[pacman-dev] Pacman-3.5.0 bug round-up

[Ray Kohler]

On Sun, Mar 20, 2011 at 4:08 AM, Allan McRae <allan at archlinux.org> wrote:
> Here is a list I made to help keep track of the status of issues in the
> 3.5.0 release that I thought may be on use to others....
>
>
> Fixed:
>
> (No FS#) - reading from stdin and piping "yes"
>  - Fix in git
>
> FS#23307 - [makepkg] unexpected EOF error when parenthesis are used in
> optdepends description
>  - Fix attached to bug report (on Allan's working branch)
>
> FS#23314 - [pacman] segfault when a database includes deltas
>  - Fix attached to bug report (Dan to clean up)
>
> FS#23315 - Makepkg gives fatal error if locale isn't set to en_US.utf8
>  - Fix identified in bug report.  Requires adjustment to translation
>
> FS#23325 - libalpm links to many libraries
>  - "Fix" to workaround libtool bug on mailing list
>
> FS#23342 - --ignore'ing a package given in the arg list
>  - Fix on mailing list (on Allan's working branch)
>
> FS#23343 - Pacman does not set world readable permissions on db files (obeys
> umask instead)
>  - Fix attached to bug report (on Allan's working branch)
>
>
> Bug remaining to be fixed:
>
> FS#23340 - Group selection mechanism and -Sp
>  - is this a case of excess output, or should the selection of packages be
> able to be performed with -Sp?  I guess the former given the -Sp is intended
> to be used in scripts?
>
> FS#23322 - zsh autocompletion no longer works with pacman 3.5
>  - The zsh autocompletion file has not been updated in a while.
>   Do we have a zsh user that wants to tackle this?
>
> (No FS#) - pacman -{S,U,R} --help does not show -p (only --print)
>  - Was this deliberate to keep it beside the --print-format line?

How about:

- usage message does not show "-T" (this is not a new bug, actually)

- difference in handling "-v" and "--debug", if this is even really to
be called a bug at all (
http://mailman.archlinux.org/pipermail/pacman-dev/2011-March/012693.html)

I don't see FS open for either of these. I'll do so if you want.