Added check for multi-result pages to prevent null exception error.

[CBXZero]

First time I attempted the squash, I followed a stackoverflow answer that was misleading. This should be good to go and is all bundled in 1 commit :panda_face:

[davidchambers]

Thank you for persevering despite your version control troubles. :)

Please git commit --amend to give the commit a sensible message. You'll then need to run git push origin master --force to update this pull request.

[CBXZero]

Updated the commit and pull request title to be informative :)

[CBXZero]

That "and" looked weird to me and the build test failed, if this build doesn't work I'll fix it when I can get to my laptop

[davidchambers]

Tests are failing. ;)

[CBXZero]

Looking at it on my dev machine now, I'll have it fixed here in a sec :) sorry for the flux of emails you must be getting. D:

[CBXZero]

Got it that time. I don't know why I thought that "and" looked funny. Then again, I've been studying Statistics all night and that might be messing with me. Lmk if there are any other changes you'd like done. :)

[CBXZero]

I did some looking into this, and 2 things stood out to me.

1. I modified the javascript library by hand that my bot was using, and told it to just spit out the html of the file that it reads, and I noticed it hits the else block that is highlighted above multiple times (the file saves 3 times to be exact), and was wondering if you had any clues to that.
2. The and is incorrect, it should be an || body.indexOf(string) >= 0 . That being said, if I actually switch it to that, I get 27 tests that break, mostly involving an Uncaught AssertionError: {} === null, which makes me think the double sided cards are now getting caught on that else block.

I was wondering if there were any ideas that you had that might be causing the issues I'm running into. I'll investigate this issue more tomorrow.

[davidchambers]

I suggest looking at the differences between the responses to the following requests:

$ curl --silent http://gatherer.wizards.com/Pages/Card/Details.aspx?name=fizzbuzzldspla

$ curl --silent http://gatherer.wizards.com/Pages/Card/Details.aspx?name=test

[CBXZero]

I'm still looking into this, just haven't been able to come back to it due to an interview and hw. I'll work on it more this next week.

[davidchambers]

Any luck so far, @CBXZero?

[CBXZero]

Taking a look at it today, was swamped with schoolwork for a while. I should have something by the end of this weekend.

[CBXZero]

Spent some time on this during the weekend, the results that I get from curling the URL and the body that I console log when testing are different and that's currently what I'm hung up on.

[davidchambers]

If you provide a reproducible example of the difference in behaviour between curl and node I will help you to investigate. :)

[CBXZero]

I console.logged the body on the line right before the callback around line 38, pasted it into a test editor, and compared it to the fixture generated and they were radically different. I think there is some recursive call that is throwing me off.

EDIT: I can send you the exact outputs tomorrow afternoon if you want.

[CBXZero]

Here are the bodies that get logged in order of when I console.log them (this is from a single tutor.card() call searching via name, in this case "test" was used):

https://gist.github.com/CBXZero/b9136fd563b7028e2165

https://gist.github.com/CBXZero/b3322224639489d4ce0b

https://gist.github.com/CBXZero/2e8ec5a0e98e844ac4ec

[davidchambers]

Superseded by #97