--- Begin Message ---
Package: wine
Version: 1.6.2-8
Severity: normal
Dear Maintainer,
This does not happens with Ubuntu packages of Wine 1.6.2 on an Ubuntu machine,
but happens with recent versions of the Debian (jessie) packaged version of
wine : I reproduced it on three Debian (jessie) machines.
The installation runs normally. When starting Winword for the first time, it
aks for name and initial, then displays a "configuration" progress bar, then
asks for a reboot.
If one clicks on yes, thge rebooot fails and gives an error message. When
asking for details, one gets the enclosed backtrace.
If one refuses to reboot, the normal Winword appears, but cannot be activated.
In short, the result is an unusable MS Office.
This seems to be a recent regression : an installation worked fine in february,
IIRC. But the Debian package is more recent than that. It might also be Debian-
specific.
Tried wine-development : ditto
Tried to create a wine64 prefix and install the 64 bit version : ditto.
I reported this on Wine's Bugzilla
(https://bugs.winehq.org/show_bug.cgi?id=37222). One other user could nor
reproduce (on openSUSE), and another people concluded (how ?) "Likely a
packaging bug".
Hence the present report.
-- System Information:
Debian Release: jessie/sid
APT prefers testing
APT policy: (650, 'testing'), (500, 'testing-updates'), (500, 'testing-proposed-updates'), (60, 'unstable'), (50, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386
Kernel: Linux 3.14-2-amd64 (SMP w/8 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Versions of packages wine depends on:
ii file 1:5.19-1
ii wine32 1.6.2-8
ii wine64 1.6.2-8
wine recommends no packages.
Versions of packages wine suggests:
ii binfmt-support 2.1.4-1
ii clamav 0.98.4+dfsg-2+b1
ii ttf-mscorefonts-installer 3.5
ii winbind 2:4.1.11+dfsg-1
pn wine-doc <none>
-- no debconf information
--- End Message ---
--- Begin Message ---
On Tue, Sep 23, 2014 at 4:14 PM, Emmanuel Charpentier wrote:
>> Wine doesn't really support unix paths anymore.
>
> Wouldn't that be a regression ?
No, it's simply a design decision made by upstream a few years ago.
It would be considered backwards compatibility breakage rather than
regression.
If you disagree with that change, you can start an upstream
discussion, but it's unlikely to go anywhere now that the new behavior
has been around for a few years.
> My question stands : should I reopen the Bugzilla report as a
> regression, a bug or leave it at that ?
No, it's not a bug. It's a user education/expectation issue.
Best wishes,
Mike
--- End Message ---