Please merge wesnoth 1.6.2-1 (universe) from Debian unstable (main)

Bug #364069 reported by Dmytro Korzhevin
14
This bug affects 1 person
Affects Status Importance Assigned to Milestone
wesnoth (Ubuntu)
Fix Released
Wishlist
Thierry Carrez

Bug Description

Related branches

Revision history for this message
Rhonda D'Vine (rhonda) wrote :

Please notice that this update would also contain the fix for the font issue for chinese users. The patch applied to the last update unfortunately was wrong - it should get dropped completely and a simple sync is suggested. I've just checked that the ttf-wqy-zenhei package both in Debian unstable and in Ubuntu jaunty contain the font as /usr/share/fonts/truetype/wqy/wqy-zenhei.ttc, according to the packages.ubuntu.com and packages.debian.org filelist listing.

Michael Terry (mterry)
summary: - wesnoth need update to fix release 1.6.1 in Ubuntu 9.04
+ Please sync wesnoth 1.6.1-1 (universe) from Debian unstable (main)
Revision history for this message
Michael Terry (mterry) wrote :

As Gerfried notes, this package can be synced from debian. The only change in Ubuntu is to change the link to wqy-zenhei.ttc to wqy-zenhei.ttf (bug 360347). But that fix is wrong anyway. So it should be overridden with debian's better fix (changing the other half of the link statement from ttf->ttc).

Unfortunately, the debian version depends upon a newer ttf-wqy-zenhei version than is currently in karmic. So this can't be synced until bug 373710 is fixed (a merge request for ttf-wqy-zenhei). I'm working on that as I type. I'll subscribe sponsors once that bug is fixed.

Revision history for this message
ubuntu_demon (ubuntu-demon) wrote : Re: Please sync wesnoth 1.6.1-1 (universe) from Debian unstable (main)

I'm wondering why Wesnoth 1.6a was packaged for Jaunty. This is a pre-release version Wesnoth 1.6. See :
http://www.wesnoth.org/forum/viewtopic.php?p=346711sid=536a75f7ddf8541574c1b0025a77e2f9#p346711

Wesnoth 1.6.2 is available.

debian unstable meta-package :
http://packages.debian.org/sid/wesnoth-all
players changelog ;
http://svn.gna.org/viewcvs/wesnoth/tags/1.6.2/players_changelog?rev=35560&view=download
complete changelog :
http://svn.gna.org/viewcvs/wesnoth/tags/1.6.2/changelog?rev=35560&view=download

Revision history for this message
ubuntu_demon (ubuntu-demon) wrote :

IMHO it would be a good idea to make Wesnoth 1.6.2 available for Karmic Koala and then backport it to earlier Ubuntu versions.

Revision history for this message
Rhonda D'Vine (rhonda) wrote : Re: [Bug 364069] Re: Please sync wesnoth 1.6.1-1 (universe) from Debian unstable (main)

* ubuntu_demon <email address hidden> [2009-05-14 13:42:04 CEST]:
> I'm wondering why Wesnoth 1.6a was packaged for Jaunty. This is a pre-release version Wesnoth 1.6. See :
> http://www.wesnoth.org/forum/viewtopic.php?p=346711sid=536a75f7ddf8541574c1b0025a77e2f9#p346711

 No. 1.6a was _not_ a pre-release version at all, in fact it was a hotfix
release to the first 1.6 release that immediately followed the first 1.6
release. The pre-releases were called Release Candidates, like with most
other sensible projects, too. :)

 One thing about the ttf-wqy-zenhei versioned recommends: It should be
enough to remove the versioning from there. The version of
ttf-wqy-zenhei in ubuntu jaunty at least already has the transitioned
name according to packages.ubuntu.com file list even in its still lower
version number. You propably though will need to change the symlink
target for backports to older releases to match the font name. I am
willing to proofread the required diff to the Debian sources and ACK it,
now that I have figured out where and how the the relevant pieces fit in
Ubuntu. :)

 So long!
Rhonda

Michael Terry (mterry)
summary: - Please sync wesnoth 1.6.1-1 (universe) from Debian unstable (main)
+ Please sync wesnoth 1.6.2-1 (universe) from Debian unstable (main)
Revision history for this message
Emilio Pozuelo Monfort (pochu) wrote : Re: Please sync wesnoth 1.6.2-1 (universe) from Debian unstable (main)

(you forgot to subscribe the sponsors)

It FTBFS in my ppa in karmic:

cc1plus: warnings being treated as errors
game_preferences.cpp: In constructor 'preferences::manager::manager()':
game_preferences.cpp:70: error: 'terrain.std::vector<t_translation::t_terrain, std::allocator<t_translation::t_terrain> >::<anonymous>.std::_Vector_base<t_translation::t_terrain, std::allocator<t_translation::t_terrain> >::_M_impl.std::_Vector_base<t_translation::t_terrain, std::allocator<t_translation::t_terrain> >::_Vector_impl::_M_start' may be used uninitialized in this function

http://launchpadlibrarian.net/27167550/buildlog_ubuntu-karmic-i386.wesnoth_1%3A1.6.2-1_FAILEDTOBUILD.txt.gz

Changed in wesnoth (Ubuntu):
status: New → Incomplete
Revision history for this message
Rhonda D'Vine (rhonda) wrote : Re: [Bug 364069] Re: Please sync wesnoth 1.6.2-1 (universe) from Debian unstable (main)

* Emilio Pozuelo Monfort <email address hidden> [2009-05-26 12:17:54 CEST]:
> (you forgot to subscribe the sponsors)

 Which "you"? ;)

> It FTBFS in my ppa in karmic:
>
> cc1plus: warnings being treated as errors
> game_preferences.cpp: In constructor 'preferences::manager::manager()':
> game_preferences.cpp:70: error: 'terrain.std::vector<t_translation::t_terrain, std::allocator<t_translation::t_terrain> >::<anonymous>.std::_Vector_base<t_translation::t_terrain, std::allocator<t_translation::t_terrain> >::_M_impl.std::_Vector_base<t_translation::t_terrain, std::allocator<t_translation::t_terrain> >::_Vector_impl::_M_start' may be used uninitialized in this function

 That's a known bug in gcc 4.4. Either get it explicitly to use gcc 4.3
or make it use gcc 4.5 in which this bug got fixed again. See also these
reports about it:
<http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=528587>
<http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40156>

 So long!
Rhonda

Revision history for this message
Emilio Pozuelo Monfort (pochu) wrote : Re: [Bug 364069] Re: Please sync wesnoth 1.6.2-1 (universe) from Debian unstable (main)

Gerfried Fuchs escribió:
> * Emilio Pozuelo Monfort <email address hidden> [2009-05-26 12:17:54 CEST]:
>> (you forgot to subscribe the sponsors)
>
> Which "you"? ;)

Whoever switched the bug report into a sync request, so not "you" ;)

>> It FTBFS in my ppa in karmic:
>>
>> cc1plus: warnings being treated as errors
>> game_preferences.cpp: In constructor 'preferences::manager::manager()':
>> game_preferences.cpp:70: error: 'terrain.std::vector<t_translation::t_terrain, std::allocator<t_translation::t_terrain> >::<anonymous>.std::_Vector_base<t_translation::t_terrain, std::allocator<t_translation::t_terrain> >::_M_impl.std::_Vector_base<t_translation::t_terrain, std::allocator<t_translation::t_terrain> >::_Vector_impl::_M_start' may be used uninitialized in this function
>
> That's a known bug in gcc 4.4. Either get it explicitly to use gcc 4.3
> or make it use gcc 4.5 in which this bug got fixed again. See also these
> reports about it:
> <http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=528587>
> <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40156>

Thanks for the pointers! I'll look into it at some point unless the gcc bug is
fixed before ;) If not I'll probably switch to building with gcc 4.3.

Do you know if I only have to define CC in rules and b-depend on gcc-4.3 ?

Revision history for this message
Ilya Barygin (randomaction) wrote : Re: Please sync wesnoth 1.6.2-1 (universe) from Debian unstable (main)

An easy workaround is just to suppress the warning (-Wno-uninitialized). However, gcc still complains:

cc1plus: warnings being treated as errors
network.cpp: In member function 'virtual void network::<unnamed>::connect_operation::run()':
network.cpp:432: error: dereferencing type-punned pointer will break strict-aliasing rules
[4 more errors]

This can be fixed by adding -fno-strict-aliasing.

Changed in wesnoth (Ubuntu):
status: Incomplete → New
Thierry Carrez (ttx)
summary: - Please sync wesnoth 1.6.2-1 (universe) from Debian unstable (main)
+ Please merge wesnoth 1.6.2-1 (universe) from Debian unstable (main)
Changed in wesnoth (Ubuntu):
status: New → Confirmed
Revision history for this message
Thierry Carrez (ttx) wrote :

Could you please resubmit the debdiff as a merge request that includes the ftbfs.patch with the following fixes:
- Set maintainer as MOTU (universe) rather than ubuntu-dev (main)
- Explain in changelog why the previous delta (the ttf-wqy-zenhei stuff) was dropped

Revision history for this message
Michael Terry (mterry) wrote :

Thierry, his maintainer change is correct. In May, the Maintainer policy was changed to reflect the upcoming main/universe changes. https://wiki.ubuntu.com/DebianMaintainerField

Note that the name is "Ubuntu Developers" not "Ubuntu Core Developers"

Revision history for this message
Ilya Barygin (randomaction) wrote :

Not sure that it's really a merge since we drop the only Ubuntu change... but here it is. Maybe it's also worth listing LP: #360347 as fixed.

Revision history for this message
Ilya Barygin (randomaction) wrote :

And Ubuntu to Ubuntu debdiff, just in case.

Thierry Carrez (ttx)
Changed in wesnoth (Ubuntu):
assignee: nobody → Thierry Carrez (ttx)
importance: Undecided → Wishlist
status: Confirmed → In Progress
Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package wesnoth - 1:1.6.2-1ubuntu1

---------------
wesnoth (1:1.6.2-1ubuntu1) karmic; urgency=low

  * Merge from Debian unstable (LP: #364069)
  * Dropped Ubuntu changes:
     - .ttc -> .ttf for Chinese font link
       (fixed in Debian for ttf-wqy-zenhei >= 0.8.38-1, fixes LP: #360347)
  * debian/rules: added compiler flags to fix FTBFS due to
    GCC bug 40156 and warnings in src/network.cpp

wesnoth (1:1.6.2-1) unstable; urgency=low

  * New upstream release.
  * Patch refreshed.
  * Updated wqy-zenhei overrides entry.
  * Added wesnoth-sof overrides entry for I in description.
  * Actually get the overrides file installed through dh_lintian, naming the
    files accordingly.

wesnoth (1:1.6.1-1) unstable; urgency=low

  * New upstream release.
  * Incorporated no-music-endless-loop, patch removed.
  * Fixed wqy-zenhei to point to proper target and added versioning to
    Recommends on ttf-wqy-zenhei for that change
    (closes: #522501, LP: #358590)

 -- Ilya Barygin <email address hidden> Mon, 15 Jun 2009 17:44:00 +0400

Changed in wesnoth (Ubuntu):
status: In Progress → Fix Released
Revision history for this message
Ilya Barygin (randomaction) wrote :

Failed to build on armel.

checking for arm-linux-gnueabi-gcc... arm-linux-gnueabi-gcc
checking for C compiler default output file name...
configure: error: in `/build/buildd/wesnoth-1.6.2':
configure: error: C compiler cannot create executables
See `config.log' for more details.
make: *** [build-arch-stamp] Error 77

Revision history for this message
Rhonda D'Vine (rhonda) wrote : Re: [Bug 364069] Re: Please merge wesnoth 1.6.2-1 (universe) from Debian unstable (main)

* Ilya B <email address hidden> [2009-06-17 09:12:53 CEST]:
> Failed to build on armel.
>
> checking for arm-linux-gnueabi-gcc... arm-linux-gnueabi-gcc
> checking for C compiler default output file name...
> configure: error: in `/build/buildd/wesnoth-1.6.2':
> configure: error: C compiler cannot create executables
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

 That sounds more like a b0rked build chroot. Is the full build log
available, or:

> See `config.log' for more details.
  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

 ... the content of the config.log file?

 Thanks,
Rhonda

Revision history for this message
Thierry Carrez (ttx) wrote :

Everything is failing to build right now with the same error, looking at https://launchpad.net/builders/nageia/+history

Revision history for this message
Ilya Barygin (randomaction) wrote :

I contacted upstream about the warnings in network.cpp, and they say that it's not a bug in Wesnoth (and that we should probably ignore the warning rather than turn off optimization... well, maybe).
http://gna.org/bugs/?13711

Revision history for this message
Rhonda D'Vine (rhonda) wrote :

* Ilya B <email address hidden> [2009-06-18 13:09:16 CEST]:
> I contacted upstream about the warnings in network.cpp, and they say
> that it's not a bug in Wesnoth (and that we should probably ignore the
> warning rather than turn off optimization... well, maybe).
> http://gna.org/bugs/?13711

 Erm ... Did you read the bug log? I commented on exactly this already
and it got acknowledged by gcc people even. Let me quote for your
pleasure again:

,-------------------------------> quote <-------------------------------
| That's a known bug in gcc 4.4. Either get it explicitly to use gcc 4.3
| or make it use gcc 4.5 in which this bug got fixed again. See also these
| reports about it:
| <http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=528587>
| <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40156>
`-------------------------------> quote <-------------------------------

 No need to bring up an issue that everyone agreed upon already. :)

 ... *puzzled*
Rhonda

Revision history for this message
Ilya Barygin (randomaction) wrote :

You're talking about "uninitialized" warnings, and I'm talking about aliasing rules violation in another source file, which is a separate issue.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.