After I said that we thought fixing exists? would be hard, I wondered if maybe we could just catch the error. I think there may have been a time where we thought using rescue was cheating. Anyway +1 on the fix.
Should we depend on RAutomation 0.6.2 or rather ~> 0.6 so that we would automatically pick up any later patches as well? Bret On Thu, Jul 7, 2011 at 12:31 PM, Jarmo <[email protected]> wrote: > Thanks for the challenge :P > > https://github.com/bret/watir/commit/4393747a6fb1ce698aa9c89fb36965985e8ddd28 > > I'd like to make a new release soon with the following changes: > * Bump RAutomation dependency version to 0.6.2 to: > - fix/improve Watir::IE#send_keys. > - fix Watir::IE loading for certain Windows XP environments > (http://jira.openqa.org/browse/WTR-484). > * Restored Watir::IE#autoit method to use RAutomation's AutoIt adapter > with a deprecation warning. > * Cache IE.version result to not access registry with each execution - > was causing slowness in TextField#set for example. > * Do not raise UnknownObjectException if parent element doesn't exist > when calling Element#exists? > > Any objections? > > Jarmo > > On Thu, Jul 7, 2011 at 7:27 PM, Charley Baker <[email protected]> > wrote: > > I also think it's a great idea. I remember talking about it with Bret at > one > > point, and it was more challenging than it seemed at first glance. I've > also > > had to do the wait_until blocks with rescue for Ajaxy pages. exists? > should > > not explode imho. > > Cheers, > > Charley > > > > > > On Thu, Jul 7, 2011 at 10:16 AM, Jarmo <[email protected]> wrote: > >> > >> Haven't digged into technicalities yet. Just wanted to make sure that > >> this behavior would make sense to everyone else too before starting to > >> spend time on it. > >> > >> Jarmo > >> > >> On Thu, Jul 7, 2011 at 7:13 PM, Bret Pettichord <[email protected]> > >> wrote: > >> > I think I tried to do this years ago, but found it to be too > difficult. > >> > How > >> > would you do it? > >> > > >> > Bret > >> > > >> > On Thu, Jul 7, 2011 at 10:17 AM, Jarmo <[email protected]> wrote: > >> >> > >> >> Hi! > >> >> > >> >> Currently if there's some elements which doesn't exist, but will > exist > >> >> after some time (for example, after some ajax request), then i have > to > >> >> do something like this: > >> >> browser.wait_until {div(:id => 'doesntexist').span(:class => > >> >> 'doesntexisteither').exists? rescue false} > >> >> or > >> >> browser.wait_until {div(:id => 'doesntexist').exists? && div(:id => > >> >> 'doesntexist).span(:class => 'doesntexisteither')} > >> >> > >> >> I don't like much both of the ways and started thinking that what if > >> >> #exists? returned a boolean all the time? What if it returns false if > >> >> the parent element doesn't exist instead of raising that unneeded > >> >> exception? Since this would be a core change, i'd like to know what > do > >> >> you think and/or how do you handle situations like these in your > >> >> tests? > >> >> > >> >> How does watir-webdriver handle these situations? > >> >> > >> >> Jarmo > >> >> _______________________________________________ > >> >> Wtr-development mailing list > >> >> [email protected] > >> >> http://rubyforge.org/mailman/listinfo/wtr-development > >> > > >> > > >> > > >> > -- > >> > Bret Pettichord > >> > Director, Watir Project, www.watir.com > >> > > >> > Blog, www.testingwithvision.com > >> > Twitter, www.twitter.com/bpettichord > >> > > >> > > >> > _______________________________________________ > >> > 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 > > > > > > _______________________________________________ > > 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 > -- Bret Pettichord Director, Watir Project, www.watir.com Blog, www.testingwithvision.com Twitter, www.twitter.com/bpettichord
_______________________________________________ Wtr-development mailing list [email protected] http://rubyforge.org/mailman/listinfo/wtr-development
