So the file history might look like this:
\nabc.properties\n+- 1.1\n+- 1.2\n+- 1.3\n| +- 1.3.2.1 ("Branch_1")\n| +- 1.3.2.2\n| +- 1.3.2.3 (not yet merged into the head)\n+- 1.4\n| +- 1.4.2.1 ("Branch_2")\n| +- 1.4.2.2 (merged into trunk as 1.5)\n+- 1.5\n+- 1.6\nSo now let's say that I want to merge Revision 1.3.2.3 (the most recent version on "Branch_1") into the head.
The commands I have been using to do this are:
cvs -q up -A (to get me onto the head revision)
cvs -q up -dP -j Branch_1 (to merge the latest changes from "Branch_1" into the head)
Let's say, for the sake of argument, that there are luckily no conflicts resulting from this merge. All changes have been in disparate parts of the file, and CVS has been able to resolve changes without human intervention.
In this case, what will happen to changes that have been made on "Branch_2" (1.4.0.2 branch)? Will they be completely lost? If so, how do I stop that from happening?
I'm asking this question because I'm in charge of merges at work, and going through CVS history on a file appears to show that changes in a case similar to this are indeed getting lost. I'm at a loss to explain how this can happen.
Some web pages that I've been using as CVS help guides:
[link|http://computing.ee.ethz.ch/sepp/cvs-1.10-to/cvsbook/main_35.html|http://computing.ee....book/main_35.html]
[link|http://www.psc.edu/~semke/cvs_branches.html|http://www.psc.edu/~...cvs_branches.html]
[link|http://wiki.x.org/wiki/CvsBranchnames|http://wiki.x.org/wiki/CvsBranchnames]
TIA,