Unfortunately it's not possible to generate that from an xslt
processor / libxml2 / lxml, and in order to trigger the xhtml output
mode (so you get <br /> with the space) you need to specify an xhtml
1.0 doctype to be output. It seems quite likely with deco / blocks /
xdv that we will have an lxml based output chain, so we will be
restricted in what's possible. This has been brought up previously on
the libxml2 list, though without resolution (I can't find the
reference to that right now).

We might want to start campaigning now for <!DOCTYPE HTML SYSTEM
"about:legacy-compat"> (and indeed -//W3C//DTD XHTML 1.1//EN) to be
added to the doctypes that trigger xhtml compatible output in
libxml2's xmlsave.c

Also here http://www.w3.org/2008/08/cleantheweb/libxml  and here
http://wiki.whatwg.org/wiki/FAQ#What_MIME_type_does_HTML5_use.3F it
states the Content-Type should be application/xhtml+xml for the xml
serialization, so I guess absolute conformity may be impossible,
though self-closing tags seem to be allowed for the html serialization
too so maybe we're ok there.

Laurence

On 16 March 2010 22:50, Alexander Limi <l...@plone.org> wrote:
> Right, I don't see a reason to do that, though — it doesn't buy us anything.
>
> The reason the HTML5 doctype is simply:
>
> <!DOCTYPE html>
>
> …is that it's the shortest possible string that will trigger
> strict/standards parsing (ie. not quirks mode) in all browsers, including
> IE6.
>
>
> On Tue, Mar 16, 2010 at 3:34 PM, Laurence Rowe <l...@lrowe.co.uk> wrote:
>>
>> It is listed as an "obsolete permitted doctype string"
>>
>> http://dev.w3.org/html5/spec/Overview.html#obsolete-permitted-doctype-string
>> - i.e. we can lie about the doctype. I'm not sure why xhtml 1.0
>> transitional is not allowed.
>>
>> Laurence
>>
>> On 16 March 2010 22:18, Alexander Limi <l...@plone.org> wrote:
>> > The way it works is that you can use the XHTML "spelling" (ie. closing
>> > your
>> > tags), but you serve it up as normal HTML.
>> >
>> >
>> > http://wiki.whatwg.org/wiki/FAQ#Should_I_close_empty_elements_with_.2F.3E_or_.3E.3F
>> >
>> > There's no Strict or similar thing in HTML5, AFAIK.
>> >
>> > (There is also something informally referred to as "XHTML5" which is
>> > serving
>> > it as XML, which isn't what we want to do)
>> >
>> > On Tue, Mar 16, 2010 at 3:06 PM, Laurence Rowe <l...@lrowe.co.uk> wrote:
>> >>
>> >> By my reading of the html 5 draft, it would seem conformant with the
>> >> (html5) spec to serve a document with a text/html Content-Type but an
>> >> XHTML Strict doctype.
>> >>
>> >> On 16 March 2010 20:14, Alexander Limi <l...@plone.org> wrote:
>> >> > What does transitional doctype have to do with geolocation?
>> >> >
>> >> > (and XHTML STRICT is a problem, since it implies serving with XML
>> >> > MIME
>> >> > type,
>> >> > which IE doesn't handle, so that's unlikely to happen)
>> >> >
>> >> >
>> >> > On Tue, Mar 16, 2010 at 12:48 PM, Veda Williams <v...@groundwire.org>
>> >> > wrote:
>> >> >>
>> >> >> This brings up the question of when we're moving away from
>> >> >> Transitional
>> >> >> DOCTYPE. Do we have a sense of when this will happen? I'm
>> >> >> particularly
>> >> >> keen
>> >> >> on knowing, as it opens up the door for us in terms of geolocation
>> >> >> in
>> >> >> the
>> >> >> next year or so.
>> >> >> Thanks,
>> >> >> - Veda
>> >> >>
>> >> >>
>> >> >> On Mar 16, 2010, at 12:40 PM, Alexander Limi wrote:
>> >> >>
>> >> >> On Tue, Mar 16, 2010 at 4:45 AM, Wichert Akkerman
>> >> >> <wich...@wiggy.net>
>> >> >> wrote:
>> >> >>>
>> >> >>> I'ld like to see a list of pros and cons of using HTML 5 as well. I
>> >> >>> am
>> >> >>> quite worried by the lack of proper support in existing browsers.
>> >> >>> None
>> >> >>> of
>> >> >>> them implement any of the existing HTML standards properly, and I
>> >> >>> fear
>> >> >>> that
>> >> >>> switching to the still unfinished HTML5 would be a several steps
>> >> >>> too
>> >> >>> far at
>> >> >>> this point in time.
>> >> >>
>> >> >> What parts in particular do you find are not working? Browsers that
>> >> >> don't
>> >> >> have dedicated support for HTML5 will just treat those tags similar
>> >> >> to
>> >> >> div
>> >> >> elements (given an HTML5 shiv for styling to apply in IE), and most
>> >> >> of
>> >> >> the
>> >> >> new form-related enhancements are additive in nature.
>> >> >>
>> >> >> In general, HTML5 renders even on IE6, there isn't much magic here
>> >> >> (but
>> >> >> of
>> >> >> course it doesn't get any of the advantages either). HTML5 is mostly
>> >> >> about
>> >> >> standardizing edge case behaviors and adding new abilities that will
>> >> >> gracefully degrade in older browsers — and then a few new tags like
>> >> >> video/audio (that are also relatively easy to make degrade) and
>> >> >> structural
>> >> >> elements like article/footer, etc.
>> >> >>
>> >> >> --
>> >> >> Alexander Limi · http://limi.net
>> >> >> _______________________________________________
>> >> >> Framework-Team mailing list
>> >> >> Framework-Team@lists.plone.org
>> >> >> http://lists.plone.org/mailman/listinfo/framework-team
>> >> >>
>> >> >> ________________________________
>> >> >> Veda Williams
>> >> >> Web Developer
>> >> >> Groundwire
>> >> >> 206.286.1235x23
>> >> >> v...@groundwire.org
>> >> >> ________________________________
>> >> >> ONE/Northwest is now Groundwire!  Read all about our new name.
>> >> >
>> >> >
>> >> >
>> >> > --
>> >> > Alexander Limi · http://limi.net
>> >> >
>> >> > _______________________________________________
>> >> > Framework-Team mailing list
>> >> > Framework-Team@lists.plone.org
>> >> > http://lists.plone.org/mailman/listinfo/framework-team
>> >> >
>> >> >
>> >
>> >
>> >
>> > --
>> > Alexander Limi · http://limi.net
>> >
>
>
>
> --
> Alexander Limi · http://limi.net
>

_______________________________________________
Framework-Team mailing list
Framework-Team@lists.plone.org
http://lists.plone.org/mailman/listinfo/framework-team

Reply via email to