- From: Alistair Turnbull <apt1002@hermes.cam.ac.uk>
- Date: Fri, 6 Dec 2002 18:11:13 +0000 (GMT)
- To: Terje Bless <link@pobox.com>
- cc: W3C Validator <www-validator@w3.org>, Nick Kew <nick@webthing.com>
Just now I tried to send you the broken URL but I can't because you've fixed the problem. There are now separate 'Validate URI' and 'Validate File' buttons, where before there was only 'Validate Page'. A perfectly satisfactory conclusion, really quickly. You're good! Alistair On Wed, 4 Dec 2002, Terje Bless wrote: > Nick Kew <nick@webthing.com> wrote: > > >I was inclined to ignore it at first ( "someone entered a URL in the > >file upload box and got confused ..." ), but on rereading your post, > >that seems unlikely. > > No, it seems clear that this is _not_ a user error. It's either an issue > wirh Konq, our web pages, or the Validator code (see below). > > > >Actually on reflection, this is probably down to the murky innards of > >the validator, specifically the redirection from a POST to a GET URL. > >Methinks it's time to thwack Terje about that one! > > This isn't due to the GET->POST thing; it actually looks like this stems > from an ambiguity on the HTML 4.01 Recommendation and Konq 2.x implementing > this slightly differently from other browsers. > > A file upload field that has no value may still be considered "successfull" > and so may be submitted. When an uploaded file is present the Validator > will prefer that over any submitted URI. And since the "file" Konq is > submitting is zero-length, we, for obvious reasons, can't extract an > encoding, hence the error message. > > > Since this doesn't appear to be a particularly widespread issue I'm not > certain to what extent we should compensate for it in the Validator. I'll > have a look at this the next time I'm digging through that part of the > code. > > > Thanks for bringing up this issue Alistair! This is pretty subtle and I > don't think we'd ever have stumbled on it by ourselves! > > -- > We've gotten to a point where a human-readable, human-editable text format > forstructured data has become a complex nightmare where somebody can safely > say "As many threads on xml-dev have shown, text-based processing of XML is > hazardous at best" and be perfectly valid in saying it. -- Tom Bradford >
Received on Friday, 6 December 2002 13:11:34 UTC