Bug 226445 - Merge Review: symlinks
Summary: Merge Review: symlinks
Alias: None
Product: Fedora
Classification: Fedora
Component: Package Review   
(Show other bugs)
Version: rawhide
Hardware: All
OS: Linux
Target Milestone: ---
Assignee: manuel wolfshant
QA Contact: Fedora Package Reviews List
Depends On:
TreeView+ depends on / blocked
Reported: 2007-01-31 21:03 UTC by Nobody's working on this, feel free to take it
Modified: 2007-11-30 22:11 UTC (History)
1 user (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2007-03-06 11:28:38 UTC
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---
wolfy: fedora-review+

Attachments (Terms of Use)

Description Nobody's working on this, feel free to take it 2007-01-31 21:03:25 UTC
Fedora Merge Review: symlinks

Initial Owner: twaugh@redhat.com

Comment 1 Ralf Corsepius 2007-02-06 16:57:40 UTC
A very interesting case ;)

1. rpmlint:
W: symlinks summary-ended-with-dot A utility which maintains a system's symbolic
Stylishness - Should be fixed.

W: symlinks invalid-license distributable
More on this below.

W: symlinks no-url-tag
Doesn't make much sense, but to satisfy the burecrats, I'd propose to use
URL: ftp://metalab.unc.edu/pub/Linux/utils/file/

2. License
Could not find an explict license, but a terse "freely distributably" inside of
the *.lsm. tsx-11 origin => Very old, widely used and known to be distributable
package - IMO "distributable" is the correct term for this.

3. *.spec:
Would you explain the getconf-call in:

getconf causes the package to use the flags it receives from system the package
is being built on
- Potential (I am inclined to think almost zero) risk of non-deterministic build
results when users rebuild the package
- Not much of an issue when building the package inside of a build system as
part of a distro, except that it might tie this package to the specific
environment it is being built on - I am not sure what to do about it. Hardcoding?

Comment 2 Tim Waugh 2007-02-06 17:20:23 UTC

The purpose of the getconf call is to build with large file support.  The exact
flags to do this vary from platform to platform.  It basically comes down to
'-D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64' for 32-bit hosts and nothing at all
for 64-bit hosts as far as I am aware.

Tagged and built 1.2-27.fc7.

Comment 3 Ruben Kerkhof 2007-02-06 19:44:06 UTC
Hi Tim,

I've assigned the ticket back to the original reviewer, that way he can see your comments (he is not on the 
CC list).

Yes, I know, the process is clear as mudd at the moment, but have a look at http://

Ralf, back to you :-)

Comment 4 Ralf Corsepius 2007-02-07 02:11:07 UTC
(In reply to comment #3)
> Ralf, back to you :-)
I never reviewed this package, I just commented.

- BuildRoot doesn't comply to Fedora standards

Comment 5 Tim Waugh 2007-02-07 10:05:04 UTC
Tagged and built as 1.2-28.fc7.

Comment 6 manuel wolfshant 2007-02-23 11:20:38 UTC
 - Package meets naming and packaging guidelines
 - Spec file matches base package name.
 - Spec has consistant macro usage.
 - Meets Packaging Guidelines.
 - License is "distributable", already discussed above
 - License field in spec matches
 - Spec is legible, in American English
 - Sources match upstream, sha1sum:
a3dafe4b55206dcf19a8b4c67252628c2ad3fab4 symlinks-1.2.tar.gz
 - No BuildRequires
 - No locales/find_lang
 - Package is not relocatable
 - Permissions are sane [*]
 - Package has a correct %clean section.
 - Package has correct buildroot
      %{_tmppath}/%{name}-%{version}-%{release}-root-%(%{__id_u} -n)
 - Package is code or permissible content.
 - Doc subpackage not needed/used (no %doc files at all)
 - no headers/static/.pc/.la libs
 - no need for ldconfig or scriptlets
 - not a GUI
 - Package builds fine in mock/devel/x86_64
 - Package has no duplicate files in %files.
 - Package doesn't own any directories other packages own.
 - Package owns all files it creates; it does not create any directories
 - rpmlint output:
Source RPM:
W: symlinks invalid-license distributable
- discussed above
W: symlinks setup-not-quiet
- please consider using setup -q
Binary RPM:
E: symlinks no-cleaning-of-buildroot %install
that;s a MUSTFIX: Package lacks cleaning the buildroot in the %install section
rpmlint of symlinks:
W: symlinks invalid-license distributable
- see above


 - Should build in mock. - OK for devel/x86_64 and i386
 - Should build on all supported archs - tested on x86_64 and i386, OK
 - Should function as described - OK
 - Should have sane scriptlets - OK (no scriptlets)
 - Should have subpackages require base package with fully versioned depend. -
not needed
 - Should have dist tag OK
 - Should package latest version OK
 - check for outstanding bugs on package. (For core merge reviews) - OK (none)

mostly OK, with one MUSTFIX and a couple of cosmetic fixes:
- please consider using the newer preferred value in %files, (-,root,root,-)
- please add -q to setup in order to silence it
- it would be nice to add usage of smp_flags to make (not that it matters for a
5K source, but the rules are the rules)

- %install should contain rm -fR $RPM_BUILD_ROOT

Please fix the above and the package is APPROVED

Comment 7 Tim Waugh 2007-02-23 12:58:42 UTC
Tagged and built as 1.2-29.fc7.

Comment 8 manuel wolfshant 2007-02-23 14:12:36 UTC
All problems fixed. Packaged is APPROVED

Comment 9 manuel wolfshant 2007-03-06 09:25:59 UTC

How about importing this one to CVS and closing the ticket ? :)

Note You need to log in before you can comment on or make changes to this bug.