Community technical support mailing list was retired 2010 and replaced with a professional technical support team. For assistance please contact: Pre-sales Technical support via email to firstname.lastname@example.org.
Hello Tony, thanks for your comment. Tony Hoyle <tony.hoyle at march-hare.com> writes: > Oliver Koltermann wrote: > > Yes, Flávio, you are right. The :1.1 is the thing I've never seen > > before. Your guess sounds somewhat reasonable, but I suspect a lot of > > dependencies to other tools and workflows if the CVS/Tag file is > > (mis-?)used that way. > > In practice nobody ever interprets that file, just sends it right back > to the server - which is exactly what we want. You can't use any > other file because of the way frontends sometimes use them. So I guess Flávio was right and what I'm seeing is some internal directory information. To be clear on that: I have *no* problem with that, now I know the background. But I am still not sure about possible side effects. If I want to "cvs add" some files they get added on the (branch-)tag the directory is on. And where do I find this information? AFAIK in the CVS/Tag file. When working on the command line I sometimes use "cat CVS/Tag" to get the directorys Tag. I know by now how to handle the :1.x part, but e.g. WinCVS does the same - without this special knowledge. So my point is that the content of CVS/Tag is not available over a specified interface (I know of) or call of cvsnt, but contains necessary information (the directorys tag). There may be more tools or scripts using this information... I don't know. That's why I don't think that "nobody ever interprets that file". But maybe there is another way I don't know of, please let me know. Best regards, O. Koltermann.