave
From: "Craig R. McClanahan" <[EMAIL PROTECTED]>
Reply-To: "Struts Developers List" <[EMAIL PROTECTED]>
To: Struts Developers List <[EMAIL PROTECTED]>
Subject: Re: HTML, XML, XHTML and <html:html>
Date: Tue, 22 Oct 2002 18:46:26 -0700 (PDT)
On Tue, 22 Oct 2002, David Graham wrote:
> Date: Tue, 22 Oct 2002 13:54:14 -0600
> From: David Graham <[EMAIL PROTECTED]>
> Reply-To: Struts Developers List <[EMAIL PROTECTED]>
> To: [EMAIL PROTECTED]
> Subject: Re: HTML, XML, XHTML and <html:html>
>
> If we're not doing xhtml in 1.1 then we'll need to change some things:
> 1. All boolean attributes need to go back to minimized form
> <select multiple> instead of <select multiple="mulitple">
> This is currently implemented incorrectly anyways because in xhtml it
> should be <select multiple="true">
>
The expanded form of boolean attributes is legal in HTML/4.01[1] (see
Section
3.3.4). When you use one:
Boolean attributes may legally take a single value:
the name of the attribute itself (eg. selected="selected").
The minimized form (selected) is *optional*. There's a warning about some
browsers not recognizing the non-minimized form, but Struts has been doing
this since version 0.5 with no complaints.
Further, according to XHTML/1.0[2], section 4.5, the format Struts renders
is also correct for XHTML/1.0. The example that the spec gives for
correct usage is:
<dl compact="compact">
and not:
<dl compact="true">
Therefore, -1 on changing this.
> 2. Remove the ending / in <input> tags.
>
+1 on removing the trailing "/" where we recently added it for XHTML
compatibility.
> Here's my view:
> - Forget xhtml in 1.1
> - In 1.2 add a global configuration parameter that tells all html tags to
> emit xhtml (default to off). I don't care where this global param. goes but
> I do care that it be global and not dependant on using <html:html> or
> setting it on a tag by tag basis. I think a separate xhtml library of
> struts tags is going overboard.
>
We'll obviously need to figure out how much we care about XHTML in future
versions, which version of XHTML we think we want to support, and so on.
That's a discussion for the future - but my preference is likely to
continue to be that the markup language used to author a particular page
should be identified on that page (and not globally). This can be done
either with an xhtml attribute on the <html:html> tag, or by using a
separate tag library that is essentially equivalent to struts-html.tld but
defaults the xhtml setting to true.
> Having said that, the whole point of xhtml is that it's valid xml webpages
> that all browsers can read. Yes, people could be using other clients
> besides browsers but those are probably few. I'm personally not worried
> about xhtml breaking backwards compatibility but do understand that others
> are. So, I think the solution suggested above is a decent compromise.
>
> Comments?
>
> David
>
Craig
[1] http://www.w3.org/TR/html4/intro/sgmltut.html#h-3.3.4.2
[2] http://www.w3.org/TR/xhtml1/#h-4.5
--
To unsubscribe, e-mail: <mailto:struts-dev-unsubscribe@;jakarta.apache.org>
For additional commands, e-mail: <mailto:struts-dev-help@;jakarta.apache.org>
_________________________________________________________________
Get a speedy connection with MSN Broadband. Join now! http://resourcecenter.msn.com/access/plans/freeactivation.asp
--
To unsubscribe, e-mail: <mailto:struts-dev-unsubscribe@;jakarta.apache.org>
For additional commands, e-mail: <mailto:struts-dev-help@;jakarta.apache.org>
