#selenium

I'm using it as a metric for how much work Microsoft has done on Edge :)

I'm still holding out hope Apple decides to join the party and implement webdriver for webkit/safari

I know they're aware of us and there are some teams there using the current safaridriver

So you're saying we should make the safaridriver worse? ;)

No, just underscoring how very little interest I have in touching that code :)

It wasn't that bad to work on, just got to the point where all that was left was the PITA hard stuff.

Honestly don't know how jimevans has stuck it out working on the iedriver for so long!

*someone at the conference told met that the secret to success was to implement something poorly and get other people frustrated enough to write code that fixes it :)

sounds like something simon would say

titusfortner: that sounds very socially responsible

Someone else told me that they considered most people in open source projects to have about 3 years of work, and that the secret to an open source project's success is to make sure you get new members excited every year or so to maintain momentum

until you realize nobody else gives a damn about the thing you produce, and if you left it, nobody would take it over.

they complain enough for not giving a damn ;-)

"but it's too haaaaaard!"

</whine>

I honestly don't know if it possible to build 1) RC only server and 2) WebDriver only server from the same code base

Should be

Well, grid complicates things

I'm back to the old conversation with simonstewart: https://gist.github.com/barancev/1323599b491639...

So we're stuck with RC? :)

:log

"The old server should not have the webdriver stuff in it" -- I'm still in doubt we need this

A selenium server based on netty is not possible with grid - which uses servlets

hey goes, how goes it

netty is just a wish :)

let's keep jetty for now

are we talking S3 ?

yes, S3

I'm going to take another try

well... as long as we get the "new server" good to go, does the old server still present a roadblock to release?

simonstewart wants "the old server" for some reason

as in, does pulling the webdriver stuff out need to hold back a new release

but if that causes a significantly increased amount of work... :)

exactly

that's what I dislike in this idea

seems like focusing on the 10% part of the problem, when the 90% is getting RC gone

let me channel simonstewart for a minute.

I should talk to him again

just dropping RC completely, with no recourse is a really nice way to say "f-u" to enterprise users who don't update

recourse is "use version 2.47"

+1

that's an incredibly rude answer to users.

Are we going to make any changes to RC code once it's split from webdriver?

* that would warrant more than a bug release

not that it's the wrong thing to do, but it's really, really impolite

I don't follow

Am I missing the point of Selenium 3?

jimevans: anyway, they can't use 3.x because it is "no RC" branch

How is it impolite to say 2.47.x is your same old server with RC support. 3.x is webdriver only

i'm just (poorly) giving you simon's arguments.

was 3 going to remove all S1 classes? accoding to the release plan, we were just going to add deprecation warnings i think

i'd like nothing more than to just get rid of it altogether.

we can create a "legacy branch", if someone is eager to fix RC issues :)

major release means major changes.

+1

definitely

and get with the fucking times.

no backwards compat

with s1 at least

No More RC! :)

but again, simon has definite views on that, and as long as he has the commit bit, there's every chance he'll revert any too-intrusive commit.

he's stated (to me) that he's not opposed to rolling stuff back that he finds offensive, for whatever reason

Not too hard to get Simon to agree to changes

well, it seems like the first step is to get the WebDriver only portion ready. Then if we want to hold up a release to get an RC-only something for people to use, that's a future decision :)

Just need to get a few drinks in him first

(says someone who updates to Se3 by deleting 2 directyories)

titusfortner: done for the client part (in Java)

random off topic question: how is MS Edge on support for ES6?

right, I know, I'm saying my part is easy

but it is not easy to perform the same with the grid...

What is ES6?

titusfortner: do you have the selenium3 roadmap? if not i can share it with you

The latest fast ring build with the latest released driver (yesterday)

is great

except for alerts

clicking something that opens an alert hangs the driver

if "removing RC support from the core download" is Selenium 3, .NET could cut the release today.

sircapsalot: we have a roadmap?

check your email

Why is Python different with the deprecation warnings?

rather than just deleting them

not sure

it's the only bindings that discuss it is all

right.. i think that's more of an "understood" that s1 is going away in all bindings

I have tests that I've tested with Firefox and Chrome. is there a different process to run the same test for Internet Explorer ?

qwebirc57849: what do you mean? have you tried it? if you have, what didn't work? that question, as asked, is far too broad to give a concise, specific answer.

jimevans: I'm not too familiar with Selenium. I just wanna know if the test should be ran with the The Internet Explorer Driver Server

yes. just like you use chromedriver.exe to run tests against chrome, you similarly use IEDriverServer.exe to run tests against IE. you will need to properly (manually) configure IE before running your tests against IE.

:wiki internetexplorerdriver

1: https://github.com/seleniumhq/selenium/wiki/int...

2: https://github.com/seleniumhq/selenium/wiki/int...

qwebirc57849: first link there ^^ has a section on required configuration.

jimevans: Thank you. I don't need to re-write the test correct ?

in general, no. however, IE tends to run a little slower than firefox and chrome, so you may have synchronization issues that you need to sort out.

titusfortner: i think i might have asked this before, but have you ever had the edgedriver hang when creating a new session?

that's happening to me all the time, on multiple machines, but it doesn't seem like it happens to anybody else

dammit it's leo's fault we get all the 'qwebirc*' handles in here... https://github.com/SeleniumHQ/www.seleniumhq.or...

i was wondering how the heck people were getting those nicks

for the first few days, i didn't even notice the numbers were different

i was just thinking, wow, this qwebirc person sure has a lot of questions

heh ;)

[13www.seleniumhq.org] 15lukeis pushed 1 new commit to 06master: 02http://git.io/vcsr2

13www.seleniumhq.org/06master 1464b5dcf 15Luke Inman-Semerau: don't default users to the random anonymous handle

there… we should see less of that in the near future… except for those who bookmarked it

:)

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