Re: gitignore: got + cvs coexistence
On Fri, Jul 07, 2023 at 02:02:49PM +0200, Alexander Hall wrote: > On July 7, 2023 12:50:55 PM GMT+02:00, Stefan Sperling wrote: > >On Fri, Jul 07, 2023 at 12:26:16PM +0200, Tobias Heider wrote: > >> For bigger changesets I have started experimenting with using got. > >> I don't like to have the whole tree on disk twice so I keep my got and CVS > >> checkouts in the same directory. > > > >Curious. I am not sure how well that will work in practice but > >if it works for you then why not. I usually keep them separate, > >using temporary CVS checkouts for commits to CVS. > > > >> A downside of this approach is of course that got always lists all the > >> unknown > >> CVS dirs in got status. Does anything speak against ignoring them via > >> gitignore? > > > >In any case, I doubt anyone would ever want to check their CVS directories > >into Git. So ok by me. > > Are there cases when CVS is not a proper directory, as in "**/CVS/" (or just > "CVS/" which would work the same AFAIK)? CVS/ does not seem to work for me, but that could be a got limitation. > > /Alexander > > > > >> diff 7efd0ea99b79bbbda1539b1ae5c635ebfa688970 > >> 99115307e40554b41e9d62708e81599b0337da96 > >> commit - 7efd0ea99b79bbbda1539b1ae5c635ebfa688970 > >> commit + 99115307e40554b41e9d62708e81599b0337da96 > >> blob - 3fdff78bcab1fcf6493998cb99b64ff5a5da4f63 > >> blob + f07392d0d0015dac869ec1d55affd353aafba670 > >> --- .gitignore > >> +++ .gitignore > >> @@ -1,2 +1,3 @@ > >> **/obj > >> **/tags > >> +**/CVS > >> > >> > >
Re: gitignore: got + cvs coexistence
On July 7, 2023 12:50:55 PM GMT+02:00, Stefan Sperling wrote: >On Fri, Jul 07, 2023 at 12:26:16PM +0200, Tobias Heider wrote: >> For bigger changesets I have started experimenting with using got. >> I don't like to have the whole tree on disk twice so I keep my got and CVS >> checkouts in the same directory. > >Curious. I am not sure how well that will work in practice but >if it works for you then why not. I usually keep them separate, >using temporary CVS checkouts for commits to CVS. > >> A downside of this approach is of course that got always lists all the >> unknown >> CVS dirs in got status. Does anything speak against ignoring them via >> gitignore? > >In any case, I doubt anyone would ever want to check their CVS directories >into Git. So ok by me. Are there cases when CVS is not a proper directory, as in "**/CVS/" (or just "CVS/" which would work the same AFAIK)? /Alexander > >> diff 7efd0ea99b79bbbda1539b1ae5c635ebfa688970 >> 99115307e40554b41e9d62708e81599b0337da96 >> commit - 7efd0ea99b79bbbda1539b1ae5c635ebfa688970 >> commit + 99115307e40554b41e9d62708e81599b0337da96 >> blob - 3fdff78bcab1fcf6493998cb99b64ff5a5da4f63 >> blob + f07392d0d0015dac869ec1d55affd353aafba670 >> --- .gitignore >> +++ .gitignore >> @@ -1,2 +1,3 @@ >> **/obj >> **/tags >> +**/CVS >> >> >
Re: gitignore: got + cvs coexistence
On Fri, Jul 07, 2023 at 12:26:16PM +0200, Tobias Heider wrote: > For bigger changesets I have started experimenting with using got. > I don't like to have the whole tree on disk twice so I keep my got and CVS > checkouts in the same directory. Curious. I am not sure how well that will work in practice but if it works for you then why not. I usually keep them separate, using temporary CVS checkouts for commits to CVS. > A downside of this approach is of course that got always lists all the unknown > CVS dirs in got status. Does anything speak against ignoring them via > gitignore? In any case, I doubt anyone would ever want to check their CVS directories into Git. So ok by me. > diff 7efd0ea99b79bbbda1539b1ae5c635ebfa688970 > 99115307e40554b41e9d62708e81599b0337da96 > commit - 7efd0ea99b79bbbda1539b1ae5c635ebfa688970 > commit + 99115307e40554b41e9d62708e81599b0337da96 > blob - 3fdff78bcab1fcf6493998cb99b64ff5a5da4f63 > blob + f07392d0d0015dac869ec1d55affd353aafba670 > --- .gitignore > +++ .gitignore > @@ -1,2 +1,3 @@ > **/obj > **/tags > +**/CVS > >
gitignore: got + cvs coexistence
For bigger changesets I have started experimenting with using got. I don't like to have the whole tree on disk twice so I keep my got and CVS checkouts in the same directory. A downside of this approach is of course that got always lists all the unknown CVS dirs in got status. Does anything speak against ignoring them via gitignore? diff 7efd0ea99b79bbbda1539b1ae5c635ebfa688970 99115307e40554b41e9d62708e81599b0337da96 commit - 7efd0ea99b79bbbda1539b1ae5c635ebfa688970 commit + 99115307e40554b41e9d62708e81599b0337da96 blob - 3fdff78bcab1fcf6493998cb99b64ff5a5da4f63 blob + f07392d0d0015dac869ec1d55affd353aafba670 --- .gitignore +++ .gitignore @@ -1,2 +1,3 @@ **/obj **/tags +**/CVS
