[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#20499: cvs-buildpackage: doesn't handle symlinks in debian only packages

>>"Fumitoshi" == Fumitoshi UKAI <ukai@debian.or.jp> writes:

Fumitoshi> I think cvs-inject should import with option "-I !" to
Fumitoshi> avoid ignoring any files at all (even ignored by default).

	I disagree. I would be very annoyed if core files and binaries
 and .o files and stuff were imported into my repository, especially
 as a part of an upgrade operation.

	The current default list of ignored file name patterns for CVS
 is the following; I find these to be a fine set of files to ignore as
 a default. This is a very reasonable set.

              RCS     SCCS    CVS     CVS.adm
              RCSLOG  cvslog.*
              tags    TAGS
              .make.state     .nse_depinfo
              *~      #*      .#*     ,*      _$*     *$
              *.old   *.bak   *.BAK   *.orig  *.rej   .del-*
              *.a     *.olb   *.o     *.obj   *.so    *.exe
              *.Z     *.elc   *.ln

	I would not like files matching these rules to be in my
 repository. (CVS is not designed to be for binary data, really). 

	Of course, this may not be true in all cases, but for the
 majority of cases this holds. I agree, though, that the user should
 have a means of putting in their own list of ignored file name

        I have reflected upon this for a while. There are three ways
 for the user to set their own set of ignore patterns: a per
 repository method, a per user method, and a means of doing this on a
 per import basis.
   * The per-repository list in `$CVSROOT/CVSROOT/cvsignore' 
   * The per-user list in `.cvsignore' in your home directory 
   * Any entries in the environment variable `$CVSIGNORE' 

        I see no reasonable way of having the scripts do anything that
 is worth the time required to spend on this, since parsing the
 locations to determine the current set of patterns, and deciding which
 file names match the patterns (since these are shell glob patterns),
 is a lot of work, and there already are ways for the user to modify
 the documented behaviour. So at the moment, I have no plans of
 passing -I patterns to the CVS commands automatically.

        The .cvsignore files, if present in the upstream source,
 should, in my opinion, be respected.

 "To err is human, to compute divine. Trust your computer but not its
 programmer." Morris Kingston
Manoj Srivastava  <srivasta@acm.org> <http://www.datasync.com/%7Esrivasta/>
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E

To UNSUBSCRIBE, email to debian-devel-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Reply to: