Maybe we should not allow to use xpath/css with other locators rather allow 
some of them. In all other cases where we construct xpath by ourselves, we can 
append index. 

If use uses xpath/css locator, he should add index by themselves or fetch 
collection. 

> On 03 февр. 2014 г., at 23:13, Jari Bakken <[email protected]> wrote:
> 
> Because without an actual XPath/CSS parser we'll quickly end up generating 
> invalid XPaths. Consider for example:
> 
> browser.element(xpath: './foo | ./bar')
> I don't think it's worth heading down that rabbit hole for a small 
> performance gain with a fairly rare use case.
> 
> However, if we control the XPath fully, it'd make sense to use XPath indexing 
> - I'm not sure if we even do that currently. There may be a good reason (and 
> probably demonstrated by the specs) that I can't remember.
> 
> —
> Reply to this email directly or view it on GitHub.
> 
> _______________________________________________
> Wtr-development mailing list
> [email protected]
> http://rubyforge.org/mailman/listinfo/wtr-development
_______________________________________________
Wtr-development mailing list
[email protected]
http://rubyforge.org/mailman/listinfo/wtr-development

Reply via email to