I’m following through chapter 8 and all seems to be working as expected except that the page saved and displayed by capybara (pg 150) is blank. If I point my browser at the page then I see the form as expected, but there is nothing on the saved page. I can reproduce this by using the sample code (code/support_code/18) and just making a simple change to the feature file so that the test fails. Any ideas what I can change to see the page as Capybarra sees it?
regards
Iain
That’s puzzling.
Can you open the HTML file in an editor, or view source on it and see what it contains?
Just a single line:
<!DOCTYPE html PUBLIC “-//W3C//DTD HTML 4.0 Transitional//EN” “http://www.w3.org/TR/REC-html40/loose.dtd”>
Right.
My only guess is that there’s something wrong in your scenario such that Capybara hasn’t actually seen anything yet. Maybe you’re missing a call to @#visit@ somewhere?
One thing you could try is to add the ~~~ @javascript ~~~ tag to the scenario, which should cause it to open up in Firefox as it runs. That might give you some more clues as to what’s happening.
I downloaded a fresh copy of support_code/18, edited the feature file so it would fail, and added the @ javascript tag. Firefox flashes up a window very briefly which looks like it contains the form as expected but the call to save_and_open_page just produces a blank page as before. All that is different from the original source is the first and last lines of the feature file. Here’s my version:
@javascript Feature: Cash Withdrawal Scenario: Successful withdrawal from an account in credit Given my account has been credited with $100 When I withdraw $20 Then $20 should be dispensed And the balance of my account should be $81
If this is a Javascript scenario then it sounds like a race condition between the driver and the “browser”. You could put a @sleep 1@ in before the @save_and_open_page@ and see if it affects the output, or you could put in a @debugger@ statement (requires the debugger or ruby-debug gems) and see what the @page.body@ is at that time (the debugger statement is generally slow enough to allow the page to finish loading)
Thanks Jon. I will try this when I can get it working again. The new version of Firefox doesn’t seem to want to play with selenium so at the moment I can get no further than an error message saying “Unable to establish a stable firefox connection in 60 seconds” This looks like a recurring problem with selenium and new versions of firefox :-/