JEP 253 webrev ready for review

[Hervé Girod]

It seems to be a very good idea!

Sent from my iPad

> On 4 sept. 2015, at 07:25, Jonathan Giles <jonathan.giles at oracle.com> wrote:
> 
> For what it's worth, I think enabling this is a really good idea. The starting point is StyleConverter.java - it currently just hard codes all available converters. Making that pluggable won't be too difficult. If you (or anyone else) is interested in looking into this, we can discuss further.
> 
> -- Jonathan
> Sent from a touch device. Please excuse my brevity.
> 
>> On 4 September 2015 17:17:14 GMT+12:00, Michael Ennen <mike.ennen at gmail.com> wrote:
>> I am trying to mentally digest the CSS API, but I am having trouble
>> answering the question of whether or not the following scenario is now
>> possible (using only public APIs). Imagine one wanted to create a new
>> class
>> that was analogous to DeriveColorConverter.java, and that it worked
>> "all
>> the way down" in terms of being parsed from CSS into a
>> StyleableProperty<Color>. For concreteness, say one wanted to implement
>> a
>> fadein function:
>> 
>> fadein
>> fadein( <color> , <number>% )
>> 
>> The fadein function takes a color and computes a more opaque version of
>> that color.
>> The second parameter is the opacity, ranging from 0% to 100%. Has no
>> effect
>> on
>> fully opaque colors.
>> 
>> Does the API allow for a way to hook this new function up into the CSS
>> parser? If one implemented FadeInColorConverter extends
>> StyleConverter<ParsedValue[], Color> that worked according to the above
>> specs, how would one hook this fadein function into the parser, so that
>> one
>> can write CSS such as:
>> 
>> my-awesome-color: fadein(my-semi-transparent-color, 30%);
>> 
>> and have a StyleableObjectProperty<Color> come out the other side?
>> 
>> If this is possible using the new APIs, then this makes those APIs very
>> powerful in terms of the possibilities of extending the built-in CSS
>> capabilities in very powerful and interesting ways. If it is not
>> possible,
>> would it be feasible to make it so in the scope of this JEP?
>> 
>> Thanks very much,
>> -- 
>> Michael Ennen