Re: Stable release of mc-4.7.0.3

2010-03-09 Thread Yury V. Zaytsev
On Sat, 2010-02-27 at 00:20 +0100, Oswald Buddenhagen wrote:

  Other than that, it's a matter of subjective preferences, I guess.
  
 yes, one can also prefer a versioning scheme which converges towards the
 value of pi. this isn't necessarily sane or even useful, though.

Actually, I've just realized that this is not such a bad idea. TeX uses
it quite successfully up to these days. Might be that your post actually
*was* an allusion to TeX though and I'm playing a slowpoke... 
 
-- 
Sincerely yours,
Yury V. Zaytsev

___
Mc-devel mailing list
http://mail.gnome.org/mailman/listinfo/mc-devel


Stable release of mc-4.7.0.3

2010-02-26 Thread Slava Zanko
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi, all.

Download page: http://www.midnight-commander.org/downloads

Major changes since 4.7.0.2.

Core
 * Reorganization of source tree structure (#2037)
 * Added 'menuinactive' skin item to draw inactive visible main menu
(#1999)

Misc
 * Documentation updates
 * Translation updates

Fixes

 * Missing includes (#2017)
 * Memory leaks (#2028, #2053, #2058)
 * Incorrect start up with some special paths (#1992)
 * MC crashes at exit due to race conditions of destroying subshell and
file manager (#2008)
 * Ctrl-\ key closes the NCurses-based MC (#1926)
 * verbose option is always switched on after MC start (#1940)
 * Selections are not visible on monochrome terminals in NCurses-based
MC (#1972)
 * Extra quoting of shell variables in user menu (#1967)
 * Editor's search parameters are not retained across editing session
(#1572)
 * EditColumnMark can't go up through newline (#1998)
 * Missed \s symbol in Syntax file (#2010)
 * ViewContinueSearch segfault on empty search (#1996)
 * Potencial security risk in mcserv (#1902)
 * The lslR VFS doesn't work with ls-lR files created in en_US.UTF-8
locale and with files and directories started with whitespaces (#1921)
 * Contents of RAR archives with filenames that contain / \d\d:\d\d /
are not listed correctly (#2029)
 * FTPFS: strcpy() is used for overlaping strings (#2018)

- --
WBR, dev-team.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)

iD8DBQFLh8DKb3oGR6aVLpoRAnHVAJ9CZYY8ox3LMLY9N6aCr4DKNtsezwCeJLjZ
xAkSVWHOnA0jg757E6XWSWY=
=wCkY
-END PGP SIGNATURE-
___
Mc-devel mailing list
http://mail.gnome.org/mailman/listinfo/mc-devel


Re: Stable release of mc-4.7.0.3

2010-02-26 Thread Oswald Buddenhagen
On Fri, Feb 26, 2010 at 02:38:34PM +0200, Slava Zanko wrote:
 Major changes since 4.7.0.2.
 
why on earth are you again inflating the version namespace? this looks
like a rather regular bugfix release, i.e. 4.7.1. and the what you
called 4.7.1 is pretty much a 4.8.
a.b.c.d releases are only justified if the release tar balls are messed
up somehow or some other kind of release showstopper slipped through.

___
Mc-devel mailing list
http://mail.gnome.org/mailman/listinfo/mc-devel


Re: Stable release of mc-4.7.0.3

2010-02-26 Thread Yury V. Zaytsev
On Fri, 2010-02-26 at 15:27 +0100, Oswald Buddenhagen wrote:

 why on earth are you again inflating the version namespace? this looks
 like a rather regular bugfix release, i.e. 4.7.1. and the what you
 called 4.7.1 is pretty much a 4.8.
 a.b.c.d releases are only justified if the release tar balls are messed
 up somehow or some other kind of release showstopper slipped through.

I was asked to answer you that what they had in mind in the

[epoch].[major].[minor].[release]

scheme, 4.7.1.x being maintenance releases and 4.7.x.0 = 4.7.x being
feature releases. Also, they say that Redhat uses the same versioning
scheme for the kernels they support.

Other than that, it's a matter of subjective preferences, I guess.

My personal opinion: I don't care as long as there's a scheme to which
they actually stick to.
 
-- 
Sincerely yours,
Yury V. Zaytsev

___
Mc-devel mailing list
http://mail.gnome.org/mailman/listinfo/mc-devel


Re: Stable release of mc-4.7.0.3

2010-02-26 Thread Oswald Buddenhagen
On Fri, Feb 26, 2010 at 03:42:47PM +0100, Yury V. Zaytsev wrote:
 I was asked to answer you that what they had in mind in the
 
 [epoch].[major].[minor].[release]
 
that's pretty much a guarantee that the epoch will never change.

 Also, they say that Redhat uses the same versioning scheme for the
 kernels they support.
 
the linux kernel's versioning scheme is an expression of a two-level
generational development model. you may have noticed that this was
dropped years ago - the major version is fixed at 2.6. and you never had
such a model in the first place, and you'll never have. so why would you
introduce such a scheme?

 Other than that, it's a matter of subjective preferences, I guess.
 
yes, one can also prefer a versioning scheme which converges towards the
value of pi. this isn't necessarily sane or even useful, though.
___
Mc-devel mailing list
http://mail.gnome.org/mailman/listinfo/mc-devel