#selenium

Well, looks like webpack really

Because I invoke webpack by calling yarn

@p0deje just the name

always loev a good pun

windows ships with robocopy too

ah, it's a good one :)

simonstewart oh, configuring webpack is fun, they say

Hi !

The problem I'm wrestling with is that the atoms source is dependent on files in two different directories

With Buck, I'd just declare a `closure_library` and set up dependencies for the libraries, then output a single file using `closure_binary` which would include an optimally small subset of source files

Time to go read the webpack docs, I guess

And now I'm here: http://2ality.com/2015/12/webpack-tree-shaking....

The closure compiler does this out of the box

Reading webpack docs now. Turns out that the atoms don't have a normal "entry" point.

you need exports, right?

Dunno.

Depends on how we're using this

I'm facing issues with latest selenium server and IEdriver in hub mode. I'm getting Element not found each time I try to find an element by id. I've seen that behavior has changed in IEdriver changelog, but I've just updated from 3.5.1 to 3.6..

In the ideal world, folks would just "import" the module that they needed from the atoms, and we'd compile the individual components that way

But we appear to be starting from an already compiled snapshot of an old IDE build

So I have no idea what we're trying to actually do here

Right now, we blindly copy all source files over to the web extension. That feels incorrect

@tomer feel free to join this conversation :)

I'm not good at those things

but I guess we either need to pull in atoms code via git submodule + exports + webpack

p0deje: we're using "peru" to manage that bit of the process

The git submodule bit

oh, looks great

It'll do

It's a hack

https://github.com/roman01la/webpack-closure-co... can it be used?

Apparently

ave, simonstewart!

G'day barancev|2!

I have a style question

Go

for

it

:)

Guys what do you prefer for Selenium? Python or Java?

Me? Java, natch

But I'm biased

Hmmm i prefer python

InternetExplorerOptions.requireWindowFocus() vs e.g. FirefoxOptions.setAcceptInsecureCerts(boolean) -- what style should we prefer for boolean options?

m-sl: ij

ok

Damn off-by-one errors

Java Bean style suggests "get/set", with the common addition of "has" for boolean values

For a builder, I'm less inclined to go with that

I think I prefer being able to explicitly set the value to `true` or `false`

But we could do the hybrid thing: `FirefoxOptions.acceptInsecureCerts(boolean)`

The options tend to be write-only

so every method is implicitly a "setter"

I just want to make all options style uniform

Yeah

I hear you

hybrid style, I think

Your thoughts?

sounds good

renaming, deprecating...

@p0deje @simonstewart I guess my reasoning is that I don’t want to break the extension

As long as we can replace small parts of it without breaking we’re moving forward

no?

Is there any known problem with latest selenium hub with latest IEDriver ? Running without hub, everything is fine, running in hub unable to get any element (I've tried by id and by class). Here is a simple test case in PHP, but it's simple enough to convert it in any language. I'm facing this issue with IE11, Selenium server 3.6, IE Driver 3.6

And here is the error message : InvalidArgumentException: Element not found.

simonstewart: also I'm going to deprecate and delete methods like SafariOptions.getUseCleanSession(), all the getters

And of course, same code is ok with chrome and my Internet option are the same for all mode

Anyone Around? Quick question about - driver.findElement(By.xpath(

Kabone: :ask

If you have a question, please just ask it. Don't look for topic experts. Don't ask to ask. Don't PM. Don't ask if people are awake, or in the mood to help. Just ask the question straight out, and stick around. We'll get to it eventually :)

Alright

pauluchon: 1) you’re sure the id is ‘loginFrm_username’? not the name or some other attr? 2) is it maybe a timing issue?

So basicly - I am trying to use - driver.findElement(By.xpath("//*[@id='i...; to type in an email address on a page. The element is <input id="inputEmailHandle" name="inputEmailHandle" value="" required="" type="text">

WhereIsMySpoon: 1) Yes sure, 2) I've tried with a basic sleep and no change

Kabone: looks good so far

whats teh issue

pauluchon: oh, its IE issue only?

When it tries it causes an eception

i missed that

This is on firefox

Kabone: can you :gist the exception?

Please paste >3 lines of code or text to https://gist.github.com and share the resulting URL

Exception in thread "main" java.lang.NullPointerException at com.package.Main.main(Main.java:251)

also can you :clarify and maybe post your whole test code in a :gist

Please paste >3 lines of code or text to https://gist.github.com and share the resulting URL

Please clarify: Are you using WebDriver, RC or IDE? What version of Selenium? What programming language? What browser and browser version? What operating system?

sure one sec

WhereIsMySpoon: Yes IE only..

ignore the first question :) i know you’re using webdriver

pauluchon: I would submit it as a bug then if it looks like an IE only issue, and if you can attach the driver logs

if you’re not sure how to do that dont worry

sounds like a bug to me

Kabone: which line is the NPE on?

Ok thank's WhereIsMySpoon

that gist doesnt have 251+ lines on :)

on that - 26

Kabone: are you sure username isnt null?

yes

you really just need to debug and step through to that line

before i print it out

then inspect each part of the line

inspect driver, inspect driver.findElement(…)

That's what I have bee trying to do. Originally I couldn't get it loading custom when I was not loading the driver as I am now it was working

well clearly something on that line is null

its either the findelement or the username

it's the element

though I would expect findelement to throw rather than return null

if it didnt exist

but I have tried adding waits, expected

but it does

throw an elementnotfound exception

or whatever it is

not an NPE

ok