On Fri, 13 Sep 2013, Ryosuke Niwa wrote: > On Sep 13, 2013, at 1:53 AM, Anne van Kesteren <ann...@annevk.nl> wrote: > > > > It's not entirely clear to me what the suggestion from Ryosuke is. Is > > it to keep getters on HTMLDocument only and expose HTMLDocument only > > in browsing contexts created from text/html resources? And use > > Document everywhere else and have Document contain all the other > > members including (former) SVGDocument members? That setup would make > > sense to me anyway. > > My proposal something along that line.
Could you be more precise? On Fri, 13 Sep 2013, Simon Pieters wrote: > > Let's list concrete cases and see what we want to do with each one. > > In a browsing context: > > * text/html document (uses interface "Document" and is an "HTML document") > * XML document (uses interface "Document" and is an "XML document") > > Not in a browsing context: > > * document.implementation.createDocument() (uses interface "XMLDocument" and > is an "XML document") > * document.implementation.createHTMLDocument() (uses interface "Document" and > is an "HTML document") > * new Document() (uses interface "Document" and is an "XML document") > * XHR response (uses interface "Document" and is an "XML document", even for > text/html if I read the spec correctly) > > For instance the createHTMLDocument() case currently supports named > getter in Gecko but not in Blink. > http://software.hixie.ch/utilities/js/live-dom-viewer/saved/2519 The list above would be a good one to start from. What features would you want in which cases? On Fri, 13 Sep 2013, Ryosuke Niwa wrote: > > But HTMLDocument also has a bunch of other junk properties like embeds, > plugins, scripts, compatMode, bgColor, fgColor, alinkColor, linkColor, > vlinkColor, etc… We even have write() and writeln() there! > > I don’t want to start adding all these legacy methods elsewhere. These aren't all junk. "scripts" for example is a perfectly valid and useful API, which it would definitely make sense to use from within an HTML fragment inside an SVG document. write() and open() are only useful in an HTML context, since it bails if the Document is an XML one. Also, the vast majority of the time, a document is the kind of document that needs these APIs. There's not really much of an "elsewhere" for them to be found in. It's not like having them in XML documents actually affects many people, so not having them doesn't actually help any appreciable number of people either. But, I'm certainly open to having a more elaborate scheme where some APIs aren't available everywhere if that's what it takes for us to make SVG-in- text/html-HTML and HTML-in-XML-SVG be able to generally use the same APIs. I think the way to move forward with this would be to describe: (a) The different categories that you think should behave differently, (b) How each category should behave, (c) Within each category, what APIs you think should be available (or, possibly better, a general rule to decide what API is available and a brief example of how this is applied to a subset of the total API surface -- we can extrapolate from there) Currently, the spec has an almost trivial answer to this; there's basically just the one category except for XMLDocument which is used in some edge cases needed for compat. -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'