[dojo-contributors] BorderContainer etc. styling

sam foster potatosculptor at gmail.com
Thu May 22 14:27:20 EDT 2008


I dont see a problem with just adding the gutters="true" property to
the tests, themeContainer and our examples - while leaving the default
(the value on the prototype) to be border-less/gutter-less. That
"upgrades" the box without causing pain for people who already
unwrapped and started using it.

In the html world arguably precedent goes against this - frames,
tables both have default borders/padding - but I think if you cast
BorderContainer as a base class from which other layout widgets are
going to be built, the default should be as minimally intrusive as
possible. The themeTester shouldnt be driving requirements here imho -
that is under our control and we can build it any way we like.

IMO I dont think it matters enough to break people. Its enough that
its only an attribute toggle away, and provided we have clear
cut-n-pasteable example code for usage in either scenario, we're good.

my 2c,
Sam

On Thu, May 22, 2008 at 10:29 AM, Adam L. Peller <adam at peller.org> wrote:
> On Thu, May 22, 2008 at 7:02 AM, Bill Keese <bill at dojotoolkit.org> wrote:
>>
>> in general, this should make BorderContainer layouts prettier,
>> regardless of whether or not they have resizable panes.
>
> I disagree, but you're the BDFL.  I'd argue you're adding something to the
> layout algorithm that didn't exist before that only makes it 'prettier' in
> certain situations.
>>
>> > I agree that 'magic' to connect the concepts of container margins with
>> > panel spacers, or 'gutters' in designer-speak, doesn't feel right.  I
>> > like the idea of an attribute (on the container should be sufficient? or
>> > maybe child selector CSS?) to control this behavior, but requiring users
>> > to put in a setting to get the 1.1 behavior might be a viewed as a
>> > breaking change.  Is it possible we could flip the logic around and have
>> > the 1.1 behavior be the default?  gutters=true?



More information about the dojo-contributors mailing list