OK, to learn something from your problem: We do not escape > for any output method
as neither XML nor specifications don't require it. On the other hand, this escaping
is also not prohibited. I think we could change this approach to make a dealing with
HTML easier.
For each method, I would suggest to escape out the following:
text: nothing
xml: <, &
html: <, &, >
xhtml: <, &, >
Petr
David Burry wrote:
> FYI, in case you ever see this problem again, installing IE 5.5 SP1 has
> fixed my problem.
>
> Dave
>
> At 03:54 PM 7/17/2001 -0700, David Burry wrote:
>
> >By the way, under RFC 1866 (which is the HTML 2.0 "standard" if you can
> >call anything that) section 3.2.1, it does "suggest" the following:
> >
> >
> > > A useful technique for translating plain text to HTML is to replace
> > > each '<', '&', and '>' by an entity reference ...
> >
> >So you may want to consider this behavior for the HTML output method only,
> >since current browsers always seem to lag behind a few HTML version
> >numbers in implementation.
> >
> >Dave
> >
> >
> >
> >At 01:35 PM 7/17/2001 -0700, David Burry wrote:
> > >Some email readers will eat my HTML example, see the attachment for
> > another copy if you need one. This is an example of how current widely
> > used browsers need ">" to be escaped as an entity in all entity-escaped
> > HTML output, just like the XML spec specifies for "<" and "&" (I just
> > looked at the spec, it appears your implementation is indeed correct for
> > XML output, all the various HTML specs are more confusing to double check on).
> > >
> > >This example fails in IE 5.50.4522.1800 on Win2k 5.00.2195 (SP2)
> > >This example functions correctly on IE 5.50.4134.0600 on Win NT
> > 4.00.1381 (SP6), and Netscape 2.02, 3.04, 4.08, 4.51, 4.7, 6.0.
> > >
> > >Hmm, after the more thorough browser test, looks like I will be doing a
> > wipe/reinstall of my win2k system soon to try to get this to work, as I
> > _really_ do need it to work on that system, sorry for taking your time
> > when this bug report probably belongs on a craposoft internet exploder
> > mailing list instead. Anyway, still something to consider if you see
> > more people like me, if not, please ignore me.
> > >
> > >Dave
> > >
> > >At 09:51 AM 7/17/2001 -0700, you wrote:
> > >
> > >>I'm not certain about the XML spec, but in HTML it very much does need
> > to be escaped, specifically inside <textarea>...</textarea> tags in IE
> > 5.5 or else it does not let the form be submitted. Call that browser
> > broken if you wish, but that's the way the real world functions. I
> > cannot "just put &gt; in the primary source," because I'm getting the
> > source from a form submission like the one below, storing it in an XML
> > file, then restoring it dynamically to a form like the one below for
> > editing the value later. XSLT doesn't contain a generic
> > search-and-replace function to convert all > signs to &gt; on the
> > fly, and no, translate() will not work, though a a complicated loop of
> > substring(), substring-before() and substring-after() might do
> > it... Seems better to me to fix Sablotron to behave the way I need it at
> > least for HTML output. Hacking the form submission receiving script to do
> > this conversion might work also though that would be really weird for any
> > non-Sablotro!
> >!
> >n uses. Go ahead and try it:
> > >>
> > >>this is a <A HREF="/">test</A>.
> > >>On my IE 5.5 Win2k, nothing at all happens when you press
> > >>"submit"... Netscape seems to work ok. IE works fine if I entity escape
> > >>the ">" characters above.
> > >>
> > >>Dave
> > >>
> > >>
> > >>At 02:36 PM 7/17/2001 +0200, Pavel Hlavnicka wrote:
> > >>
> > >>>the '>' sign doesn't have to be escaped, so it is not escaped. If you
> > >>>really need > in xml output file, you need put &gt; in the primary
> > >>>source.
> > >>>
> > >>>Pavel
> > >>>
> > >>>David Burry wrote:
> > >>>
> > >>>>Hi, I'm relatively new to xslt and even newer to this list, so hope this
> > >>>>is the right place...
> > >>>>Is there a good reason why this xml:
> > >>>><?xml version="1.0" encoding="ISO-8859-1" ?>
> > >>>><tag>
> > >>>>this is a <A HREF="/">test</A>.
> > >>>></tag>
> > >>>>plus this xsl:
> > >>>><?xml version="1.0" encoding="ISO-8859-1" ?>
> > >>>><xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
> > >>>>version="1.0">
> > >>>><xsl:output method="xml" encoding="ISO-8859-1"/>
> > >>>><!-- nothing needed here, since default output is tag contents -->
> > >>>><!-- we could also put a template here that contains one of -->
> > >>>><!-- value-of select="." or value-of select="text()" -->
> > >>>><!-- and we'd get the exact same results -->
> > >>>></xsl:stylesheet>
> > >>>>produces this output:
> > >>>><?xml version="1.0" encoding="ISO-8859-1"?>
> > >>>>this is a <A HREF="/">test</A>.
> > >>>>instead of this output:
> > >>>><?xml version="1.0" encoding="ISO-8859-1"?>
> > >>>>this is a <A HREF="/">test</A>.
> > >>>>?????
> > >>>>This kind of behavior is really messing up my HTML, making forms not
> > >>>>function (i.e. not submit) in IE because of technically badly formatted
> > >>>>default values, etc...
> > >>>>Dave
> > >>>
> > >>>
> > >>>
> > >>>--
> > >>>Pavel Hlavnicka
> > >>>Ginger Alliance
> > >>>www.gingerall.com
> > >
> > >
--
Petr Cimprich
Ginger Alliance
www.gingerall.com
