#selenium

lukeis, we're not going to support java8 until it has been EOL'd and java10 is out

>.>

jleyba: also, sorry i was a bit wrong on the dates i gave you before for selenium conf (not that it's likely to change for you…) it's sept. 8-10 (not 10-12)

lukeis, yeah that works even less :)

vacation after memorial day i take it? ;)

s/memorial/labor/

can't we just rename these holidays to "start of summer" and "end of summer" :)

jimevans: welp… that's one test failure for FF39 that I can't account for ^^

okay, doing a local build

[13selenium] 15jimevans pushed 2 new commits to 06master: 02http://git.io/vYAAg

13selenium/06master 1428a167a 15Jim Evans: Syncing IE driver for impending 2.47 release

13selenium/06master 145bb60fa 15Jim Evans: Updating version numbers for .NET assemblies for impending 2.47 release

ruh-roh raggy

uhh, I really need to fix that while I still have access..

once and forever

ugh. our developers make sure I'm never bored with automation...

they create such pages that selenium refuses to work with them -_-

very considerate of them

I mean, what the hell? Selenium scrolls to element and clicks on it. The problem is, at that point element is located at the top of the screen and is hidden by semi-transparent topbar.

so it is not clicked.

I'm actually surprised it has to scroll element to the top of the page, you would think that at least middle would suffice.

but oh well, what do I know

octomeow: the scroll behavior is entirely consistent with the javascript element.scrollIntoView() function.

jimevans, that much I figured. I am trying to think now of a way to go around this, cause even if I click, scroll up and click again, second click scrolls it back again.

and I'm curious as to why did it happen only recently. I thought our design wasn't changed as much.

depending on the browser you're having trouble with, you can alter the scroll behavior to scroll to the bottom of the viewport instead.

IE and firefox implement the elementScrollBehavior capability

m

can I ship yet?

set to 1, it behaves as if scrollIntoView(true) were called, which scrolls only enough to get the element into the viewport.

oh

i

i'll try that, thanks

set to 0, it scrolls to the top of the viewport.

AutomatedTester: had an unexpected failure in firefox 39 tests in :ci

Jenkins: http://ci.seleniumhq.org:8080/

waiting to see if it was a transient failure.

should know in the next 15 minutes or so

okey dokey

jimevans, basically it just calls element.scrollIntoView(false)?

octomeow: it calls element.scrollIntoView(<whatever value based on the capability>)

though honestly, i forget which value (true or false) means which behavior

false is bottom

as we just tested in console

should do. at least, firefox *used* to support that capability.

dunno if it still does.

ok, I'll try it now and tell you

and the driver should be smart enough not to scroll if the element is already in the view port.

so the solution of 'scrolling it back into the middle' should work fine.

for the subsequent click.

lukeis: there you go!

AutomatedTester: give me 5 minutes

getting ready to tag and bag the release

lukeis: ....... yeah i guess i'm next in line to be harassed. i'm supposed to find out how the CI works from seva one of these days? ... probably?

ok then… i'll call it a "flapper"

juangj: heh… please do soon! :)

[13selenium] 15jimevans pushed 1 new commit to 06master: 02http://git.io/vYxOd

13selenium/06master 140e4837e 15Jim Evans: Updating version information for 2.47.0 release

AutomatedTester: release the hounds!

release has been tagged

ok then… let's see if I can release maven in ~15min ;)

[13selenium] 15AutomatedTester pushed 1 new commit to 06master: 02http://git.io/vYxZr

13selenium/06master 14fa839d3 15AutomatedTester: Bump python to 2.47

chocks away!

titusfortner: you'll want to push a ruby version asap.

great, thanks for heads up

Hey jimevans: what does this "Native event support has been discontinued for versions of Firefox later than 33" piece mean? It sounds self-explanatory, but I'm doing some native event stuff using Firefox 39 right now, and it's working fine.

reminder… use java7 to build javadocs sigh

brma: no, you're really not. :)

even if you specifically request native events, you don't get them

heh

Is ActionChain synthetic?

check the driver.capabilities (after it's been created) you'll see "NativeEvents": false

it can be

hmmm, interesting…

c'mon maven… upload faster!

what's worse than downloading the internet? … uploading it!

python released