Don't disregard trailing newlines in Differ input. #70

[phiggins]

A fix for #70.

String#split takes an optional second argument. From the docs:

If the limit parameter is omitted, trailing null fields are suppressed. If limit is a positive number, at most that number of fields will be returned (if limit is 1, the entire string is returned as the only entry in an array). If negative, there is no limit to the number of fields returned, and trailing null fields are not suppressed.

The difference being output like this:

 >   "foo\nbar\n".split("\n")
 => ["foo", "bar"] 
 > "foo\nbar\n".split("\n", -1)
 => ["foo", "bar", ""] 
 > "foo\nbar\n\n\n".split("\n", -1)
 => ["foo", "bar", "", "", ""]

I modified the methods that preprocessed the input to Differ to keep these empty strings for any trailing newlines. This worked to fix the issue in #70, but I had to modify lots of Differ's test cases since they used a trailing newline to trigger their diffability.

[phiggins]

💭 This wasn't really related to this issue, but it was a missing test case so I threw it in just for completeness's sake.

[JonRowe]

Currently the build is failing with incorrect or unexpected diffs...

[phiggins]

@JonRowe Thanks for the heads up, I opened a PR against rspec-expectations to fix the specs that depended on Differ.

[phiggins]

The branch I opened on rspec-expectations, rspec/rspec-expectations#556, is passing, but this build is still failing. Is there some way I can make this use my branch of rspec-expectations on CI?

[JonRowe]

Generally our procedure is to add a temporary commit linking the Gemfile to the branch, and then when passing, delete the commit and merge things simultaneously.

[phiggins]

Generally our procedure is to add a temporary commit linking the Gemfile to the branch, and then when passing, delete the commit and merge things simultaneously.

I did that, unless I'm misunderstanding what you mean.

[JonRowe]

You're not, for whatever reason your build is still failing against your branch

[myronmarston]

Is this not an issue anymore, @phiggins? Sorry I never got around to reviewing it...

[phiggins]

It seemed like I was having to make pretty invasive changes to fix a minor edge case bug. The issues are still open so if there's interest these can be reopened.