I transcribed your test a closely as possible into html, and your
intuition is born out (I'll send you my code if you like).
$style should look up the value of the css attribute on the enclosing
node (edgar). Assuming that by default attributes should inherit,
the cascade for testdata in edgar should be:
rule 3 (tag selector on ancestor node)
rule 1 (attribute selector on ancestor node)
rule 2 (tag selector on this node)
If you are getting something else, that seems like a bug to me. This
doesn't seem like that subtle a case, so I would be concerned.
On 2006-09-08, at 22:34 EDT, Benjamin Shine wrote:
> LPP-2618, combined with an example from Gregor, started an inquiry,
> which led me to something weird.
>
> Take these three rules:
>
> /* rule 1, specificity 10 */
> [name="thename"] {
> testdata: "i am thename";
> }
>
> /* rule 2, specificity 1 */
> testclass {
> testdata: "i am testclass";
> }
>
> /* rule 3, specificity 1 */
> view {
> testdata: "i am view";
> }
>
> And this view structure:
>
> <view name="thename">
> <testclass id="edgar">
> <attribute name="testdata" value="$style
> {'testdata'}" />
> </testclass>
> </view>
>
>
> What should the value of edgar.testdata be?
> Intuitively, I want edgar.testdata to be "i am testclass". What
> *edgar* is should matter more than one of edgar's ancestors'
> attributes.
> That's not what happens.
> edgar.getAttribute("testdata") returns "i am view"
>
> All the nearby screw cases I can think of seem to be handled the
> way I expect them to be, but inconsistently with the edgar case.
>
> I'm attaching a tar with the test cases, and a couple other test
> cases I've been working on. Adam and Gregor -- what do you think?
> Am I perseverating here, or is this really a sign of a problem that
> we care about?
>
> This test is in
> http://localhost:8080/trunk-csswork/test/style/attrselector/main.lzx
>
>
> _______________________________________________
> Laszlo-dev mailing list
> [email protected]
> http://www.openlaszlo.org/mailman/listinfo/laszlo-dev
_______________________________________________
Laszlo-dev mailing list
[email protected]
http://www.openlaszlo.org/mailman/listinfo/laszlo-dev
