I've taken this one step futher. Rather than having the user add the Change #, the verify message script queries the database for the next change number, and appends it to the top of the comments entered by the user. To do this a patch to cvs is required to allow the comments to be updated by the verify message script.
Allen Jensen wrote: > We do it outside of CVS by having the users always include a Change or Defect > number assigned out of out change/defect tracking database. You can use the > RCS templates and a script to check the log message to verify a valid ID was > entered (if you have such a database - or you could just use a simple file with >valid ID's....) > > You can then, by using various UNIX tools, get a list of all of the files modified > relative to a particular change ID. Take a look at the automatic changelog > generation scripts for CVS.... (http://cvsbook.red-bean.com/ is a great reference) > > Allen Jensen > ----- Original Message ----- > From: "Gregoire Banderet" <[EMAIL PROTECTED]> > Newsgroups: gnu.cvs.help > To: <[EMAIL PROTECTED]> > Sent: Wednesday, January 16, 2002 6:58 AM > Subject: commit IDs ? > > > How could I know which files were modified together (same commit). > > I mean, I want to know which files (if any) were commited in the same time as file >f version v. > > > > Is there a command for that ? > > Looking at the CVSROOT/history file I can imagine that it's possible to do. > > > > Thanks, > > Greg. > > > > _______________________________________________ > > Info-cvs mailing list > > [EMAIL PROTECTED] > > http://mail.gnu.org/mailman/listinfo/info-cvs > > _______________________________________________ > Info-cvs mailing list > [EMAIL PROTECTED] > http://mail.gnu.org/mailman/listinfo/info-cvs _______________________________________________ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
