On 2019-04-18 18:29, Andrew Cagney wrote:
On Thu, 18 Apr 2019 at 11:26, Martin Husemann <martin%duskware.de@localhost> wrote:On Thu, Apr 18, 2019 at 11:18:06AM -0400, Andrew Cagney wrote:So again, which commit broke the branch? With subversion, I can't answer that question.I am not sure I understand. svn log on the branch certainly shows all change sets, and all files touched. The point you are trying to make is that (if not put in the commit log) you can't identify the original trunk commit that got pulled up for each change on the branch?That just tells me the history as recorded by subversion. When two fully tested commits hit the repo at the same time, and the result is broken, who do I blame? Subversion? We can hardly wave a finger at the developer who had the simple misfortune of being second with their push?
Why do you insist on the "at the same time"? It can be at any time, and the problem is the same.
With ACID, since the second developer's change gets rejected, they can be rest assured that things didn't break.
git instead goes even more crazy. You have your local copy, which might be totally out of sync with the master, and you might be doing all kind of changes, and then when you sync up with the master, you have all kind of conflicts on files that have been changed by both, but files that have only been changed by one side will happily pass through, and you actually have the exact same situation you are complaining about. So git does absolutely nothing to solve this.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt%softjar.se@localhost || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol