where can I address death threats to people developing phantomJS?
hazmeister-sl joined the channel
hazmeister-sl
why would you do that?
:phantomjs
selbot2
No users will ever visit your site using a browser even approximating phantomjs. While phantomjs is based on a rendering engine used by one of the major desktop browsers (webkit), it's not the current version, and more importantly, the javascript engine is completely different.
GodFather_ joined the channel
Numline1
hazmeister-sl meh, I just inherited some stuff that uses phantomJS to take snapshots into PDFs
for whatever reason it always generates second page with some random stuff that shouldn't ever be there
hazmeister-sl shudders
hazmeister-sl
phantomjs is being depreacted
for reasons such as the bugs you describe
checkout headless chrome
Numline1
is it really?
hazmeister-sl
which does a similar job
Numline1
yeah, we're using headless chrome for other tasks
hazmeister-sl
but is nearly full on chrome under the hood
Numline1
I kinda hacked around Laravel Dusk to use it, this is just full legacy trash
basically it grew legs, and pretending to be a browser is hard
Numline1
hazmeister-sl makes sense, headless chrome is great. Even regular Xvfb Chrome was okay, if it worked
no reason to use something custom
it's like using Netscape
hazmeister-sl
i see no benefit in your scenario
Numline1
hazmeister-sl what do you mean?
hazmeister-sl
in using phantomjs
you’ve identified an area where it performs badly
Numline1
hazmeister-sl yeah, the thing is, reworking this part would take a while
it generally works as it is
it just does one thing it shouldn't
sadly it just being "bad" doesn't grant me a budget to rework it from scratch
kyuwonchoi joined the channel
hazmeister-sl
are you using webdriver, or talking to phantom direct?
because webdriver would let you drop in a replacement browser
GodFather_ has quit
Faucelme joined the channel
stryx` joined the channel
mfsi_ankitm joined the channel
Numline1
hazmeister-sl we're using WebDriver for the part that uses headless chrome
the part using phantom is using a PHP/LAravel package from Spatie
hazmeister-sl has quit
saphyel joined the channel
SitamJana joined the channel
Faucelme has quit
tsglove has quit
tsglove joined the channel
moldy joined the channel
moldy
hi
SitamJana has quit
TacticSiege has quit
SitamJana joined the channel
TacticSiege joined the channel
KaibutsuX joined the channel
LastWhisper____ has quit
dude-x
:ask
selbot2
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 :)
dude-x
hello!
loophole
I need to construct the driver Object without actually launching a browser or communicating with selenium-server (I don't have the luxury in this case of waiting for the Object by communicating with selenium-server and launching marionette/ff). I had asked yesterday how to switch the session_id for a given driver object and it's not officially supported from the API, but I spent yesterday hacking around and I have successfully switched the session_id and url t
dude-x
in python?
loophole
ruby
dude-x
that's pretty much it
loophole
Would have been easier in python
dude-x
why's that?
loophole
There were examples of people switching the session_id... things aren't as private
dude-x
oh that's right ruby enforces private attribute access right?
dude-x: just because python is easier to make hacky things with doesnt mean its better :P
loophole
:)
dude-x
python runs faster than ruby too
lucast
i'd say it's easier to monkeypatch ruby than python
WhereIsMySpoon
dude-x: I’d contest that and say it’s heavily dependent on what you’re doing :)
supersri
I am trying to upload a large file using selenium , using `browser.setFileDetector(new remote.FileDetector());, remote.FileDetector() fails with OOM if the file is too large`https://github.com/seleniumhq/selenium-go...
is there any other workaround to upload a large file using selenium?
dude-x
let an http client do the uploading
mfsi_ankitm has quit
supersri
@dude-x you mean like a REST request?
@dude-x you mean like a REST call i mean?
dude-x
i mean like a HTTP POST
supersri
yeah my bad , i get it , thanks let me give it a try
Cheers
LastWhisper____ joined the channel
supersri has quit
kggr has quit
lukeis joined the channel
stryx` joined the channel
michl68 has quit
Orphis joined the channel
hazmeister-sl joined the channel
hazmeister-sl
Running my tests through a software proxy on IE
should IE be reverting proxy changes when the test finishes?
ddavison-sl has quit
WhereIsMySpoon
hazmeister-sl: what do you mean
chele has quit
hazmeister-sl
so I’m using BrowserMob Proxy to assert certain network requests are sent
chrome creates disposable profiles, so, the proxy changes are only for that individual profile
but when you change the browser proxy on IE, it changes the setting for the rest of Windows
which is mildly annoying, as the software proxy disappears when my automated test ends
I could workaround it, but I’m wondering if it should be doing that in the first place
I go to manually verify some bits in IE, and I get told the proxy no longer exists
Lio_ joined the channel
Lio_ has left the channel
pnaranja-sl joined the channel
lucast
I don't think IE is sandboxed
WhereIsMySpoon
hazmeister-sl: yeah I don’t think there’s really any way around that short of putting it all in a container and proxying the container
dude-x
hazmeister-sl in general changes to IE are "permanent"
becuse there's no private profile created
hazmeister-sl
cool, I thought as much
I guess I could spin up a new webdriver at the end of the test with no proxy