Marco wrote, On 10/30/2008 02:16 AM:
> Hi all
> Can someone give me some advices how to do below ruquest?
>
> Two modules: A and B
> one file: f
> f exists in A and B.
>
> upgrade f, check in to A or B not (A and B).
> whenever check out A or B, the f is synced most current one.
chaitu wrote, On 10/30/2008 05:38 AM:
Hi Marco,
AFAIK, this can be done only by writing a wrapper check-in program/
script which is coded to commit to both branches (if you omit B, it
will detect and take care automatically.....you have to code it to
consider a list of modules you want to be in sync, like "A B C ...").
All your check-ins must be done through this script and not directly
cvs commit command.
Anyone else knows a better way?
- Chaitanya
The usual suggestion is to make a third module/directory C, which contains
files that have been in common between A and B, then in the
$CVSROOT/CVSROOT/modules file setup modules within modules, i.e., :
ProjA &A &C
ProjB &B &C
after that you have to update the makefiles in A & B (now ProjA & ProjB) to
either point to where the files are in the new sandboxes (VPATH could be
helpful here), or build C independently as a lib that A and B use.
--
Todd Denniston
Crane Division, Naval Surface Warfare Center (NSWC Crane)
Harnessing the Power of Technology for the Warfighter
