Re: Note regarding Proposal for issue #162: 'MediaStreamError should not be an interface'

No one objected, so the conclusion is that the editors will incorporate 
Dan's proposal (as outlined further below).

Stefan for the chairs

On 29/05/15 08:46, Stefan Håkansson LK wrote:
> Note: the definition of Errors is something we have been struggling
> with. Dan has now (after discussing with people) developed a
> proposal, see below.
>
> The proposal was sent Tuesday this week, unless we hear objections
> by Tuesday next week we will ask the editors to incorporate the
> proposal in the Media Capture and Streams document.
>
> Stefan for the chairs.
>
>
> On 26/05/15 15:11, Dan Burnett wrote:
>> I have been tasked with providing a proposal for how to proceed to
>> address issue #162 [1].
>>
>> Based on a conversation among Domenic Denicola, Jan-Ivar, and
>> others in PR #170 [2] (which was merely a draft of how to add a new
>> error using ECMAScript 6-style language), I suggest the
>> MediaStreamError defined in the Media Capture and Streams
>> specification be removed and the errors in section 7.2 be addressed
>> as follows.
>>
>> The following DOMException errors already defined in WebIDL will
>> use their WebIDL names: - AbortError - NotSupportedError -
>> NotFoundError
>>
>> The following errors that are new but semantically similar to
>> existing error names will use those names, as suggested by Domenic
>> Denicola in the above PR: - PermissionDeniedError will be replaced
>> with the DOMException SecurityError - SourceUnavailableError will
>> be replaced with the DOMException NotReadableError
>>
>> The two errors OverconstrainedError and ConstraintNotSatisfiedError
>> will be combined into one new OverconstrainedError to be defined
>> using EMCAScript 6-style language and include a new property
>> 'constraint' to hold the name of a mandatory constraint that is
>> overconstrained/not satisfied.  It is not possible to use WebIDL
>> here (see the PR conversation), so the ES6 definition is the way we
>> would need to proceed.
>>
>>
>> Does anyone on the list have objections to this approach that they
>> would like to discuss?  If not, I will write up the pull request
>> making these changes.
>>
>> Any remaining events, if any, that could be generated based on the
>> new set of errors (beyond the asynchronous returns already
>> addressed by Promises) would require at least one additional pull
>> request after this one to spec out.
>>
>> -- dan
>>
>> [1] https://github.com/w3c/mediacapture-main/issues/162 [2]
>> https://github.com/w3c/mediacapture-main/pull/170
>>
>
>
>


Received on Thursday, 4 June 2015 07:20:51 UTC