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

Re: Bug#18822: cvs-buildpackage: can't pass -I ! to cvs import

>>"Roderick" == Roderick Schertler <roderick@argon.org> writes:

Roderick> [I asked why not make -I ! the default in cvs-upgrade/cvs-inject?]
>>  Because there are upstream sources with CVS directories in them
>> that shall totally destroy the import. (Yes, such upstream sources
>> exist).
>> And ignoring whole directories seems to be ignored while generating
>> the diffs.

Roderick> I'm not sure what you mean by this second paragraph.  If the
Roderick> source is imported with -I ! -I CVS the upstream CVS
Roderick> directory won't be there when the diff is generated, so I
Roderick> can't see how it can affect that.  Would you expand on the
Roderick> problem?

	This is correct. I was saying that -I ! by it self is not such
 a good idea.

	The question I have is: why would one want files normally
 ignored by CVS in the repository? The file patterns are chosen to
 ignore object files, backup files, binaries, libraries, etc, of
 various sorts.

	Currently, the default list of ignored file name patterns is:
              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 think I never want this set of files in my repository (gack!
 think about a core file in there!)

	If I did want to do this, I would set up a ~/.cvsignore file
 with a ! followed by whatever I did want ignored.  In fact, there are
 thre places where I can still set it up:
   * The per-repository list in `$CVSROOT/CVSROOT/cvsignore' 
   * The per-user list in `.cvsignore' in my home directory 
   * Any entries in the environment variable `$CVSIGNORE' 

	Since it is already configurable on a site, user, and instance
 basis, I think it is correct to let the default be the (IMHO)
 reasonable set of ignored files that CVS has buit in.

	The other option is to parse all these locations, and descend
 the upstream source tree, look for matches in each directory,
 and offer to stop if any ignored files are found. I think that is a
 lot of work; and I do not really see the benefits (I mean, we can
 already tailor the behaviour of the programs as we wish!)

	I also think that is the upstream author has provided us with
 a .cvsignore file in a directory, we should go with their wishes.

 Seen on the wall in a New York subway station: "There are no integers
 n > 2 and x, y, z > 0, such that x^n + y^n = z^n I have found a truly
 wonderful proof of this. Unfortunately, my train is coming.
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 FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-devel-request@lists.debian.org . 
Trouble?  e-mail to templin@bucknell.edu .

Reply to: