[Bug 195] Review request: perl-Crypt-IDEA - Perl interface to IDEA block cipher

Ralf Corsepius rc040203 at freenet.de
Sun Jan 18 16:50:07 CET 2009


Going off-bugzilla.

RPM Fusion Bugzilla wrote:
> http://bugzilla.rpmfusion.org/show_bug.cgi?id=195
> 
> 
> 
> 
> 
> --- Comment #10 from Michael Schwendt <mschwendt at gmail.com>  2009-01-18 12:49:21 ---
> The original rationale for recommending "install -p" and "cp -p" when
> installing files manually (inside the spec file e.g.) has been in preserving
> timestamps for _prebuilt_ files in tarballs. Such as various forms of
> documentation files.

> It is considered helpful by many package users, because they can judge about
> the age of documentation files simply by checking timestamps. Particularly
> helpful with but not limited to larger pdf/ps files and html trees. No need to
> revisit such files after package updates, if the documentation is still
> unchanged since 2001, for example, and other files are several months old, too.
> User would "cd /usr/share/doc/..." and quickly notice that only a README file
> has changed for this update.

Provided the fact many pdf/ps/man files are _generated_ (doxygen, 
texinfo, pod2man), this rationale is of very limited use, as well as a 
simple "INSTALL=install -p" would not help in many occasions.

> A few corner-cases have been found where install -p helped,
Correct, there have been _very few_ such cases.

Off-head I don't recall any ;)

> I think related to
> .rpmnew creation of config files just because the mtime changed (and not just
> the checksum).

.rpmnew's are being generated for %config files. Handling them correctly 
is rpm's job. And yes, IIRC, rpm once had been broken wrt. them.

> Some reviewers have expanded the recommendation to use "install -p ..." to also
> run "make install DESTDIR=$RPM_BUILD_ROOT INSTALL="install -p", which I think
> is somewhat over the top

Agreed, but I'd express it a bit harder: Enforcing such a rule 
demonstrates a reviewer's lack of technical skills.

He should at least examine whether a package recognises "INSTALL" (All 
automake based ones do, most others don't), rsp. whether preserving 
would make any sense in a particular package.

In most cases, it doesn't.

> even if covers a few more prebuilt files. Some
> tarballs mix cp/install and mkdir/install, so one would need to switch to "cp
> -p" for the full show.
additionally there are other means of generating files ...

E.g. perl modules typically copy around their sources (lib/blib) during 
builds, generate man-pages on the-ply (using pod2man) etc.

> So, conclusively: Historically it has only been pedantic eye-candy (albeit
> considered helpful by our users). If nowadays there is knowledge that it
> actually fixes anything else, please document that.
Actually, I would finally see any actual bug, this "install -p" fixes, 
these days.

In the past, have been some case, but AFAICT, all of them actually have 
been side-effects of bugs elsewhere (e.g. rpm).

Ralf




More information about the rpmfusion-developers mailing list