Bug 206512 - sysutils/coreutils: Update to 8.25
Summary: sysutils/coreutils: Update to 8.25
Status: Closed FIXED
Alias: None
Product: Ports & Packages
Classification: Unclassified
Component: Individual Port(s) (show other bugs)
Version: Latest
Hardware: Any Any
: --- Affects Only Me
Assignee: Kurt Jaeger
URL:
Keywords: patch-ready
Depends on:
Blocks:
 
Reported: 2016-01-23 02:38 UTC by w.schwarzenfeld
Modified: 2016-02-26 20:50 UTC (History)
4 users (show)

See Also:
pi: maintainer-feedback+


Attachments
svn-diff-coreutils-8.25 (10.17 KB, patch)
2016-01-23 02:38 UTC, w.schwarzenfeld
koobs: maintainer-approval-
Details | Diff
poudriere-coreutils-8.25.log (143.64 KB, text/x-log)
2016-01-23 02:39 UTC, w.schwarzenfeld
no flags Details
portlint-coreutils-8.25 (576 bytes, text/plain)
2016-01-23 02:41 UTC, w.schwarzenfeld
no flags Details
patch to correct OP's breakage when NLS isn't selected (4.24 KB, patch)
2016-02-02 21:57 UTC, jharris
jharris: maintainer-approval+
Details | Diff
current-patch (6.24 KB, patch)
2016-02-05 21:54 UTC, Kurt Jaeger
no flags Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description w.schwarzenfeld 2016-01-23 02:38:34 UTC
Created attachment 165984 [details]
svn-diff-coreutils-8.25

Update coreutils to 8.25.

ChangeLog:

* Noteworthy changes in release 8.25 (2016-01-20) [stable]

** Bug fixes

-  cp now correctly copies files with a hole at the end of the file,
  and extents allocated beyond the apparent size of the file.
  That combination resulted in the trailing hole not being reproduced.
 
-  cut --fields no longer outputs extraneous characters on some uClibc configs.
  
-  install -D again copies relative file names when absolute file names
   are also specified along with an absolute destination directory name.
 
-  ls no longer prematurely wraps lines when printing short file names.

- mv no longer causes data loss due to removing a source directory specified
  multiple times, when that directory is also specified as the destination.
  
-  shred again uses defined patterns for all iteration counts.
 
- sort --debug -b now correctly marks the matching extents for keys
  that specify an offset for the first field.
 
- tail -F now works with initially non existent files on a remote file system.
  
http://savannah.gnu.org/forum/forum.php?forum_id=8445
Comment 1 w.schwarzenfeld 2016-01-23 02:39:41 UTC
Created attachment 165985 [details]
poudriere-coreutils-8.25.log
Comment 2 w.schwarzenfeld 2016-01-23 02:41:46 UTC
Created attachment 165986 [details]
portlint-coreutils-8.25
Comment 3 jharris 2016-02-02 21:29:14 UTC
Approved, thanks!
Comment 4 jharris 2016-02-02 21:57:11 UTC
Created attachment 166462 [details]
patch to correct OP's breakage when NLS isn't selected

This patch is approved.
Comment 5 Kurt Jaeger freebsd_committer freebsd_triage 2016-02-05 21:39:49 UTC
Testbuilds are fine after a bit of tweaking, but there seems to be a usability change, which is discussed on the coreutils mailinglist:

http://lists.gnu.org/archive/html/coreutils/2016-02/msg00000.html

etc.

The following ports depend on coreutils:

deskutils/kdepim4/Makefile
devel/quilt/Makefile
devel/hs-git-annex/Makefile
graphics/variety/Makefile
mail/sqlgrey/Makefile
misc/ansiweather/Makefile
net/dropbox-uploader/Makefile
net/opensips/Makefile
security/openvas-plugins/Makefile
sysutils/renameutils/Makefile
sysutils/tenshi/Makefile
sysutils/createrepo/Makefile
sysutils/confman/Makefile
www/mod_spdy/Makefile

I wonder if they need to be tested ?
Comment 6 Kurt Jaeger freebsd_committer freebsd_triage 2016-02-05 21:40:38 UTC
portmgr, please comment: Does this need a exp-run ?
Comment 7 Antoine Brodin freebsd_committer freebsd_triage 2016-02-05 21:50:43 UTC
The "info" change is totally wrong, please revert.

And I don't think this needs an exp-run, you can test 5 or 6 of these ports yourself.
Comment 8 Kurt Jaeger freebsd_committer freebsd_triage 2016-02-05 21:54:46 UTC
Created attachment 166632 [details]
current-patch

This is the patch which I tested (has the fixed INFO).
Comment 9 Kurt Jaeger freebsd_committer freebsd_triage 2016-02-05 21:57:45 UTC
maintainer: Can you check the usability change from the mailinglist ? Any comments on that ?
Comment 10 jharris 2016-02-08 15:31:41 UTC
Actually, with a little more grepping, I see that only sysutils/renameutils explicitly uses coreutils' gls, the only program identified in the mailing list message you referenced.  Matteo has been added to this bug's CC list.

All other ports explicitly (grep|awk|sort|uniq -c) use just one coreutils program:

   1 gcp
   2 ginstall
   2 gmd5sum
   1 gprintf
   3 greadlink
   1 grep
   1 gsha256sum
   1 gtail

except for graphics/variety, which I have grepped and can't determine where it actually uses any of the coreutils programs (no instances of gcp, gls, etc.)
So, nivit@ is also CC'd.

Matteo and nivit@, will the coreutils gls change described here:

  http://lists.gnu.org/archive/html/coreutils/2016-02/msg00000.html

break your ports I just mentioned?  If you don't mind, please test against this coreutils 8.25 patch:

  https://bugs.freebsd.org/bugzilla/attachment.cgi?id=166632&action=diff

Thanks.
Comment 11 Nicola Vitale freebsd_committer freebsd_triage 2016-02-08 16:14:47 UTC
For graphics/variety see ports/graphics/variety/files/patch-variety_VarietyWindow.py

I don't think that change will break variety.
Comment 12 commit-hook freebsd_committer freebsd_triage 2016-02-26 20:48:58 UTC
A commit references this bug:

Author: pi
Date: Fri Feb 26 20:48:21 UTC 2016
New revision: 409637
URL: https://svnweb.freebsd.org/changeset/ports/409637

Log:
  sysutils/coreutils: 8.23 -> 8.25

  - cp now correctly copies files with a hole at the end of the file,
    and extents allocated beyond the apparent size of the file.
    That combination resulted in the trailing hole not being reproduced.
  - cut --fields no longer outputs extraneous characters on some uClibc
    configs.
  - install -D again copies relative file names when absolute file names
    are also specified along with an absolute destination directory name.
  - ls no longer prematurely wraps lines when printing short file names.
  - mv no longer causes data loss due to removing a source directory specified
    multiple times, when that directory is also specified as the destination.
  - shred again uses defined patterns for all iteration counts.
  - sort --debug -b now correctly marks the matching extents for keys
    that specify an offset for the first field.
  - tail -F now works with initially non existent files on a remote file system.

  Please note: usability issue for 'ls' is discussed in
  http://lists.gnu.org/archive/html/coreutils/2016-02/msg00000.html

  PR:		206512
  Submitted by:	w.schwarzenfeld@aon.at
  Approved by:	jharris@widomaker.com (maintainer)

Changes:
  head/sysutils/coreutils/Makefile
  head/sysutils/coreutils/distinfo
  head/sysutils/coreutils/files/patch-configure
  head/sysutils/coreutils/files/patch-src_stty.c
  head/sysutils/coreutils/files/patch-stty.c
  head/sysutils/coreutils/pkg-plist
Comment 13 Kurt Jaeger freebsd_committer freebsd_triage 2016-02-26 20:50:26 UTC
Committed, thanks!