Simple conflict situation during git merge. E.g.:
index e910fdc,492c972..0000000 @@@ -1,10 -1,5 +1,18 @@@ ++<<<<<<< HEAD +1 +2 +a +4 +5 +6 +b +8 +9 - newnewnew ++newnewnew ++======= + aa + aa + cc + aa -aa ++aa ++>>>>>>> br2 This is diff for not staged both modified file. I'm interested why at the end of each branch block we have lines like:
+8 +9 - newnewnew ++newnewnew but not just
+8 +9 +newnewnew Thanks.
TL;DR
a merge conflict needs to ask the user to judge which side should do what.
It seems the double plus sign line signifies the end of a problematic section - since it is a line BOTH sides of a merge conflict agree should be added into the final merge product.
Long Version
Amazing, I couldn't find any valuable documentation either, but it seems like this is a syntax meant for merge-conflict diff display - at least this is the only location where i've seen this happen.
I think it is based upon git diff syntax, and i will explain: when git performs a merge, it attempts to merge file A into file B. if there is a conflict, since git wants the user to do minimal work, it creates a partial-merge (work-in-progress) file C, and then asks the user to finish the merge by putting A and B on both sides, and letting the user decide how C should look in the end.
in this case, it means there are 3 files being diffed, and a syntax is required per line to show the relationship between all 3. so no we have more than 2 states for a line, depending on whether the merge decided to enter it into the merge (C) file.
a double plus just means that the line exists in both A and B, but is not in C currently, this is a good hint that it SHOULD be in C, but the merger for some reason couldn't make a choice to add it.
so the minus and plus on both 2 first columns of the diff just means whether the line is missing, existant, or added in the right or left files A and B.
here is a basic example:
A:
my original line 1. my original line 2. a unique A line. a new code line B:
my original line 1. my original line 2. a unique B line. a new code line C (the partial-diff):
my original line 1. my original line 2. a new code line lets say A is merged into B to create the partial diff C.
The unified diff window will show something like this:
(replace 0 by a regular space: when its neither minus nor plus)
00my original line 1. 00my original line 2. +0a unique A line. 0+a unique B line. ++a new code line As you can see, the line that is in both files on both sides of the merge is marked as double plus - to say "both sides agree this should be added" while the other ones say "each wants a different part to be added inside" - and this is the hint to the section on which the merge actually failed.
Hope i helped a bit, sorry for the messy look, im not good at using the syntax yet..
- character in column N means..."
- line, ++lineis the same as just simple ` +line`