- From: James Graham <james@hoppipolla.co.uk>
- Date: Tue, 03 Mar 2015 17:49:37 +0000
- To: "public-browser-tools-testing@w3.org" <public-browser-tools-testing@w3.org>
The "no such element" error is, I think, currently doing double duty
(or, might be, the spec isn't really clear). One scenario in which it
comes up is
GET /session/{sessionId}/element/{ELEMENT}/selected
In this case {ELEMENT} not corresponding to a known element should
return a 404 with a body indicating it's a "no such element" error.
The other situation is
POST /session/{sessionId}/element/{ELEMENT}/element
with a body like
{"using": "css selector", value:"some selector"}
Where {ELEMENT} is a real element, but the selector doesn't match
anything. This clearly isn't a 404 error (maybe a 500; I'm not sure what
makes 501 appropriate here). However it is still a "no such element"
error in the spec.
Are people OK with this or do they expect each webdriver status code to
have a single http status?
Received on Tuesday, 3 March 2015 17:50:03 UTC