Re: Bad Git Commit / Advice?
I'm going to leave master as-is and merge in the PRs into develop that I merged into master. It's unfortunate that master will have a snapshot POM on it, but at least those two branches will be consistent. I won't rush to cut a release now, but I'll read up on the ASF incubator release process to get an idea about the effort involved (and hopefully a good idea about planning and deciding when to cut a release). Thanks All! --Aaron On Tue, Dec 22, 2015 at 10:33 PM Josh Elserwrote: > Haha, that's a good representation of how difficult the first one will be > :) > > Releasing will get that easy -- the first few will not be. > > I used to try to keep master history "nice", but I've long since given > up. I'm convinced that there's no real value in trying to this. You just > end up ignoring useful tasks your tools provide. > > It's up to you, but accidentally committing code to a branch shouldn't > be cause to make a release. You can start thinking about a 3.2.10 > release, but these are unrelated ideas. :) > > Puja Valiyil wrote: > > Yes-- though I'm not sure how involved that is. > > > > Sent from my iPhone > > > >> On Dec 22, 2015, at 7:19 PM, Sean Busbey wrote: > >> > >> When y'all say "cutting a 3.2.10 release" do you mean going through the > ASF > >> incubator process for having a release? > >>> On Dec 22, 2015 16:45, "Aaron D. Mihalik" > wrote: > >>> > >>> yep, master didn't like force-push (results below). > >>> > >>> You don't like the idea of cutting a 3.2.10 release? > >>> > >>> I'm hesitating to do a revert because > >>> > >>> (1) my push onto the master involved 13 commits and it seems like that > >>> would require 13 reverts and > >>> > >>> (2) that leaves the master history really ugly... > >>> > >>> --Aaron > >>> > >>> > >>> === > >>> > >>> remote: Rewinding refs/heads/master is forbidden. > >>> > >>> remote: > >>> > >>> To https://git-wip-us.apache.org/repos/asf/incubator-rya.git > >>> > >>> ! [remote rejected] master -> master (pre-receive hook declined) > >>> > >>> error: failed to push some refs to ' > >>> https://git-wip-us.apache.org/repos/asf/incubator-rya.git' > >>> > On Tue, Dec 22, 2015 at 12:21 PM Josh Elser > wrote: > > Since there's still no force-push on master (AFAIK), just use `git > revert` on master. > > Aaron D. Mihalik wrote: > > I've made the first big mistake on the git repo for Rya. > > > > Basically, I pulled in three of the latest commits onto master > instead > >>> of > > develop. Now Master is ahead of develop (and the POM on master > > 3.2.10-SNAPSHOT instead of 3.2.9). > > > > What's the best way to remedy this? > > > > I'm tempted to cut a 3.2.10 release and open 3.2.11-SNAPSHOT on > >>> Develop. > > Or perhaps a git-pro can undo my push onto master the apache repo. > > > > Any advice? > > > > --Aaron >
Re: Bad Git Commit / Advice?
yep, master didn't like force-push (results below). You don't like the idea of cutting a 3.2.10 release? I'm hesitating to do a revert because (1) my push onto the master involved 13 commits and it seems like that would require 13 reverts and (2) that leaves the master history really ugly... --Aaron === remote: Rewinding refs/heads/master is forbidden. remote: To https://git-wip-us.apache.org/repos/asf/incubator-rya.git ! [remote rejected] master -> master (pre-receive hook declined) error: failed to push some refs to ' https://git-wip-us.apache.org/repos/asf/incubator-rya.git' On Tue, Dec 22, 2015 at 12:21 PM Josh Elserwrote: > Since there's still no force-push on master (AFAIK), just use `git > revert` on master. > > Aaron D. Mihalik wrote: > > I've made the first big mistake on the git repo for Rya. > > > > Basically, I pulled in three of the latest commits onto master instead of > > develop. Now Master is ahead of develop (and the POM on master > > 3.2.10-SNAPSHOT instead of 3.2.9). > > > > What's the best way to remedy this? > > > > I'm tempted to cut a 3.2.10 release and open 3.2.11-SNAPSHOT on Develop. > > > > Or perhaps a git-pro can undo my push onto master the apache repo. > > > > Any advice? > > > > --Aaron > > >
Re: Bad Git Commit / Advice?
When y'all say "cutting a 3.2.10 release" do you mean going through the ASF incubator process for having a release? On Dec 22, 2015 16:45, "Aaron D. Mihalik"wrote: > yep, master didn't like force-push (results below). > > You don't like the idea of cutting a 3.2.10 release? > > I'm hesitating to do a revert because > > (1) my push onto the master involved 13 commits and it seems like that > would require 13 reverts and > > (2) that leaves the master history really ugly... > > --Aaron > > > === > > remote: Rewinding refs/heads/master is forbidden. > > remote: > > To https://git-wip-us.apache.org/repos/asf/incubator-rya.git > > ! [remote rejected] master -> master (pre-receive hook declined) > > error: failed to push some refs to ' > https://git-wip-us.apache.org/repos/asf/incubator-rya.git' > > On Tue, Dec 22, 2015 at 12:21 PM Josh Elser wrote: > > > Since there's still no force-push on master (AFAIK), just use `git > > revert` on master. > > > > Aaron D. Mihalik wrote: > > > I've made the first big mistake on the git repo for Rya. > > > > > > Basically, I pulled in three of the latest commits onto master instead > of > > > develop. Now Master is ahead of develop (and the POM on master > > > 3.2.10-SNAPSHOT instead of 3.2.9). > > > > > > What's the best way to remedy this? > > > > > > I'm tempted to cut a 3.2.10 release and open 3.2.11-SNAPSHOT on > Develop. > > > > > > Or perhaps a git-pro can undo my push onto master the apache repo. > > > > > > Any advice? > > > > > > --Aaron > > > > > >
Re: Bad Git Commit / Advice?
Yes-- though I'm not sure how involved that is. Sent from my iPhone > On Dec 22, 2015, at 7:19 PM, Sean Busbeywrote: > > When y'all say "cutting a 3.2.10 release" do you mean going through the ASF > incubator process for having a release? >> On Dec 22, 2015 16:45, "Aaron D. Mihalik" wrote: >> >> yep, master didn't like force-push (results below). >> >> You don't like the idea of cutting a 3.2.10 release? >> >> I'm hesitating to do a revert because >> >> (1) my push onto the master involved 13 commits and it seems like that >> would require 13 reverts and >> >> (2) that leaves the master history really ugly... >> >> --Aaron >> >> >> === >> >> remote: Rewinding refs/heads/master is forbidden. >> >> remote: >> >> To https://git-wip-us.apache.org/repos/asf/incubator-rya.git >> >> ! [remote rejected] master -> master (pre-receive hook declined) >> >> error: failed to push some refs to ' >> https://git-wip-us.apache.org/repos/asf/incubator-rya.git' >> >>> On Tue, Dec 22, 2015 at 12:21 PM Josh Elser wrote: >>> >>> Since there's still no force-push on master (AFAIK), just use `git >>> revert` on master. >>> >>> Aaron D. Mihalik wrote: I've made the first big mistake on the git repo for Rya. Basically, I pulled in three of the latest commits onto master instead >> of develop. Now Master is ahead of develop (and the POM on master 3.2.10-SNAPSHOT instead of 3.2.9). What's the best way to remedy this? I'm tempted to cut a 3.2.10 release and open 3.2.11-SNAPSHOT on >> Develop. Or perhaps a git-pro can undo my push onto master the apache repo. Any advice? --Aaron >>
Re: Bad Git Commit / Advice?
Haha, that's a good representation of how difficult the first one will be :) Releasing will get that easy -- the first few will not be. I used to try to keep master history "nice", but I've long since given up. I'm convinced that there's no real value in trying to this. You just end up ignoring useful tasks your tools provide. It's up to you, but accidentally committing code to a branch shouldn't be cause to make a release. You can start thinking about a 3.2.10 release, but these are unrelated ideas. :) Puja Valiyil wrote: Yes-- though I'm not sure how involved that is. Sent from my iPhone On Dec 22, 2015, at 7:19 PM, Sean Busbeywrote: When y'all say "cutting a 3.2.10 release" do you mean going through the ASF incubator process for having a release? On Dec 22, 2015 16:45, "Aaron D. Mihalik" wrote: yep, master didn't like force-push (results below). You don't like the idea of cutting a 3.2.10 release? I'm hesitating to do a revert because (1) my push onto the master involved 13 commits and it seems like that would require 13 reverts and (2) that leaves the master history really ugly... --Aaron === remote: Rewinding refs/heads/master is forbidden. remote: To https://git-wip-us.apache.org/repos/asf/incubator-rya.git ! [remote rejected] master -> master (pre-receive hook declined) error: failed to push some refs to ' https://git-wip-us.apache.org/repos/asf/incubator-rya.git' On Tue, Dec 22, 2015 at 12:21 PM Josh Elser wrote: Since there's still no force-push on master (AFAIK), just use `git revert` on master. Aaron D. Mihalik wrote: I've made the first big mistake on the git repo for Rya. Basically, I pulled in three of the latest commits onto master instead of develop. Now Master is ahead of develop (and the POM on master 3.2.10-SNAPSHOT instead of 3.2.9). What's the best way to remedy this? I'm tempted to cut a 3.2.10 release and open 3.2.11-SNAPSHOT on Develop. Or perhaps a git-pro can undo my push onto master the apache repo. Any advice? --Aaron