- From: Charles McCathieNevile <charles@w3.org>
- Date: Tue, 21 Dec 1999 10:59:09 -0500 (EST)
- To: Vincent Quint <quint@w3.org>
- cc: Sergio Gelato <bglbv@my-deja.com>, www-amaya@w3.org
I agree that it is a more useful way to spend the very limited resources of the Amaya team to deal with well-formed HTML than to work out how to provide recovery for rubbish that looks like HTML. Charles McCN On Tue, 21 Dec 1999, Vincent Quint wrote: Sergio Gelato wrote: > > I actually have several concerns with this. > > 1) It would be nice, when loading an initially invalid document into > Amaya for editing, to have it transformed into something that doesn't > require *extensive* clean-up. In this case, the NAME= attribute > shouldn't have been inherited by every <A> element in sight. What makes > sense for fonts doesn't necessarily for anchors. Right. As you may have noticed, Amaya tries to fix *some* *common* HTML errors. A common error is a character level element such as font that encloses several block elements such as paragraphs, headings, lists, etc. In that regard, anchors are currently handled in the same way as font elements, just because they belong to the same category: character level elements. Obvioulsy, that's wrong and Amaya should make a special case for anchors. >In this case, I'd be happy with the <A NAME="Top"> being floated down > to the beginning of the BODY contents, and the missing </A> being > placed immediately afterwards (although that is perhaps less critical, > as long as the user is in a position to manipulate the anchor with > Amaya's editing tools). That'ss exactly what will be done for the next release. > 2) Somehow that out-of-place, unclosed anchor triggers other bugs in > Amaya. There is no excuse for breaking a line in the middle of a word, > for example. Nor for having such a garbled rendition of the document > in the structure view (both the Linux and the Solaris version print > several lines on top of each other, making the whole > thing hard to read). [I must say the X server I used was on Linux in > both cases; if it's an XFree86 bug, that would be interesting to know.] Don't blame Amaya for not recovering properly from an error. Blame the producer of this garbled pseudo-HTML file first. It's much easier to produce valid HTML by simply following the specification than trying to recover from a virtually infinite number of undocumented situations. Anyway, we have learned from that new case, and Amaya will now handle it properly, but I am sure someone else will report another strange HTML error soon, and we will have to cope with it. That's an endless effort and we prefer to spend our limited resources to develop new features, rather than trying to fix all possible HTML errors. > 3) I know I ought to look this up in the spec, but is it legal for several anchors to be given the same NAME attribute value? Isn't NAME like ID, in that the values are meant to be unique across a document? No, it's not legal. It will be fixed in the next release. > On Mon, 20 Dec 1999 14:18:14 Irene.Vatton wrote: > > > >Hi, > > > >Amaya detects that the anchor <A NAME="Top"> is misplaced. That's true, > >but I agree the result is surprising. In fact Amaya tries to apply the > >same fix as it does for misplaced <font> elements. > >What are you expecting: Amaya ignore the invalid anchor? Amaya creates an > >empty anchor as the first element of the body? what else? ------------------------------------------------------- Vincent Quint INRIA Rhone-Alpes W3C/INRIA ZIRST e-mail: Vincent.Quint@w3.org 655 avenue de l'Europe Tel.: +33 4 76 61 53 62 38330 Montbonnot St Martin Fax: +33 4 76 61 52 07 France -- Charles McCathieNevile mailto:charles@w3.org phone: +61 409 134 136 W3C Web Accessibility Initiative http://www.w3.org/WAI 21 Mitchell Street, Footscray, VIC 3011, Australia (I've moved!)
Received on Tuesday, 21 December 1999 10:59:18 UTC