- From: Michael A. Puls II <shadow2531@gmail.com>
- Date: Fri, 13 Mar 2009 00:25:25 -0400
- To: "Martin Duerst" <duerst@it.aoyama.ac.jp>, uri@w3.org
On Thu, 12 Mar 2009 06:49:26 -0400, Martin Duerst <duerst@it.aoyama.ac.jp> wrote: >> A note about this in the mailto spec (or some new mailto form >> submission spec) might help clear this up. > > I'm still not sure why there's a need to "help clear this up". In <http://shadow2531.com/opera/userjs/BeforeMailtoURL.js>, look at the first line of createMailtoURIFromEnabledFormControls(form). It's var uri = "mailto:?";. You should see how that could be an error (as it always drops [to]). The sole reason for doing it like that was because the mailto spec says that 'mailto:x' is equivalent to 'mailto:?to=x'. So, because I followed the mailto spec a little too literally in that case, I messed things up. So, I'm saying that if I messed up because of what the mailto spec said, others might too and the spec should leave no margin for interpretation in that area. And, I know that HTML form handling specifies what to do, but in this case the mailto spec was contradicting and I chose to give it priority. It's the "equivalent" part in the mailto spec that got me. Using "different" would be better. Example suggested text: There are two different ways to specify a To: hfvalue in a mailto URI that Users Agents SHOULD support: <mailto:foo1@example.com%2C%20foo2@example.com> and <mailto:?to=mailto:foo1@example.com%2C%20foo2@example.com>. However, you SHOULD NOT use both ways at the same time. Again, it's "equivalent" that threw me off. -- Michael
Received on Friday, 13 March 2009 04:26:00 UTC