On Apr 7, 2005 1:33 PM, Frank W. Zammetti <[EMAIL PROTECTED]> wrote: > On Thu, April 7, 2005 1:15 pm, Fogleson, Allen said: > > Huh? > > > > SVN is like CVS. You don't lock files when you check them out. I can't > > write to the svn repository but almost anyone can do an anonymous > > checkout, do updates (one way, to their copy of the repository). > > Otherwise it would be near impossible to write patches :) > > This is a lesson for me. I thought one of the primary purposes of any > source control system was to disallow concurrent modifications (unless > specifically allowed). The whole point of checking code out, as I > understand it (and use it on a daily basis) is to be sure I am the only > one modifying a particular source file at any given time. Sure, you can > always get the latest code from the repository and modify it to your > hearts' content, but can you apply your changes anonymously then? Perhaps > I have some stuff to learn here, but that is completely contrary to how we > use source control here.
You can create a patch that represents the modifications, and then a committer can apply the patch. Appying patches is generally an automatic process and only takes a few minutes. The change logs sent to the commits@ lists are patches. It's not unusual for a mature project to ask that all changes be submitted as patches first, even by the committers. The list can then review the changes before they are made. Around here, we are still operating in lazy consensus mode. We can commit a change, and if nobody squawks, it stick. But, if a PMC member does veto a product change, we have to do undo it. -Ted. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
