- From: Adam Bergkvist <adam.bergkvist@ericsson.com>
- Date: Mon, 16 Jun 2014 09:24:25 +0200
- To: Martin Thomson <martin.thomson@gmail.com>, Harald Alvestrand <harald@alvestrand.no>
- CC: Stefan Håkansson LK <stefan.lk.hakansson@ericsson.com>, tim panton <thp@westhawk.co.uk>, "Makaraju, Maridi Raju (Raju)" <Raju.Makaraju@alcatel-lucent.com>, Justin Uberti <juberti@google.com>, Kiran Kumar Guduru <kiran.guduru@samsung.com>, "public-webrtc@w3.org" <public-webrtc@w3.org>
On 2014-06-14 21:22, Martin Thomson wrote: > On 14 June 2014 02:19, Harald Alvestrand <harald@alvestrand.no> wrote: >>> FWIW, I'm with Tim on this one. I also think it makes it clearer - >>> either the UA throws, or it sends all symbols. >> >> FWIW, I think the spec will be simpler if we throw >> IllegalCharacterException on strings that contain any character except >> [0-9A-Da-d*#,], and have no special treatment at sendout time (illegal >> characters can't occur there). > > > FWIW, FWIW isn't necessary. This is simply good design. Having > something blow up early avoids nasty deferred surprises. So after a little detour we're back at the same set of recognized characters, but we throw instead of ignore. As a result, the problem that started this thread doesn't exist anymore. Let's put this into the spec. /Adam
Received on Monday, 16 June 2014 07:24:53 UTC