Remove cascadeScope

Remove cascadeScope

As <style scoped> is removed, |cascadeScope| parameters
which was used for determining which style will be applied
is no longer necessary.
Only cascadeOrder is enough.

BUG=none
TEST=no new layout test failure

CLOSED IN FAVOR OF https://codereview.chromium.org/577423002

[tasak, 2014-08-08 07:00:26]

I'm just curious: compareRules is:

static inline bool compareRules(const MatchedRule& matchedRule1, const
MatchedRule& matchedRule2)
{   
    if (matchedRule1.cascadeScope() != matchedRule2.cascadeScope())
        return matchedRule1.cascadeScope() > matchedRule2.cascadeScope();

    unsigned specificity1 = matchedRule1.specificity();
    unsigned specificity2 = matchedRule2.specificity();
    if (specificity1 != specificity2)
        return specificity1 < specificity2;

    return matchedRule1.position() < matchedRule2.position();
}


So cascadeOrder is position(). cascadeScope is cascadeScope().
If removing cascadeScope, we cannot make some rule whose specifity is very small
win.

[kochi, 2014-09-19 01:38:45]

kochi@chromium.org changed reviewers:
+ esprehn@chromium.org

[kochi, 2014-09-19 01:38:58]

+esprehn

[esprehn, 2014-09-19 01:41:40]

On 2014/09/19 at 01:38:58, kochi wrote:
> +esprehn

I took this over in https://codereview.chromium.org/577423002, not sure about
tasak@'s comment but given that cascadeOrder is always zero now I don't think
we've lost anything here.

[kochi, 2014-09-19 01:46:49]

On 2014/09/19 01:41:40, esprehn wrote:
> On 2014/09/19 at 01:38:58, kochi wrote:
> > +esprehn
> 
> I took this over in https://codereview.chromium.org/577423002, not sure about
> tasak@'s comment but given that cascadeOrder is always zero now I don't think
> we've lost anything here.

Please do so, once https://codereview.chromium.org/585683002/ gets committed.
I agree that cascadeOrder corresponds to <style scoped>, so now no user uses
the value.

[tasak, 2014-09-19 02:31:04]

On 2014/09/19 01:46:49, Takayoshi Kochi wrote:
> On 2014/09/19 01:41:40, esprehn wrote:
> > On 2014/09/19 at 01:38:58, kochi wrote:
> > > +esprehn
> > 
> > I took this over in https://codereview.chromium.org/577423002, not sure
about
> > tasak@'s comment but given that cascadeOrder is always zero now I don't
think
> > we've lost anything here.
> 
> Please do so, once https://codereview.chromium.org/585683002/ gets committed.
> I agree that cascadeOrder corresponds to <style scoped>, so now no user uses
> the value.

Do you have any plan about removing / deprecating webkit-prefixed custom pseudo,
e.g. input::-webkit-slider-thumb?

If not, we should apply input::-webkit-slider-thumb in a parent treescope to
current treescope.

Since TreeScope == CascadeScope, we cannot remove CascadeScope, I think.