CVS is your friend

I've read through most of the "CVS directory structure" thread, and it feels like the "solution" here simply involves a bit of end-user education regarding cvs(1). It seems like the majority of the PEAR developers check out the entire 'pear/' hierarchy out of CVS. In fact, I could go as far as to assume that many of you are under the impression that this is the only way to get your files. It's not. If I just want to check out the source code for the LiveUser package, I can just do: $ cvs co pear/Perm_LiveUser This will produce a 'pear/Perm_LiveUser' directory. As acknowledged on this list, it's difficult to work in this directory structure because it does not mirror the layout of the installed package. Fortunately, cvs(1) allows us to specify an alternate directory name in which the checked-out files will be placed: $ cvs co -d LiveUser pear/Perm_LiveUser This will produce a 'LiveUser' directory which contains the contents of the 'pear/Perm_LiveUser' CVS module. Unfortunately, the '-d' option does handle subdirectories (e.g. '-d foo/bar' won't work), so you need to give it a little help if you need to nest your checkout: $ mkdir Category; cd Category $ cvs co -d Package pear/Category_Package The nice thing about this solution is that it only needs to be done once, when the files are initially checked-out from CVS. Subsequent operations are performed in the resulting directory just like they are today. I *think* this should make everyone happy. -- Jon Parise (jon@php.net) :: The PHP Project (http://www.php.net/)