Tom Udale wrote: > > We have just installed cvs and are finally starting to get a grip on how > to use it. One conceptual problem so far update vs checkout. What is > the real difference between the two?
You can't update a sandbox until you've checked it out. If you run 'cvs checkout' on an already checked-out sandbox, it will revert and acutally do an 'update'. Update re-syncs your local sandbox with the repository, and checkout is used to get a new sandbox. In a typical 'cvs development cycle', you would do something like: 1. cvs checkout my_module -- get a new sandbox 2. [make some changes] 3. cvs update -- get any new changes from the repository, merging with local changes if necessary 4. cvs commit -- save local changes to the repository 5. goto 2 You only need to 'checkout' if you want to start over fresh. > I orginally thought that checkout > was the only way to get newly added directories (and create the local > CVS subdirectories). Now, from this newsgroup, I see that there is > update -d which seems very much like checkout. 'cvs update -d' behaves *like* checkout, in that it creates missing directories in your sandbox, but that's where the conceptual similarities end. > If you always wanted to > update -d, is there any reason not to use checkout? Is checkout simply > a 'convienient' way of saying update -d (or perhaps with some other > switches) or is there a more fundemental difference? > > Thanks, > > Tom Udale If you're still confused, take a look at Karl Fogel's book. It is an *excellent* intro to CVS. -Matt _______________________________________________ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs