glusterfs-hadoop
glusterfs-hadoop copied to clipboard
support for major version incrementation
Lets decide how to do major version upgrades,,,, can this be automated via commit message? Or should we even worry about it at all?
(just some brainstorming...) Right now
- we explicitly grep for "ci-skip in commit message".
- we always assume only the minor version is being incremented.
Lets make these key-value pairs sent into the bash script, example:
[ci-skip="false", version="auto"] translates to --ci-skip=false --version=auto, which results in autoincrement of minor version along with auto publishing of the new version, that way someone who wants to upgrade the major version can do
commit -m "Lets bump to 2.2 [version="2.2.0"]
and then the version, rather than auto incrmented via python script, is explicitly set in the pom.
note , this is VERY low priority... because major releases are rarely implemented, and we have an easy workaround : Just turn build server off when we do a major release bump,do a pull request for the version bump, comment out the auto incrementer, and then manually run the build.
Then turn the CI back on and re-enable the auto-incrementer .
AFTER FURTHER BRAINSTORMING WITH JOHN, IM NOW OPEN TO CLOSING THIS ISSUE ENTIRELY. :)