Provides narrower regex for conflict marker detect

[allenhumphreys]

I recently ran into a situation when parsing the acknowledgements file within my Podfile. One of the pods (libidn v1.33) has the following in their license:

OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.

===============================================

Libidn COPYING -- Explanation of licensing conditions.
Copyright (C) 2002-2015 Simon Josefsson

The regex currently used in file_in_conflict is detecting this section separator as a merge conflict, thus Plist.read_from_path is refusing to open the file.

I'm proposing a new regex for matching the entire conflict marker sequence which is:

1. beginning of line, followed by exactly 7 left angle brackets
2. any characters
3. beginning of line followed by exactly 7 equality symbols
4. any characters
5. beginning of line followed by exactly 7 right angle brackets

A trivial script for testing the regex

#!/usr/bin/env ruby

def file_in_conflict?(contents)

    contents.match(/^<{7}(?!<)[\w\W]*^={7}(?!=)[\w\W]*^>{7}(?!>)/m)
end

conflict_example = "<<<<<<< HEAD
words and stuff go here
=======
other, different words go here
>>>>>>> 8ab358aeba3f
"

if file_in_conflict?(conflict_example)
    puts "This is in conflict" # Expect this
else
    puts "This is not in conflict"
end

non_conflict_example = "<<<<<<<<
words and stuff go here
========
other, different words go here
>>>>>>>>
"

if file_in_conflict?(non_conflict_example)
    puts "This is in conflict"
else
    puts "This is not in conflict" # Expect this
end

[segiddins]

Can you please break the regexp over multiple lines and add inline comments to it?
Looks 💯 with added test coverage

[dnkoutso]

nit, typo exactly

[allenhumphreys]

@segiddins I'm not sure why the CI is failing. Reading the output it seems to be a setup issue, but I'm new to ruby/gem/rake/allofit. Is it something I need to address?

[segiddins]

I don't know, @orta is the one who set up the danger integration, so I don't know why it's stopped working. we're tracking that over at #493

[dnkoutso]

@allenhumphreys can you try rebasing?

[allenhumphreys]

@dnkoutso Squashed and rebase on master. ✅

[dnkoutso]

@allenhumphreys as the barista says, need to add a changelog entry ;)

[segiddins]

👍 Thanks!

[dnkoutso]

Looks like rebase did it.

[allenhumphreys]

@segiddins @dnkoutso It's been a pleasure learning and participating, hope to contribute more in the future!

[allenhumphreys]

Apparently, in the changelog, the 2 trailing spaces at the end of the first line must mean something in markdown that I was unfamiliar with. So, that's my bad.

[segiddins]

It just forces GitHub to render the next line of markdown on a new line, no big deal, we can fix it up during release