Michael Sims tried to tell us something, and all I got was: > At 06:33 PM 11/22/2001 +0000, Michael L. Hostbaek wrote: > > ./htdocs/foo > ./htdocs/bar > > Have a "foo" and a "bar" module in CVS, not just "htdocs". Then after you > have imported the directories, delete the ones out of htdocs, then do a > "cvs co foo" and "cvs co bar" directly from ./htdocs, and this will put > these CVS working directories directly into your production > environment. Then when you need to update them just do a "cvs update" from > within ./htdocs/foo and ./htdocs/bar. There shouldn't be any need to do > any manual copying. The whole point of CVS is avoid stuff like that, at > least that is my understanding. >
Well, let us imagine my docroot looks something like this: members/ usage/ mrtg/ phpMyAdmin/ images/ @mp3/ @download/ stuff/ index.html hello.html foo.txt Then, let us say that I have the following modules: main_www (that is the docroot /) members_www (members dir) admin_www (phpMyAdmin dir) images_www (images dir) usage, mrtg are not in the cvsrep, cause they are dynamically generated. mp3 and download are not in the cvsrep, cause they are symlinks. (and symlinks are not imported into cvsrep, and I do not want 120 MB of binary files in my rep.) Let's just say, that someone commits a usage dir. Then the next time, I am going to update my docroot working dir, I am going to have a nasty conflict, right ? Also, I was under the impression that people are always exporting their projects. Cause they do not want the CVS dir hanging around in docroot. And furthermore, I would basically give everyone with commit access to CVS, write access to my docroot production environment. (I'd like to be able verify that process) Again thanks for your time and efforts... I hope you see my dilemma.. ? -- Regards, Michael L. Hostbaek -= Thanks for all the fish.. =- _______________________________________________ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
