W3C home > Mailing lists > Public > www-svg@w3.org > October 2013

Re: Missing values in SVGPathElement API: What to do?

From: Stephen Chenney <schenney@chromium.org>
Date: Thu, 24 Oct 2013 11:43:28 -0400
Message-ID: <CAObCcUrpD3CvS7aSBPpqU2nNoAHmSMofHF9hq6pu9dHTPJDi_Q@mail.gmail.com>
To: Rick <graham.rick@gmail.com>
Cc: "www-svg@w3.org" <www-svg@w3.org>
OK. So option 1 is correct, which the addendum that invalid input should
not modify the underlying path element.

Thanks.

On Thu, Oct 24, 2013 at 11:40 AM, Rick <graham.rick@gmail.com> wrote:

> The document is in error if required attributes are not supplied.
>
> I'm not sure where it says that, probably in the XML spec.
>
> Here's the defined behaviour.
>
> http://www.w3.org/TR/SVG/implnote.html#PathElementImplementationNotes
>
>
> On Thu, Oct 24, 2013 at 11:33 AM, Stephen Chenney <schenney@chromium.org>wrote:
>
>> I just came across some undefined behavior in the spec that could easily
>> be defined. An example is the
>> http://www.w3.org/TR/SVG/paths.html#InterfaceSVGPathElement spec. It has
>> absolutely nothing to say about cases where the calling script does not
>> provide all of the necessary parameters. I could find no blanket statement
>> in the spec about what to do in such cases. The canvas spec is similarly
>> unhelpful.
>>
>> e.g. what to do with
>> pathSegment = pathElement.createSVGPathSegCurvetoCubicAbs(1.0)
>>
>> Logical choices are either throw an exception or zero the missing values.
>>
>>  Right now Chromium, and presumably WebKit, pass NaN through to the back
>> end, which is bad and which I plan to somehow fix. I haven't checkout other
>> browsers yet.
>>
>> The question is, what fix?
>> 1) Exception and return null object
>> 2) Exception but fill missing values with 0
>> 3) Silently fill in missing values with 0.
>>
>> I'm liking 1 right now.
>>
>> Regardless of which option we choose, I think it would be helpful to have
>> a generic statement about this situation somewhere in the spec.
>>
>> Stephen.
>>
>
>
>
> --
> *Would be fair to suggest that perhaps Obama didn't lie when you clearly
> saw his lips move?*
> *
> *
>
>
Received on Thursday, 24 October 2013 15:43:56 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:54:46 UTC