- From: Terje Bless <link@tss.no>
- Date: Sun, 22 Oct 2000 02:27:41 +0200
- To: W3C Validator <www-validator@w3.org>
- cc: Christian Smith <csmith@barebones.com>, Bertilo Wennergren <bertilow@hem.passagen.se>
On 21.10.00 at 14:29, Christian Smith <csmith@barebones.com> wrote: >If there is a popup that lists DTDs there needs to be an option by which >the user selects to have a DICTYPE in the document override the DTD chosen >from the popup. Selecting this option means > > Use the DOCTYPE in the document but if there is no DOCTYPE use the > one selected in the popup. > >This is useful when the user only wants to specify a default DTD for when >there is no DOCTYPE. > >But, you also want the user to be able to choose a set of options that >translate to > > Use the DOCTYPE selected in the popup regardless of the DOCTYPE in > the document. > >so that the user can easily see (without having to edit the documents) how >close to compliancy with a specific version the pages are. This would be >useful for instance when validating HTML 4.0 Transitional documents to see >how much work it will be to make them compliant with 4.01 Strict. This is already done[0]. See <URL:http://lists.w3.org/Archives/Public/www-validator/2000OctDec/0037.html>. The way this works is that the front page gets a new popup that lists a subset of common known HTML Versions. The default entry is "Inline", which means "Use the DOCTYPE declaration in the document" and this is what happens if you don't specify anything at all. If a different HTML version is selected, we parse the document and comment out any existihng DOCTYPE, inserting instead the selected one. Such documents can never be valid; merely without errors. So the default is as your first case above, and if you select anything you get the second case. If you use the defaults, the current DOCTYPE guessing is used to autodetect XML, XHTML, and unlabeled HTML. My intent, and ISTR Gerald's as well, is that once the DOCYPE override has been tested and found solid the DOCTYPE guessing is removed in favor of simple XML-sniffing[1]; i.e. "Is this SGML or XML?", with a special case for XHTML[2]. [0] - Or it will be when I merge in Gerlad's latest flurry of updates and resubmit a patch that will actually apply cleanly. Depending on the phase of the moon I may even do some testing this time. :-) [1] - Which we seem to be unable to get rid of[3]. :-( [2] - Which we seem to be unable to get rid of[4]. :-( [3] - Thanks /so/ much XML WG[5]! :-( [4] - Thanks /so/ much HTML WG[5]! :-( [5] - :-) -- "Python 2.0 beta 1 is now available from BeOpen PythonLabs. There is a long list of new features since Python 1.6, released earlier today. We don't plan on any new releases in the next 24 hours." - From Python 2.0b1 Announcement
Received on Saturday, 21 October 2000 21:05:42 UTC