W3C home > Mailing lists > Public > www-archive@w3.org > September 2014

Re: W3C Proposed Recommendation: HTML5

From: Anne van Kesteren <annevk@annevk.nl>
Date: Mon, 22 Sep 2014 16:35:54 +0200
Message-ID: <CADnb78h3MwLfEv+H23sQ0Ub06U=mL3tS_2shgoBBetCTamaf=Q@mail.gmail.com>
To: Robin Berjon <robin@w3.org>
Cc: Henri Sivonen <hsivonen@hsivonen.fi>, "L. David Baron" <dbaron@dbaron.org>, "www-archive@w3.org" <www-archive@w3.org>
On Mon, Sep 22, 2014 at 4:30 PM, Robin Berjon <robin@w3.org> wrote:
> I was thinking about that. We have a test suite for URL but apparently it's
> not complete enough to cover all the ground (I haven't checked but IIRC Mike
> has). Would you estimate that we're at a point where we can try to prod
> browsers in that direction or do you reckon you need more tinkering on the
> spec first? If the former, I would think that a test suite would be a decent
> place to start.

I wrote a test suite that's a decent start for parsing (based on test
suites from browsers). We've made some fixes in Gecko, but currently
URL parsing in Gecko means JavaScript can get invoked due to
extensions. We effectively have to overhaul the entire URL system,
which might break a ton of things. And without volunteers to take this
on, it'll take a while before Mozilla deems it's important enough I
think.

Having said that, the parsing algorithm in the specification might
change due to feedback from Simon Sapin. He wrote something in Rust
that would be a better basis for a more readable version of the
current algorithm without losing the important characteristics.


-- 
https://annevankesteren.nl/
Received on Monday, 22 September 2014 14:36:21 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:44:33 UTC