Nice! One test started failing, but the test was wrong:
http://localhost:8080/trunk/test/style/constraints/constraint-test.lzx
which comes down to a comparison between two selectors
styledbox {...} vs #gSB { ... }
for this node:
<styledbox x="20" y="20" id="gSB" />
The #gSB selector should win; id selectors win over element
selectors. The test wanted the element selector to win. Change the
test to:
this.assertEquals( 0xAA0000, gSB.bgcolor);
then you're approved.
-ben
On Sep 11, 2006, at 5:50 PM, Adam Wolff wrote:
> Change change.XXXXXXXXX.xQax4u3O.txt by [EMAIL PROTECTED] /Users/adam/
> laszlo/src/svn/openlaszlo/trunk/WEB-INF/lps/lfc/ on 2006-09-11
> 17:05:21 PDT
>
> Summary: Fix problems with CSS rule *applicability.*
> This is different from specificity; specificity depends
> solely on
> a rule description. Applicability is a function of both
> the rule
> description and the node for which the style is looking up
> the
> ruleset.
>
> In our case, when two rules have matching specificity, we
> consider how many steps total up the class hierarchy are
> required
> to meet each class selector in the rule. The rule with the
> fewer
> number of steps wins.
>
> New Features:
>
> Bugs Fixed:
> applicability testcase in trunk/test/style/specificity/
> applicability.lzx
> also fixed performance hit caused by implementation of
> classAndAttr
> type selector
>
> Technical Reviewer: ben
> QA Reviewer: (pending)
> Doc Reviewer: (pending)
>
> Documentation:
> This does have doc impact, as it refines the W3C rules for CSS the
> applicability of CSS selectors. That appears to be unspecified
> by our
> reading of the spec.
>
> Release Notes:
>
> Details:
> This change counts how many times a rule must hop to superclass to
> match any of its selectors. This is stored on the rule as it is
> tested
> for matching. The sort function then looks at this value when
> sorting
> the rule list.
>
> Note that these counts are in aggregate. The behavior of complex
> selectors whose counts happen to match for these operations is
> unspecified.
>
> Also, a minor change here to the _addRule function to take
> advantage of the name= optimization for Diamond.
>
> Tests:
> http://localhost:8080/trunk/test/style/specificity/
> applicability.lzx
> http://localhost:8080/trunk/test/style/specificity/specificity-
> tests.lzx
> http://localhost:8080/trunk/test/style/descendantselector/
> complex-descendant-selector-test.lzx
> http://localhost:8080/trunk/test/style/attrselector/main.lzx
>
> This test had slowed down due to earlier changes:
> http://localhost:8080/trunk/diamond/client/test/speedtest.lzx
>
> The caching change above fixes this
>
> Files:
> M WEB-INF/lps/lfc/services/LzCSSStyle.js
> M test/style/specificity/applicability.lzx
> <patch.adam.4DDI.tgz>
_______________________________________________
Laszlo-dev mailing list
[email protected]
http://www.openlaszlo.org/mailman/listinfo/laszlo-dev
