1214725 - Add info about attributes to SVG info DB

Status: RESOLVED WORKSFORME

(Developer Documentation Graveyard :: Macros/Templates, enhancement)

Description

[Sebastian Zartner [:sebo]]

The SVG JSON DB should contain information about the SVG attributes.
See bug 1212826 comment 2.

Sebastian

Comment 1

[Sebastian Zartner [:sebo]]

(In reply to Jeremie Patonnier :Jeremie from bug 1212826 comment #7)
> Hi!
> 
> SVG attributes are quite complex. It's true that the global vs specific
> distinction is "wrong". To be accurate there is 3 kind of attributes on SVG
> elements:
> 
> * Attributes that can be used by all SVG elements (likely some sort of
> "global" attributes)
>   https://svgwg.org/svg2-draft/struct.html#TermCoreAttribute
> 
> * Attributes that can be used by a given group of SVG elements
>   Some spec try to define groups of attributes that apply in bulk to some
> given elements:
>   https://svgwg.org/specs/animations/#AdditionAttributes
>   https://svgwg.org/specs/animations/#TermAnimationEventAttribute
>   https://svgwg.org/specs/animations/#AttributeNameAttribute
>   https://svgwg.org/specs/animations/#TimingAttributes
>   https://svgwg.org/specs/animations/#ValueAttributes
>   https://svgwg.org/svg2-draft/struct.html#TermARIAAttribute
>   https://svgwg.org/svg2-draft/struct.html#TermConditionalProcessingAttribute
>   https://svgwg.org/svg2-draft/interact.html#TermDocumentEventAttribute
>  
> http://www.w3.org/TR/2014/CR-html5-20140204/webappapis.
> html#globaleventhandlers
>   https://svgwg.org/svg2-draft/interact.html#TermGraphicalEventAttribute
>   https://svgwg.org/svg2-draft/styling.html#PresentationAttributes
>   https://svgwg.org/svg2-draft/styling.html#ElementSpecificStyling
>   https://svgwg.org/svg2-draft/linking.html#XLinkRefAttrs
>   ...
>   And I'm pretty sure I miss some.
> 
> * Attributes that can be used by a single SVG elements
>   Basically all attributes that are listed individually within elements
> definition
> 
> The fun parts begin with the fact that some attributes can have the same
> name but behave very differently from one element to the another. For
> example: fill which is different for basic shapes and animation elements:
> https://svgwg.org/svg2-draft/painting.html#FillProperty 
> https://svgwg.org/specs/animations/#FillAttribute
> 
> And of course, there are the presentation attributes than can be used both
> as CSS properties or attributes.
> 
> Because of all that mess, the spec do very little difference between
> attributes:
> https://svgwg.org/svg2-draft/attindex.html
> 
> Oh, and for what it worth, we should relay on the SVG2 spec draft (and all
> its independent modules) for all things about SVG if we want an accurate
> vision of what really happens within browsers:
> 
> SVG 2: https://svgwg.org/svg2-draft/ 
> SVG Animation: https://svgwg.org/specs/animations/
> SVG Markers: https://svgwg.org/specs/markers/
> SVG Paths: https://svgwg.org/specs/paths/ (no attributes defined)
> SVG Strokes: https://svgwg.org/specs/strokes/
> (Note: SVG fonts have been removed from SVG 2 clear and simple, we shouldn't
> bother documenting that)
> 
> I have no idea what should be done in the JSON, but I would tend to make a
> single attribute/element association and if it's possible to do some
> factorization (to avoid documenting things twice), that will be at another
> level.

Comment 2

[Jean-Yves Perrier [:teoli]]

I think we should list all attributes individually in the 'element' part of the json.

In the attribute part we may link to a group.

To document, we would have 1 page by elemennt, 1 page per attribute-element (that is the same attribute name in different pages would have a different pages) and summary pages that woould list all attribute of a group (Like 'Addition attributes').

That way people looking for a specific attribute on an element will find it, people looking for all attributes of an element too, as well as people looking for a group of attributes (E.g additive attributes).

Comment 3

[Florian Scholz (Open Web Docs)]

This will be tracked on GitHub going forward.
See https://github.com/mdn/data/issues/59