Re: adding on branch
"Paul Edwards" <[email protected]>
| Newsgroups | gmane.comp.version-control.cvs.bugs |
|---|---|
| Organization | BigPond Internet Services (http://www.bigpond.net.au) |
| Message-ID | <[email protected]> |
"Max Bowsher" <[email protected]> wrote in message news:[email protected]... > Paul Edwards wrote: > > "Max Bowsher" <[email protected]> wrote in message > > news:[email protected]... > >>> Ok, but an import is another way of adding a "head" version. > >>> So the import should also go through the same logic, moving a > >>> file out of the Attic and putting it into the normal spot, > >> > >> No, if you import on top of a dead head revision, you get a conflict, as > >> expected, and then you merge, and the commit after that commits and alive > >> revision, causing the RCS file to move out of the Attic. > > > > What I meant was, if you do an import, and no file currently > > exists, it will automatically add it, and a head will now exist, > > where none previously existed. > > > > So import is different from adding on a branch. > > > > Import is expected to create a head version. > > > > So if in the Attic there is an existing file, but it only has a branch > > (file was added on a branch), there is no explicitly-dead head, > > then that should also be reason to move out of the Attic. > > Um, no. There cannot not be a head. Therefore, when a file is added on a > branch, there *must* be a dead head added. Correspondingly, importing > creates a live head revision it creates a new RCS file - but if one already > exists, there is no reason for import to touch head, and no reason to affect > the Attic-status of a file. Logically, I don't think a user submitting an advanced version of a file, on his own branch, should interfere with the natural behaviour of an import, which would create a non-dead-head, not in the Attic. The whole point of a branch is for people to have their own playpen and not cause harm to the natural processing outside that branch. > > At work, our multiple projects are branched off various drops of > > various imports. The head is never used. So we never get files > > added on a branch moved out of the Attic unless I do it manually. > > And this is a problem because...? I have had other problems, can't remember exactly what, and I solved them by moving them out the Attic. I don't think a file should be in the Attic if it's counterparts (imported in the same way) are not. BFN. Paul.