#selenium

anyone know how to expose the chromedriver log in ruby? I can only find info from 2013 which is not working well for me

tried both of the following

```:service_log_path => LOG_FILE``

and

tried the all languages portion of this: https://sites.google.com/a/chromium.org/chromed... (via watir)

it goes in driver_opts now

we split up service commands from browser commands

[13www.seleniumhq.org] 15User253489 opened pull request #179: Fix link configurations and technical expressions (06master...06master) 02https://git.io/vN9Gh

I'm trying to automate the screenshotting of elements with selenium, but unlike chrome I can't get firefox to even start with webdriver+python. after requesting the firefox driver, the browser instance starts, but the code hangs for several seconds before failing out with a connection refused error

all the details, with code, errors, and versions are here: https://gist.github.com/AbstractBeliefs/04543b7...

does anyone have suggestions on where to go from here?

Can you update your Firefox version?

Your version looks fairly old, the latest version is 58

im on debian so nominally "no", but ill have a poke around!

no, neither firefox nor the debian licensing version iceweasel packages have updated versions in the backports repo

but thank you, its another avenue to explore down :)

it's frustrating that chrome doesn't express the element-screenshot endpoint for webdriver - im looking at implementing it but i understand that developing anything for chromium is a herculean effort

i am seeing issues with calling driver.get using Chrome 64 and chromedriver 2.35 or 2.34 using the python bindings. it's a bit bizarre when connecting to an IP address

[13selenium] 15dnknitro opened pull request #5406: Added no-cache header to HTTP GET command of RemoteWebDriver (06master...06FixRemoteWebDriverCacheCommands) 02https://git.io/vN92D

Hello! I would like to understand what is the difference between visibility_of(element) and visibility_of_element_located(locator)

why do we need visibility_of(element) if visibility_of_element_located(locator) is enough

different parameter types (element vs locator)

but in order to find element I still need to use locator, right?

why do I need to do 2 steps: find element by locator and then place it to visibility_of(element)

yes. but later some other element can be found by the same locator

whereas you want to be visible this exact element found by a locator some time ago

ok. now it is clear. thank you very much.

@martynas congrats :)

@jackofspaces I have another pattern for you

Added an image: https://i.imgur.com/DghAPA0.png with comment: "any work around on how can I install selenium on mac firefox"

use Firefox ESR

or install this version: 54.0.1

thanks :) you saved my day https://www.mozilla.org/en-US/firefox/organizat...

bytheway I'm just new to Selenium I'm trying to learn Selenium IDE

then Selenium Web Driver in the future, my question to you guys is if you have little or no programming background, Will it be hard to learn Selenium Web Driver?

It will be hard, but it is worth the effort

once you get the basics, webdriver isn’t hard

but you will have to build some infrastructure around running your tests, which can be tricky if you haven’t done something like it before

well worth it though, and usually you can find a developer to pair with who can help

jerome-sl

a python or java developer! xP jerome-sl

how about PHP?

if is your main programming language, don’t think more

not my main programming language but in our company its PHP that we use on our ecommerce platform

I'm not sure if I should focus on PHP or learn Java/Python

i work with php e-commerce too and my main is Python

for b2b tests is possible

nice, I asked it because it be easier for to learn python than PHP

*might be easier I mean

sure!

if you need something ask me

I’ll try to help u

I have an issue where an iCheck js lib component is obscuring the element I want to click.

Are there any good solutions/workarounds to it?

`Element input id="inp-en-US" name="language" type="radio"> is not clickable at point (81,474.0666809082031) because another element <ins class="iCheck-helper" obscures it`

My selector works in browser console (with click()). And I can click in FF by adding the "ins" to my selector. However, in Chrome it fails with "not clickable" error.

jimboslice-sl what is iCheck?

and do you test it at all?

It's a helper lib for JS

hazmeister-sl

The other question I don't understand.

it looks like it styles form controls

so you don’t want to click the input

but the div which is copying the state of the input

Yes, it's a radio button I'm trying to click

use the browser’s dev tools

The surrounding div?

quite possibly

it will take some trial and error

but some form controls have limited styling options

Ok, cool, thanks a lot! I will try it.

so some frameworks draw the form control as a div

the input is in the dom, but its hidden (or underneath)

I see.

Right.

you might see an event handler if the browser inspector if you are lucky

but there is usually an obvious pattern once you figure it out

in this example, it looks like you might to click the ins tag

I’ve never seen an ins before, but that’s what the inspector is finding

Ok, we're using react. So this jquery lib is shoehorned in. There is a plan to replace it.

react is a whole ’nother ball game

Right. So clicking the ins works in FF. But not in Chrome

as devs stop using classes and start using react properties, its get harder to automate

hmmm, try the div parent

firefox might be implicitly clicking the div in the way

I will. I just have to figure out a good way of finding it. I have 10 radio buttons (language selector) and the unique identifier is in the input. So I have to find the parent. :pensive:

I don’t often use xpath, but there is a good parent selector you can use

Yeah, I'll use that if I can't solve it without.