02:15
<lhb>
morning, anyone awake
02:31
<lhb>
mmm.. obviously not
02:33
<lhb>
i have been watching a webrat tutorial
02:33
<lhb>
things dont always work like in the tutorial :(
02:34
<lhb>
simple login screen... why do i get this error
02:34
<lhb>
You are being redirected. <false> is not true.
02:34
<lhb>
am i missing a step?
03:00
<manveru>
that looks like a very useful error :)
03:23
<lhb>
mmm... always hard to continue someone elses work
03:24
<lhb>
solved... got a new failure... user didnt exists
05:03
raldred_away joined
10:06
technicalpickles joined
11:52
thewoolleyman joined
14:12
technicalpickles joined
15:26
<atmos>
has the syntax changed for have_selector ?
15:45
gaffo_laptop joined
16:22
technicalpickles joined
16:33
<atmos>
is there an easy way to leave my selenium rc server running ?
16:45
technicalpickles joined
17:08
technicalpickles joined
21:12
<snusnu>
hey mabes_ .. thx for pulling my fork in!
21:13
<snusnu>
now if only i were a better git user and knew why the github fork queue thinks that your additional commits fail to apply :-)
21:47
technicalpickles joined
21:49
<mabes_>
snusnu: np, thanks for adding it! I always had plans to add DM support but haven't really messed with DM in a long time
23:01
<snusnu>
mabes_: have a look at my latest commit ... your gemspec had TODO in it's description again, which doesn't make (my?) rake happy ... i added a little explanation to the commit message
23:02
<snusnu>
mabes_: http://github.com/snusnu/database_cleaner/commit/cf981bfd081af6af45c53508bfd86b360c3a4360
23:03
<mabes_>
snusnu: ahh, thanks. I'll merge it in in a minute
23:03
<snusnu>
mabes_: thx
23:21
<zdennis>
hey mabes_ you still around?
23:21
<mabes>
zdennis: yep
23:22
<zdennis>
i msg'd brynary last night about this, but since you're on i'd like to your thoughts to
23:22
<mabes>
snusnu: looks like I need to update my jeweler settings- i'll do that now, thanks for pointing it out
23:22
<zdennis>
i've started giving webrat a better automated suite for selenium functionality
23:22
<zdennis>
I've spiked out an approach, do you have a minute to check out a spec?
23:23
<zdennis>
pull up check_spec.rb locally and then pull up http://github.com/zdennis/webrat/blob/bd5fa6432cab708d91f96e79165fa27e4015daea/spec/public/check_spec.rb
23:23
<mabes>
are you the first person to do this BTW? i though gaffo_work or someone was wrapping a suite around the selenium stuff..
23:24
<zdennis>
i haven't seen anything from gaffo, and I don't see a branch on his github account
23:24
<zdennis>
or even a webrat fork
23:25
<zdennis>
if he's got something that's great, but if not, we need to get selenium support moving along
23:25
<zdennis>
I've been waiting for months :)
23:27
<zdennis>
wdyt of the updated check_spec?
23:27
<tomoj>
trying to get selenium to work.. it boots up a browser, and then appears to hang. my Then step doesn't get executed
23:27
<tomoj>
I read on the cucumber wiki that firefox 3 has issues and needs a patched jar, but that was for the Selenium gem, and I can't find a jar to patch in selenium client
23:27
<tomoj>
anyone else seen this?
23:28
<zdennis>
tomoj: you get a timeout error?
23:28
<mabes>
tomoj: the jar that ships with webrat should work
23:28
<tomoj>
no it just sits there
23:28
<zdennis>
tomoj: does everything in the browser runs up until your Then step?
23:28
<tomoj>
it waits for the selenium RC and rails app servers, says Ready! for both, and then does nothing
23:29
<zdennis>
ah, check to see if you have a mongrel_rails process running
23:29
<tomoj>
actually my When I go to the homepage doesn't get executed either
23:29
<zdennis>
if you are running with rails
23:29
<tomoj>
yeah, mongrel's running
23:29
<tomoj>
and the app shows up in the browser selenium starts
23:29
<tomoj>
it just never goes on with the scenario
23:31
<mabes>
zdennis: so, did you add simulated and automated or did that already exist?
23:32
<mabes>
in the specs
23:32
<zdennis>
mabes: I added
23:32
<zdennis>
mabes: that spec actually runs right now too
23:32
<zdennis>
it's not just pseudo-code spike
23:32
<zdennis>
although all of the setup changes is definitely "spike" code
23:32
<mabes>
I like the use of context too BTW
23:33
<zdennis>
tomoj: what's the last thing your scenario does successfully? And what's the first thing it hangs on?
23:33
<zdennis>
you may need to use a debugger, or some put statements to find out
23:34
<mabes>
zdennis: is there anything in particular you wanted me to look at, or just the general approach?
23:34
<zdennis>
mabes: just in general
23:35
<zdennis>
mabes: if it looks good to those involved in webrat i want to really push it and get it done and contributed back
23:35
<zdennis>
because the selenium support needs love, but I know its hard to add w/o good regression and coverage it actually works
23:35
<mabes>
yeah, I like it. but I'm not that regular of a contributor to webrat
23:36
<zdennis>
that's okay, i value your opinion, and i know others in the community do as well
23:36
<mabes>
one question.. on the automated and simulated methods
23:36
<mabes>
i don't think you need the instance eval
23:36
<tomoj>
zdennis: it runs right up to the 'visit' in the first step
23:37
<tomoj>
it looks like it starts firefox once, closes it, and starts it again.. is that normal?
23:37
<zdennis>
tomoj: visit is the first thing you do that would try to tal k to selenium?
23:37
<tomoj>
zdennis: yeah
23:37
<tomoj>
all I'm doing is When I go to the homepage I should see "Hello!", and it's just the public.html
23:37
<zdennis>
mabes: just remove it completely, with a block.call ?
23:38
<mabes>
hmm. well I didn't see the automate block.. I was thinking that you could just do
23:38
<zdennis>
or pass it in to: simulate &blk
23:38
<mabes>
it(desc, &block)
23:38
<tomoj>
the backtrace when I kill it includes several "in `timeout'"s
23:38
<mabes>
but, it has the simulate and automate block.. hmm
23:39
<mabes>
maybe you have to do an instance_eval
23:40
<zdennis>
tomoj: have you ran it multiple times and has it hung multiple times? if not, run it again to see if the behaviour persists.
23:40
<mabes>
zdennis: the other suggestion I would make is to make it easy to turn off the selenium specs and just run the automated ones. like so: http://gist.github.com/109113
23:41
<mabes>
er.. just run the simulated ones
23:41
<tomoj>
zdennis: yup, happens every time. I also killed my already running firefox, and it does the same thing
23:41
<mabes>
since selenium is.. uh.. slow :)
23:41
<zdennis>
mabes: yeah i totally agree with you on that, will make sure that exists before request a patch back in from brynary or gaffo_work
23:41
<mabes>
zdennis: you weren't at railsconf were you?
23:42
<zdennis>
mabes: no, i went to JSConf a few weeks ago, 3 colleagues went though
23:42
<zdennis>
did you go?
23:42
<mabes>
ok, just wanted to make sure I didn't miss a chance to meet you in person. yeah, I went
23:42
<zdennis>
mabes: you hitting rubyconf later this year?
23:43
<mabes>
I've exhausted my conference budget for the year... I submitted a proposal to LSRC though so I may be going to that
23:43
<zdennis>
tomoj: do you have any mongrel_rails or selenium processes running right now? (Now that you have stopped your scenarios?)
23:43
<mabes>
but I don't think I can make it to rubyconf due to the budget concerns
23:44
<tomoj>
zdennis: if selenium processes can be grepped for 'selenium', no
23:45
<zdennis>
mabes: ah, i don't think i'm going to make that one, august is too early for our conference budget. We're hoping rubyconf is far enough away to replenish the funds
23:45
<zdennis>
mabes: good luck with the submission
23:46
<zdennis>
tomoj: are you running webrat 0.4.4?
23:46
<tomoj>
zdennis: yep
23:47
<tomoj>
tried setting the browser_key to safari, it does the same thing
23:48
<zdennis>
tomoj: can you put a debugger before your visit line that hangs?
23:49
<zdennis>
and then look at log/selenium.log and log/mongrel.log to see if any errors or warnings are printed
23:49
<zdennis>
also, while in debugger, type "ARGV.clear" <enter>, and then go into irb mode, and type visit "/"
23:50
<zdennis>
to see if selenium hangs on that or responds, also, look to see if anything additional was printed to selenium.log or mongrel.log
23:50
<zdennis>
If nothing shows up, before leaving the debugger, look in the selenium control window, and click "Show Log" to view the in browser log
23:50
<zdennis>
see if any errors show up in thee
23:55
<tomoj>
when I visit "/", selenium starts up, but the feature still hangs
23:55
<tomoj>
no errors in the selenium control window's log
23:55
<tomoj>
no errors in mongrel.log either. I don't seem to have a selenium.log
23:56
<zdennis>
tomoj: right now is your site pulled up in the firefox window? Or is it just white?
23:56
<tomoj>
the site is there, it runs the javascript successfully (just did a document.write to test selenium)
23:57
<zdennis>
tomoj: are you using default webrat settings, or do you change the port that selenium or the app run on? (besides 3001 and 4444)
23:58
<tomoj>
it successfully goes to the url I put in visit, but then does nothing
23:58
<tomoj>
default webrat settings
23:58
<zdennis>
ah, the visit did work
23:59
<zdennis>
is your debugger session hanging right now from the last visit call?
23:59
<zdennis>
or are you back at a prompt
23:59
<tomoj>
it hangs, no prompt
23:59
<tomoj>
last thing selenium does is open(/)