#selenium

I'm specifically looking to log HTTP requests and responses. Do you think this is something log4n is capable of doing?

sandramedina-sl: you want to do this from within selenium tests?

sandramedina-sl: log4n is for logging to something or even multiple things at a time, not for gathering logs

(and the follow up question is, "why?")

Yes. I'd like my tests to run, and for every page it hits, I'd like to get the HTTP response. I'm testing many pages (takes multiple hours to run the test) and sometimes certain pages fail, so I'd like to see the header response in a log when I come in the next day

sandramedina-sl: so, then, you know that selenium doesn't give you access to the http response, right? and that there's no cross-browser way to retrieve that from the browser, right?

jimevans: g'day. I feel like stalking you, but there was a question basicly why you don't promote page object factory, and what are consequences of page object factory not compliant with .net core despite portability (does this approach affects e.g. performance?)

GrzegorzSwiecPL: i don't promote it because it offers no value whatsoever.

there is no value in identifying elements via attribute.

none.

zero.,

zip.

nada.

bupkis.

:)

How can I know if two elements have the same dom in them?:/

CtrlC: you can't. i mean, you might compare their innerHtml values, but that doesn't mean they're the same element.

jimevans: making sensible logging is way easier in page object factory

The problem is they don't even have a innerHTML value. It's android.

GrzegorzSwiecPL: how so?

GrzegorzSwiecPL: i don't believe that to be the case, and it's an extraordinary claim, so you'll need extraordinary evidence to convince me.

CtrlC: you asked earlier how to tell if two elements are "equal." that's not the same as having the same DOM.

ok, will prepare snippet code for you tommorow :)

@jimevans I did hear that, but someone on my team said that Selenium used to give access to the responses and that I should try with Selenium first. This seems inaccurate to me based on what I have read here: http://jimevansmusic.blogspot.com/2012/07/webdr...

idk man, I'm looking for anything to cling to.

sandramedina-sl: given that i wrote that article, i stand by its content completely.

:)

hahahah

perfect

however, you *can* use a proxy in conjunction with selenium to give you that information.

elsewhere within that blog is an example of how to do that.

right. I'll jump right into trying to set that up then, instead of messing with the (non existent) logging capabilities

jimevans: despite code snippet i'm going to convince you with. Why would you spend hours on developing that factory, thinking so?

came late with that conclusion ?:)

why? because people looked at the java implementation and complained, "but .NET doesn't have everything the same as java!"

that's why.

i've never been sold on it as a solution

got it :)

and i've never advocated that anything in WebDriver.Support.dll be used as anything more than example code (with the exception of WebDriverWait)

so maybe you should advocate the other way round. There is many web pages on page object factory as good example, none with the other opinion.

I'm feeling like i'm trying to overcomplicate our testing framework atm :)

it's on my ever-increasing to-do list.

I'm even wondering if any additional logging during test execution is needed now

using attributes (or annotations in java) to define locators in a page object just seems like asking for trouble and difficulty in debugging.

I mean, you have a stacktrace once it fails

why bothering when its green

not to mention incredibly difficult to maintain

I agree with debugging - more difficult

Hi

I need some information, in which selenium standalone server supports firefox version ?

trying again. Anyone seeing problems with using sendKeys(filepath) to a file input for upload? I’m getting File Not Found errors. But only with Firefox 55. 54 is fine withe same code. Chromedriver is fine with the same code. seen with Se 3.5.2 and 3.4.0.

I need some information, in which selenium standalone server supports which firefox version ?

pls help me om this

woodybrood-sl: update to 3.5.3

suman_: selenium 3.5, geckodriver 0.18, 55+

FF 55+

i thought I had tried that too. I’ll give it another go in a little bit.

ok, thanks

Hi, I'm trying to update my selenium to 3.4 and am running into this error: java.lang.NoSuchMethodError: org.openqa.selenium.support.ui.WebDriverWait.until

I've tried adding guava 21.0 to my maven dependencies, but it still doesn't work. Anyone have any tips?

Guest92707: upgrade to selenium 3.5.3 and guava 23

I just tried that and it still doesn't work. I'm afraid I'm adding the dependency wrong? or perhaps it is conflicting with one of my other dependencies

Hi can anyone help me with this error: java.lang.NoSuchMethodError: org.openqa.selenium.support.ui.WebDriverWait.until

Yeah. The recent updates triggered some guava hell. I realized that I was only specifying Selenium-server in my POM file. When I looked at dependencies, then it showed I was using the version i wanted for that, but I had 3.5.2 for the other selenium libraries. I’ve since changed my pom to specify all of the libraries. I’m also excluding all other sources of guava.

looking at my dependencies through eclipse, I see a (managed:#.#.#) next to each dependency do you know what this means?

[13selenium] 15barancev pushed 1 new commit to 06master: 02https://git.io/v5cGz

13selenium/06master 14a38aec8 15Alexei Barantsev: Cleaning required capabilities out of tests, step 1: deleting tests directly checking required capabilities. They were enabled in legacy Firefox driver only because there no other driver implemented required capabilities.

Hi people, i'm currently creating a dockerized selenium grid

For comands involving element search i'm receiving responses like this from the nodes {"ELEMENT":"0.773277606074561-2"}

are this elements id valid for a node response?

Yeah - they are, I see the same when running the things on Browserstack

[13selenium] 15barancev pushed 1 new commit to 06master: 02https://git.io/v5cnB

13selenium/06master 1405bc5bf 15Alexei Barantsev: Cleaning required capabilities out of tests, step 2: driver suppliers should not use driver constructors that accept required capabilities

Is this a known issue with Java bindings 3.5.3 and GeckoDriver 0.18.0: 'InvalidArgumentException: direct was not a valid proxyType value'

running python scripts on Ubuntu. I am using Firefox and geckodriver. I am noticing I have a lot (>10) processes of Firefox running without any script. I do a .quit() at the end of the script. How do I get Firefox to go away?

I think I found a code bug where it skipped the .quit()