- From: David Ornstein <davido@objarts.com>
- Date: Thu, 25 Apr 1996 15:15:14 -0700
- To: erik <erik@inch.com>, Foteos Macrides <MACRIDES@sci.wfbr.edu>
- Cc: www-html@w3.org
At 05:39 PM 4/25/96 -0400, erik wrote:
>This syntax is comletely unacceptable for a language which may
>have </script> or other non-sgml sequences embedded within the
>script.
>
>All we need to do is then add a separator tag to
><script>....this would prevent this from occurring in the cases
>where a language may not be SGML compatible.
>
>The lexer would have to "shift gears" after encountering a
><script> tag with a separator.
As I said before, I'd *like* to avoid having a lexer have to know about the
tags it's dealing with. Could we use a combination of techniques like this?
<script type=txt/smalltalk>
<![ CDATA [
Transcript show: 'This is a cool idea'.
Disk format.
]]>
</script>
I don't much like all the funny punctuation, but it substantially eases the
burden on the parser/lexer and allows for more sophisicated systems to be
built...
Or one could do something along the lines of the <object> proposal that
Foteos referenced (could someone send me a good URL for this proposal so I
can finally read it?):
<object
id="smalltalk"
classid="progid:Smalltalk"
codebase="http://www.vendor.com/bin.smalltalk.cab"
>
</object>
<script SCRIPTENGINE="#smalltalkl">
<![ CDATA [
Transcript show: 'This is a cool idea'.
Disk format.
]]>
</script>
Received on Thursday, 25 April 1996 18:16:24 UTC