Bug#1057565: state of kalzium package, and metapackage dependencies on it.

[Peter Green]

kalzium needs to be rebuilt for the time64 transition, but it has had
a FTBFS bug with no maintainer response for 4 months. The only reverse
dependencies seem to be a number of metapackages.

In particular, the kdeedu package is a key package and has a hard
dependency on kalzium. This means that it can't be autoremoved from
testing, making it a blocker the time64 transition.

Is there someone who can step up and fix kalzium? or should
it be dropped from the metapackages so it can be removed from testing?

Metapackages built from the meta-kde source (key, hard dependencies)

* kdeedu

Metapackages built from the debian-edu source (key, but only reccomends):

* education-chemistry
* education-highschool
* education-primaryschool
* education-secondaryschool

Metapackages built from the debian-science source (not key, only reccomends):

* science-chemistry

Metapackages built from the debichem source (not key, only reccomends):

* debichem-visualisation

Bug#1067906: qtwebengine-opensource-src - FTBFS on armhf.

[Peter Green]

Package: qtwebengine-opensource-src
Version: 5.15.15+dfsg-2
Severity: serious

qtwebengine-opensource-src failed to build on armhf when binnmu'd for the time_t
transition due to symbol changes.
(qtwebengine does not support any of the other architectures affected by
the time64 transition.

grep MISSING qtwebengine-opensource-src.log | grep -v optional
+#MISSING: 5.15.15+dfsg-2+b3# 
_ZN15QtWebEngineCore14ProfileAdapter21determineDownloadPathERK7QStringS3_RKl@Qt_5
 5.14.1
+#MISSING: 5.15.15+dfsg-2+b3# _ZN7sandbox18localtime_overrideEPKl@Qt_5 5.12.2
+#MISSING: 5.15.15+dfsg-2+b3# _ZN7sandbox20localtime64_overrideEPKl@Qt_5 5.12.2
+#MISSING: 5.15.15+dfsg-2+b3# _ZN7sandbox20localtime_r_overrideEPKlP2tm@Qt_5 
5.12.2
+#MISSING: 5.15.15+dfsg-2+b3# _ZN7sandbox22localtime64_r_overrideEPKlP2tm@Qt_5 
5.12.2

Bug#973703: cantor-backend-scilab: dependency no longer available on i386 and armhf

[peter green]

Package: cantor-backend-scilab
Version: 4:20.04.3-1
Severity: serious

As a result of swt dropping support for 32-bit targets, libjogl2-java can no 
longer be built on i386 and armhf and the
existing binaries have been removed in unstable. This in turn has lead to the 
removal of scilab on those architectures.

Please adjust the architecture list for cantor-backend-scilab accordingly.

Bug#949177: libaccounts-glib appears to be unmaintained.

[peter green]

The libaccounts-glib package does not seem to be getting maintained in Debian, 
it has had a rc bug (deprecated declarations
and -Werror) for 8 months with no maintainer response and has recently picked 
up a second one (unversioned python removal).
It has not had an upload for 3 years.

Both of the rc bugs have been fixed in ubuntu. The ubuntu package has a higher 
"upstream version" than the Debian one,
but this does not seem to actually represent a newer version from upstream. The 
changes outside the Debian directory
between the "upstream version" in Debian and the one in Ubuntu seem to be 
limited to a small number of changes in the
"tests" directory (changing a timeout and removing some gitignore files) and I 
could not find corresponding changes
in the upstream vcs. It seems the Debian and Ubuntu packaging diverged back in 
2016.

There does seem to be activity upstream with a release 5 months ago, but 
neither Debian or Ubuntu seem to have packaged it.

So I started looking at the reverse dependencies to see if this was a package 
that was still being used and I found quite
a bit. Looking at the first two levels I get:

direct:

ayatana-indicator-datetime
kaccounts-integration
kaccounts-providers
kio-gdrive
ktp-accounts-kcm
libaccounts-qt
purpose
telepathy-accounts-signon

second levels:

ktp-common-internals (via 
kaccounts-integration/libaccounts-qt/telepathy-accounts-signon)
ktp-auth-handler (via kaccounts-integration/libaccounts-qt)
meta-kde-telepathy (via ktp-accounts-kcm)
kaccounts-integration (via libaccounts-qt)
akonadi (via libaccounts-qt)
signon-ui (via libaccounts-qt)
choqok (via purpose)
falkon (via purpose)
gwenview (via purpose)
kamaso (via purpose)
kde-spectacle (via purpose)
kdenlive (via purpose)
kdeplasma-addons (via purpose)
kdevelop (via purpose)
kphotoalbum (via purpose)
libkf5pimcommon (via purpose)
okular (via purpose)
plasma-browser-integration (via purpose)

So, it doesn't look like removal is an option. Can someone (perhaps the 
maintainer of one of the reverse-dependencies
take a look at getting this package back into shape?). Bringing across the 
fixes from the Ubuntu package looks like it
would be fairly easy but I have no idea how to test the results.

Re: build failure while updating pyside2 in raspbian

[peter green]

On 24/05/2020 16:01, peter green wrote:


(minimal) Clang: 1 diagnostic messages:
  /pyside2/sources/shiboken2/tests/libminimal/listuser.h:32:10: fatal: 'list' 
file not found
   /tmp/global_pjsLqX.hpp:1:10: note: in file included from 
/tmp/global_pjsLqX.hpp:1:
/pyside2/sources/shiboken2/tests/minimalbinding/global.h:32:10: note: in file 
included from /pyside2/sources/shiboken2/tests/minimalbinding/global.h:32:




Any idea what might be going wrong and what can be done about it?


I found the issue myself, it seems that the issue was caused by having 
libgcc-10-dev but not libstdc++-10-dev installed in the build environment and 
this causes clang to use the gcc-10 directories (which contain no c++ stuff 
because libstdc++-10-dev is not installed) rather than the gcc-9 directories 
(which do contain the c++ stuff).

See https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=961547 for more details.

build failure while updating pyside2 in raspbian

[peter green]

Hi, I am trying to update pyside2 in raspbian bullseye to bring it back into 
sync with Debian, we carry a couple of local patches to the packaging to build 
without qtwebengine (which we don't have in raspbian, but serveral Debian 
architectures don't have it either) and to allow the documentation packages to 
be built on 32-bit.

Unfortunately after merging our patches and trying to build I get


(minimal) clang_parseTranslationUnit2(0x0, 
cmd[6]=-isystem/usr/lib/llvm-9/lib/clang/9/include -fPIC 
-Wno-constant-logical-operand -std=c++14 
-I/pyside2/sources/shiboken2/tests/libminimal /tmp/global_pjsLqX.hpp)
/pyside2/sources/shiboken2/tests/libminimal/listuser.h:32:10: fatal error: 
'list' file not found
(minimal) Errors in /tmp/global_pjsLqX.hpp:
/pyside2/sources/shiboken2/tests/libminimal/listuser.h:32:10: fatal: 'list' 
file not found
   /tmp/global_pjsLqX.hpp:1:10: note: in file included from 
/tmp/global_pjsLqX.hpp:1:
/pyside2/sources/shiboken2/tests/minimalbinding/global.h:32:10: note: in file 
included from /pyside2/sources/shiboken2/tests/minimalbinding/global.h:32:


(minimal) Clang: 1 diagnostic messages:
  /pyside2/sources/shiboken2/tests/libminimal/listuser.h:32:10: fatal: 'list' 
file not found
   /tmp/global_pjsLqX.hpp:1:10: note: in file included from 
/tmp/global_pjsLqX.hpp:1:
/pyside2/sources/shiboken2/tests/minimalbinding/global.h:32:10: note: in file 
included from /pyside2/sources/shiboken2/tests/minimalbinding/global.h:32:


Keeping temporary file: /tmp/global_pjsLqX.hpp
shiboken: Error running ApiExtractor.
Command line: 
--project-file=/pyside2/pyside3_build/py3.8-qt5.12.5-32bit-relwithdebinfo/shiboken2/tests/minimalbinding/minimal-binding.txt
make[4]: *** 
[tests/minimalbinding/CMakeFiles/minimal_generator.dir/build.make:66: 
tests/minimalbinding/mjb_rejected_classes.log] Error 1
make[4]: Leaving directory 
'/pyside2/pyside3_build/py3.8-qt5.12.5-32bit-relwithdebinfo/shiboken2'
make[3]: *** [CMakeFiles/Makefile2:2789: 
tests/minimalbinding/CMakeFiles/minimal_generator.dir/all] Error 2
make[3]: *** Waiting for unfinished jobs


Any idea what might be going wrong and what can be done about it?

Bug#948764: pyside2: arch all FTBFS on 32-bit architectures.

[peter green]

Package: pyside2
Version: 5.13.2-2.2
Severity: important
Tags: patch

Doing a full build (architecture dependent and independent packages) of pyside2 
fails on 32-bit architectures (a build that only builds architecture-dependent 
packages seems to be fine). I noticed this while working on updating pyside2 in 
raspbian, but it's happening on the reproducible builds site too, so it's not 
raspbian-specific.

https://tests.reproducible-builds.org/debian/rb-pkg/unstable/i386/pyside2.html


dh_installdocs -O--buildsystem=pybuild
dh_installdocs: Cannot auto-detect main package for python-pyside2-doc.  If the 
default is wrong, please use --doc-main-package
dh_installdocs: Cannot find (any matches for) 
"pyside3_build/py3.7-qt5.12.5-64bit-relwithdebinfo/shiboken2/doc/html" (tried 
in ., debian/tmp)

The fix is trivial, just a replacement of 64 with ?? in the path.

https://github.com/raspbian-packages/pyside2/commit/db69429c836fcdc21a07cd60dd32980a5d525790

Bug#942129: kio: no multiarch support.

[peter green]

Package: kio

While trying to install a kde app for a foreign architecture. I noticed that 
while libkf5declarative5 is multi-arch: same it can only be installed for one 
architecture at a time, because it depends on kio and kio has no multi-arch 
field.

Unfortunately it seems kio contains a bunch of stuff in architecture-specific 
paths which look like they may be used from other packages, but also contains 
native binaries in /usr/bin , so it seems adding multi-arch support would 
require splitting the package.

Thoughts?

(I did not put a version number on the bug because this situation seems to be 
the same in all versions of Debian I looked at)

Bug#917200: pyside2: build-dependencies no longer satisfiable on mips64el.

[peter green]

Package: pyside2
Version: 5.11.2-1
Severity: serious

It seems that patchelf has been removed on mips64el due to testsuite failures, 
but pyside2 still bulid-depends on it. So pyside cannot be built on mips64el 
anymore.

Possible fixes (in roughly descending order of preferences)
1. Fix the testsuite failures in patchelf
2. Break the dependency chain (I notice the patchelf build-dependency was only 
added recently, is it mandatory?)
3. Request removal of the pyside2 binaries on mips64el

Bug#911502: calligra FTBFS 'memCheck' is not a member of 'Object'

[peter green]

Package: calligra
Version: 1:3.1.0+dfsg-3
Severity: serious.

During the rebuild for the poppler transition calligra failed to build with


/<>/calligra-3.1.0+dfsg/filters/karbon/pdf/PdfImport.cpp: In member function 
'virtual KoFilter::ConversionStatus PdfImport::convert(const QByteArray&, const 
QByteArray&)':
/<>/calligra-3.1.0+dfsg/filters/karbon/pdf/PdfImport.cpp:106:13: 
error: 'memCheck' is not a member of 'Object'
  Object::memCheck(stderr);


Since this is in a pdf related file I presume the new poppler was the cause of 
the build failure.

Bug#907013: calligra FTBFS, missing QT includes.

[peter green]

Package: calligra
Version: 3.1.0+dfsg-2
Severity: serious
Tags: patch

Looks like calligra recently started to FTBFS, I first saw this in raspbian but 
it is also happening on the reproducible builds test infrastructure, so it's 
not raspbian specific.

http://buildd.raspbian.org/status/fetch.php?pkg=calligra=armhf=1%3A3.1.0%2Bdfsg-2%2Bb5=1534965502
https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/calligra.html


/<>/calligra-3.1.0+dfsg/libs/widgets/KoPageLayoutWidget.cpp:157:33: 
error: invalid use of incomplete type 'class QButtonGroup'
  Q_ASSERT(d->orientationGroup->button( layout.orientation ));
  ^~
In file included from 
/usr/include/arm-linux-gnueabihf/qt5/QtWidgets/qradiobutton.h:44,
  from 
/usr/include/arm-linux-gnueabihf/qt5/QtWidgets/QRadioButton:1,
  from 
/<>/calligra-3.1.0+dfsg/obj-arm-linux-gnueabihf/libs/widgets/ui_KoPageLayoutWidget.h:22,
  from 
/<>/calligra-3.1.0+dfsg/libs/widgets/KoPageLayoutWidget.cpp:22:
/usr/include/arm-linux-gnueabihf/qt5/QtWidgets/qabstractbutton.h:53:7: note: 
forward declaration of 'class QButtonGroup'
  class QButtonGroup;


I presume this was triggered by a new version of one or more qt libraries.

The FTBFS can be fixed by explicitly adding includes for the incomplete types.

Debdiff attatched, no intent to NMU.

diff -Nru calligra-3.1.0+dfsg/debian/changelog 
calligra-3.1.0+dfsg/debian/changelog
--- calligra-3.1.0+dfsg/debian/changelog2018-03-09 08:37:57.0 
+
+++ calligra-3.1.0+dfsg/debian/changelog2018-08-22 20:34:52.0 
+
@@ -1,3 +1,9 @@
+calligra (1:3.1.0+dfsg-2+rpi1) buster-staging; urgency=medium
+
+  * Add missing qt includes.
+
+ -- Peter Michael Green   Wed, 22 Aug 2018 20:34:52 
+
+
 calligra (1:3.1.0+dfsg-2) unstable; urgency=medium
 
   * Team upload.
diff -Nru calligra-3.1.0+dfsg/debian/patches/add-missing-includes.patch 
calligra-3.1.0+dfsg/debian/patches/add-missing-includes.patch
--- calligra-3.1.0+dfsg/debian/patches/add-missing-includes.patch   
1970-01-01 00:00:00.0 +
+++ calligra-3.1.0+dfsg/debian/patches/add-missing-includes.patch   
2018-08-22 20:34:52.0 +
@@ -0,0 +1,102 @@
+Description:  Add missing qt includes.
+Author: Peter Michael Green 
+Last-Update: 2018-08-22
+
+Index: calligra-3.1.0+dfsg/libs/widgets/KoPageLayoutWidget.cpp
+===
+--- calligra-3.1.0+dfsg.orig/libs/widgets/KoPageLayoutWidget.cpp
 calligra-3.1.0+dfsg/libs/widgets/KoPageLayoutWidget.cpp
+@@ -22,6 +22,7 @@
+ #include 
+ 
+ #include 
++#include 
+ 
+ class Q_DECL_HIDDEN KoPageLayoutWidget::Private
+ {
+Index: calligra-3.1.0+dfsg/libs/widgets/KoCsvImportDialog.cpp
+===
+--- calligra-3.1.0+dfsg.orig/libs/widgets/KoCsvImportDialog.cpp
 calligra-3.1.0+dfsg/libs/widgets/KoCsvImportDialog.cpp
+@@ -37,6 +37,8 @@
+ 
+ #include "ui_KoCsvImportDialog.h"
+ 
++#include 
++
+ class KoCsvImportWidget : public QWidget, public Ui::KoCsvImportWidget
+ {
+ Q_OBJECT
+Index: calligra-3.1.0+dfsg/plugins/chartshape/dialogs/TableEditorDialog.cpp
+===
+--- calligra-3.1.0+dfsg.orig/plugins/chartshape/dialogs/TableEditorDialog.cpp
 calligra-3.1.0+dfsg/plugins/chartshape/dialogs/TableEditorDialog.cpp
+@@ -33,6 +33,7 @@
+ #include "ChartTableView.h"
+ #include "ChartDebug.h"
+ 
++#include 
+ 
+ using namespace KoChart;
+ 
+Index: calligra-3.1.0+dfsg/words/part/dialogs/KWRunAroundProperties.cpp
+===
+--- calligra-3.1.0+dfsg.orig/words/part/dialogs/KWRunAroundProperties.cpp
 calligra-3.1.0+dfsg/words/part/dialogs/KWRunAroundProperties.cpp
+@@ -28,6 +28,8 @@
+ 
+ #include 
+ 
++#include 
++
+ KWRunAroundProperties::KWRunAroundProperties(FrameConfigSharedState *state)
+ : m_state(state)
+ {
+Index: calligra-3.1.0+dfsg/words/part/dialogs/KWAnchoringProperties.cpp
+===
+--- calligra-3.1.0+dfsg.orig/words/part/dialogs/KWAnchoringProperties.cpp
 calligra-3.1.0+dfsg/words/part/dialogs/KWAnchoringProperties.cpp
+@@ -36,6 +36,7 @@
+ #include 
+ 
+ #include 
++#include 
+ 
+ const int KWAnchoringProperties::vertRels[4][20] = {
+ { // KoShapeAnchor::AnchorAsCharacter
+Index: calligra-3.1.0+dfsg/plugins/formulashape/FormulaToolWidget.cpp
+===
+--- calligra-3.1.0+dfsg.orig/plugins/formulashape/FormulaToolWidget.cpp
 calligra-3.1.0+dfsg/plugins/formulashape/FormulaToolWidget.cpp
+@@ -31,6 +31,7 @@
+ #include 
+ #include 
+ #include 
++#include 
+ 
+ FormulaToolWidget::FormulaToolWidget( KoFormulaTool* tool, QWidget* parent )
+   : QTabWidget( parent )
+Index: 

Bug#879094: Status of the marble package.

[peter green]

Marble currently FTBFS in unstable due to a couple of missing symbols and this 
is preventing the libgps transition from finishing up. The version in 
experimental fixes the issue but it looks like uploading it to unstable would 
start a transition.

What are the plans here? are the symbols that disappeared just template cruft 
or are they real ABI changes? are there any plans to transition the 
experimental version to unstable in the near future.

Bug#840071: calligra FTBFS, failure to find postgresql.

[peter green]

tags 840071 +patch
thanks

On 07/10/16 23:50, peter green wrote:
I belive this was triggered by the removal of postgresql 9.5 from the 
list of supported postgresql versions. I belive the fix is to add 9.6 
to the list of known postgresql versions in 
FindCalligraPostgreSQL.cmake , testing (in raspbian) now.

I can confirm that this fixed the issue for us.

Our debdiff is at 
http://debdiffs.raspbian.org/main/c/calligra/calligra_2.9.11+dfsg-2+rpi1.debdiff 
(note that said debdiff also contains a workaround for an ICE we 
encountered in raspbian)

Bug#840071: calligra FTBFS, failure to find postgresql.

[peter green]

package: calligra
serverity: serious
version: 1:2.9.11+dfsg-2

Calligra FTBFS because it fails to find postgresql resulting in missing 
files.


I first saw this in when doing a build for raspbian stretch, but it can 
also be seen in the Debian reproducable builds tests and in the 
autobuild results for minor debian architectures. For example from the 
mips64el build log.



 -- Could NOT find PostgreSQL (missing:  PostgreSQL_TYPE_INCLUDE_DIR) (found version 
"9.6.0")

<--snip-->

-
-- The following OPTIONAL packages could NOT be located on your system.
-- Consider installing them to enable more features from this software.
-

<--snip-->

   * libpq
 C application programmer's interface to PostgreSQL
 Required by Kexi PostgreSQL driver

<--snip>

dh_install --list-missing
dh_install: Cannot find (any matches for) "usr/lib/kde4/kexidb_pqxxsqldriver.so" (tried in 
"." and "debian/tmp")
dh_install: kexi-postgresql-driver missing files: 
usr/lib/kde4/kexidb_pqxxsqldriver.so
dh_install: Cannot find (any matches for) "usr/lib/kde4/keximigrate_pqxx.so" (tried in 
"." and "debian/tmp")
dh_install: kexi-postgresql-driver missing files: 
usr/lib/kde4/keximigrate_pqxx.so
dh_install: Cannot find (any matches for) 
"usr/share/kde4/services/calligra/kexidb_pqxxsqldriver.desktop" (tried in "." and 
"debian/tmp")
dh_install: kexi-postgresql-driver missing files: 
usr/share/kde4/services/calligra/kexidb_pqxxsqldriver.desktop
dh_install: Cannot find (any matches for) 
"usr/share/kde4/services/calligra/keximigrate_pqxx.desktop" (tried in "." and 
"debian/tmp")
dh_install: kexi-postgresql-driver missing files: 
usr/share/kde4/services/calligra/keximigrate_pqxx.desktop



I belive this was triggered by the removal of postgresql 9.5 from the 
list of supported postgresql versions. I belive the fix is to add 9.6 to 
the list of known postgresql versions in FindCalligraPostgreSQL.cmake , 
testing (in raspbian) now.

Bug#832777: muon-updater depends on cruft package

[Peter Green]

package: muon-updater
version: 4:5.7.1-1
severity: serious

muon-notfier and muon-updater depend on plasma-discover-updater which is 
no longer built by the plasma-discover source package.

Bug#797389: calligra: FTBFS, multiple issues.

[Peter Green]

On 03/07/16 16:55, Peter Green wrote:

On 14/06/16 07:10, Scott Kitterman wrote:

On Tuesday, June 14, 2016 01:21:26 AM peter green wrote:
I am reluctant to NMU this In Debian myself because I don't use 
calligra

personally and I have not had any feedback from anyone on the patched
package.
Considering it doesn't build at all and the binaries have been 
removed from
Testing and Unstable, I think the chance of you making it worse is 
nil.  If

you can get it building again, then I wouldn't wait.
Ok, I just set up a sid VM, built calligra with my, installed it and 
did a quick smoke test and it seemed to run (I was able to start the 
writer tool and create a blank document). So i've uploaded it. Debdiff 
attatched (which is the same as what I had in raspbian except for 
changelog changes)



Well I tried to upload it. Unfortunately I was unsuccessful.

The first attempt was rejected because I screwed up the target distribution.

The second attempt was rejected because source only uploads to new are 
not allowed.


The third attempt went into the new queue where it was rejected with the 
following

unfortunately I have to reject the package.

As 
calligra-2.8.5+dfsg/kexi/webforms/webroot/default/media/javascripts/lightbox.js
is licensed under CC 2.5, please remove it from the source tarball.

As well, lintian complains about the lena picture, which is non-free.

While you are at it, please take also care of all those lintian warnings
related to debian/copyright.

   

Bug#797389: calligra: FTBFS, multiple issues.

[Peter Green]

On 14/06/16 07:10, Scott Kitterman wrote:

On Tuesday, June 14, 2016 01:21:26 AM peter green wrote:
   

I am reluctant to NMU this In Debian myself because I don't use calligra
personally and I have not had any feedback from anyone on the patched
package.
 

Considering it doesn't build at all and the binaries have been removed from
Testing and Unstable, I think the chance of you making it worse is nil.  If
you can get it building again, then I wouldn't wait.
   
Ok, I just set up a sid VM, built calligra with my, installed it and did 
a quick smoke test and it seemed to run (I was able to start the writer 
tool and create a blank document). So i've uploaded it. Debdiff 
attatched (which is the same as what I had in raspbian except for 
changelog changes)


diff -Nru calligra-2.8.5+dfsg/debian/changelog 
calligra-2.8.5+dfsg/debian/changelog
--- calligra-2.8.5+dfsg/debian/changelog2015-06-21 07:38:53.0 
+0100
+++ calligra-2.8.5+dfsg/debian/changelog2016-07-03 16:51:02.0 
+0100
@@ -1,3 +1,14 @@
+calligra (1:2.8.5+dfsg-1.3) stretch-staging; urgency=medium
+
+  * Non-maintainer upload.
+  * Add build-depends on libsoprano-dev (Closes: 797389)
+  * Drop kexi-map-form-widget and calligra-reports-map-element  binary packages
+and drop build-dependencies on libmarble-dev. It seems marble has moved
+entirely to qt5 and is hence incompatble with this package. (Closes: 
818395)
+  * Remove dependency on kde-runtime-dbg which no longer exists.
+
+ -- Peter Michael Green <plugw...@debian.org>  Sun, 03 Jul 2016 12:06:15 +
+
 calligra (1:2.8.5+dfsg-1.2) unstable; urgency=medium
 
   * Non-maintainer upload.
diff -Nru calligra-2.8.5+dfsg/debian/control calligra-2.8.5+dfsg/debian/control
--- calligra-2.8.5+dfsg/debian/control  2015-06-21 07:38:20.0 +0100
+++ calligra-2.8.5+dfsg/debian/control  2016-06-12 23:17:41.0 +0100
@@ -27,7 +27,6 @@
libkactivities-dev (>= 4:4.10),
libkdcraw-dev (>= 4:4.12),
liblcms2-dev (>= 2.4),
-   libmarble-dev (>> 4:4.12),
libmysqlclient-dev,
libodfgen-dev (>= 0.1),
libopencolorio-dev,
@@ -59,7 +58,8 @@
pstoedit,
shared-desktop-ontologies,
shared-mime-info,
-   zlib1g-dev
+   zlib1g-dev,
+   libsoprano-dev
 Standards-Version: 3.9.5
 XS-Testsuite: autopkgtest
 Homepage: http://www.calligra-suite.org/
@@ -273,23 +273,6 @@
  .
  This package is part of the Calligra Suite.
 
-Package: kexi-map-form-widget
-Architecture: any
-Section: database
-Depends: kexi (>= ${source:Version}), ${misc:Depends}, ${shlibs:Depends}
-Breaks: kexi (<< 1:2.3.70)
-Replaces: kexi (<< 1:2.3.70)
-Description: map form widget for Kexi
- Kexi is an integrated data management application. It can be used for
- creating database schemas, inserting data, performing queries, and
- processing data. Forms can be created to provide a custom interface to
- your data. All database objects - tables, queries and forms - are stored
- in the database, making it easy to share data and design.
- .
- This package provides a map widget for Kexi using the Marble library.
- .
- This package is part of the Calligra Suite.
-
 Package: calligra-reports-web-element
 Architecture: any
 Depends: ${misc:Depends}, ${shlibs:Depends}
@@ -303,20 +286,6 @@
  .
  This package is part of the Calligra Suite.
 
-Package: calligra-reports-map-element
-Architecture: any
-Depends: ${misc:Depends}, ${shlibs:Depends}
-Description: map element for Calligra Reports
- Calligra Suite is a set of applications written to help you to accomplish
- your work. It includes office applications such as a word processor,
- a spreadsheet, a presentation program, a database application, etc., and
- raster and vector graphics tools.
- .
- This package provides a map element for Calligra Reports, based on the
- Marble library.
- .
- This package is part of the Calligra Suite.
-
 Package: krita
 Architecture: any
 Section: graphics
@@ -573,7 +542,7 @@
 Section: debug
 Architecture: any
 Priority: extra
-Depends: calligra-libs (= ${binary:Version}), kde-runtime-dbg, ${misc:Depends}
+Depends: calligra-libs (= ${binary:Version}), ${misc:Depends}
 Description: debugging symbols for Calligra
  This package contains the debugging symbols associated with Calligra.
  They will automatically be used by gdb for debugging calligra-related
diff -Nru calligra-2.8.5+dfsg/debian/kexi-map-form-widget.install 
calligra-2.8.5+dfsg/debian/kexi-map-form-widget.install
--- calligra-2.8.5+dfsg/debian/kexi-map-form-widget.install 2014-08-12 
14:42:38.0 +0100
+++ calligra-2.8.5+dfsg/debian/kexi-map-form-widget.install 1970-01-01 
01:00:00.0 +0100
@@ -1,2 +0,0 @@
-usr/lib/kde4/kformdesigner_mapbrowser.so
-usr/share/kde4/services/kformdesigner/kformdesigner_mapbrowser.desktop

Bug#797389: calligra: FTBFS, multiple issues.

[peter green]

On 14/06/16 01:21, peter green wrote:
I've just done a new upload of calligra in Raspbian that on top of 
previous fixes removes the dependency on which no longer exists. 
Debdiff at
Apologies for not actually linking the debdiff, it's at 
http://debdiffs.raspbian.org/main/c/calligra/calligra_2.8.5%2bdfsg-1.2%2brpi3.debdiff

Bug#797389: calligra: FTBFS, multiple issues.

[peter green]

Martin Michlmayr wrote (as bug 818395)

calligra fails to build.  I don't see any obvious errors (but I didn't look
through the 22 MB of logs) but it ends with:
   
I already mentioned this in my posts to 797389. It seems that marble no 
longer supports qt4. The fix I applied for raspbian was to disable the 
map functionality in calligra, not great but IMO better than losing the 
package completely.


I wrote (in bug 797389)

In addition to this debdiff during the most recent binnmu of calligra in
raspbian stretch I found I needed to fix pstoedit. See
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=813316  for details.

pstoedit has now been fixed.

I've just done a new upload of calligra in Raspbian that on top of 
previous fixes removes the dependency on which no longer exists. Debdiff at


I am reluctant to NMU this In Debian myself because I don't use calligra 
personally and I have not had any feedback from anyone on the patched 
package.

Bug#826599: kdepim-runtime depends and build-depends on packages that are no longer built.

[peter green]

package: kdepim-runtime
severity: serious
version: 2.2.0-2

kdepim-runtime depends on libkgapi2-2 and build-depends on libkgapi-dev 
. These packages were previously built by the source package libkgapi 
but are no longer built.


This appears to be fixed in the version in experimental, maybe it's time 
to upload that to sid?

Neon enablement in qt for raspbian.

[Peter Green]

Back when Mike and I founded raspbian, there was only the pi1. Packages 
that used neon often set off our armv7 contamination checker and at the 
time we saw neon support in raspbian as at best a pointless waste of 
space and at worst a problem, so we simply disabled it wherever we ran 
into it.


However times have changed. The raspberry pi 2 and 3 have been released 
and Raspberry Pi want to continue shipping a single image for all their 
products for the forseable future.


As a result I am currently working on re-enabling neon in raspbian where 
appropriate. I have been doing the work in raspbian jessie and intend to 
forward port it to raspbian stretch later. Clearly since we still 
support the original raspberry pi and the new pi0 only packages with 
correctly functioning neon runtime detection are candidates for neon 
enablement.


There are two qt source packages in raspbian jessie where we currently 
have neon disabled. qt4-x11 and qtbase-opensource-src . I have a couple 
of questions.


1: are these packages supposed to have runtime neon detection (either in 
the packages own code or through ld.so hwcaps).
2: For each of the packages can anyone suggest a simple benchmark/test 
that will hit the neon codepaths which I can use to check that neon 
detection is doing it's job (i.e. that the package still works on the 
pi1 and there is a performance improvement over the package without neon 
on the pi2)

Bug#797389: calligra: FTBFS, missing build-dep on libsoprano-dev

[peter green]

On 29/11/15 14:42, peter green wrote:

On 29/09/15 21:05, peter green wrote:


On further investigation it appears that libmarble has moved 
completely to qt5.


In raspbian I disabled libmarble support in calligra. A debdiff can be 
found at.


http://debdiffs.raspbian.org/main/c/calligra/calligra_2.8.5%2bdfsg-1.2%2brpi2.debdiff 

In addition to this debdiff during the most recent binnmu of calligra in 
raspbian stretch I found I needed to fix pstoedit. See 
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=813316 for details.

Bug#807277: krfb build-depends on cruft package

[Peter Green]

tags 807277 +patch
thanks

In a raspbian stretch-staging environment (which has sid's 
ktp-common-internals) I was able to build this package after changing 
the build-dependency from libktpcommoninternalsprivate-dev to libktp-dev

Bug#807277: krfb build-depends on cruft package

[Peter Green]

Package: krfb
Version: 4:15.04.2-1
Severity: serious

krfb build-depends on libktpcommoninternalsprivate-dev which is no longer built 
by ktp-common-internals

This applies to both versions 4:15.04.2-1 and 4:15.08.3-1 of krfb.

This blocks the ktp-common-internals transition.

Bug#806848: libsignon-glib FTBFS on architectures other than amd64.

[peter green]

Package: libsignon-glib
Severity: serious

Your package failed to build on all architectures other than amd64. This 
is blocking the build of telepathy-accounts-signon which in turn 
libktpcommoninternals9 installable in debian sid on all architectures 
other than amd64.


The fix is trivial, just change /usr/lib/x86_64-linux-gnu 
/libsignon-glib.so.1* to /usr/lib/*/libsignon-glib.so.1*


I will be fixing this in raspbian stretch (where we are ahead of Debian 
stretch regarding ktp-common-internals) no intent to NMU in Debian.

Bug#797389: calligra: FTBFS, missing build-dep on libsoprano-dev

[peter green]

On 29/09/15 21:05, peter green wrote:


On further investigation it appears that libmarble has moved 
completely to qt5.


In raspbian I disabled libmarble support in calligra. A debdiff can be 
found at.


http://debdiffs.raspbian.org/main/c/calligra/calligra_2.8.5%2bdfsg-1.2%2brpi2.debdiff

Re: ongoing libkdegames transition.

[peter green]

On 17/10/15 10:48, Maximiliano Curia wrote:

Hi,

On 17/10/15 03:55, peter green wrote:
   

There appears to be a transition going on with the libkdegames source
package which seems to require sourceful changes in the rdeps. A
substantial number of rdeps have transitioned but a substantial number have
not and no bug reports seem to have been filed on said rdeps, nor does
there seem to be a transition tracker.
 

The new package names (kf5 versions) don't clash with the old ones and the so
files have different lib names, so a transition is not really needed, at least
not in the way that the release team tracks them. It would have been better if
we had upload the new lib with a different source name.

Agreed.


Is there any other package affected by this?
   

Yes, quite a few.
https://ftp-master.debian.org/cruft-report-daily.txt
* source package libkdegames version 4:15.08.0-1 no longer builds
  binary package(s): kde-games-core-declarative libkdegames-dev 
libkdegames6abi1 libkdegames6abi1-dbg libkdegamesprivate1abi1
  on 
amd64,arm64,armel,armhf,i386,kfreebsd-amd64,kfreebsd-i386,mips,mipsel,powerpc,ppc64el,s390x

  - suggested command:
dak rm -m "[auto-cruft] NBS (no longer built by libkdegames)" -s 
unstable -a 
amd64,arm64,armel,armhf,i386,kfreebsd-amd64,kfreebsd-i386,mips,mipsel,powerpc,ppc64el,s390x 
-p -R -b kde-games-core-declarative libkdegames-dev libkdegames6abi1 
libkdegames6abi1-dbg libkdegamesprivate1abi1

  - broken Depends:
kgoldrunner: kgoldrunner
kigo: kigo
klickety: klickety
kmahjongg: kmahjongg
knavalbattle: knavalbattle
knetwalk: knetwalk
knights: knights
kolf: kolf
konquest: konquest
kreversi: kreversi
kshisen: kshisen
ksirk: ksirk
ksnakeduel: ksnakeduel
kspaceduel: kspaceduel
ksudoku: ksudoku
ktuberling: ktuberling
kubrick: kubrick
lskat: lskat
tagua: tagua
  - broken Build-Depends:
bomber: libkdegames-dev (>= 4:4.11)
bovo: libkdegames-dev (>= 4:4.11)
granatier: libkdegames-dev (>= 4:4.11)
kajongg: libkdegames-dev (>= 4:4.11)
kapman: libkdegames-dev (>= 4:4.11)
katomic: libkdegames-dev (>= 4:4.11)
kblackbox: libkdegames-dev (>= 4:4.11)
kblocks: libkdegames-dev (>= 4:4.11)
kbounce: libkdegames-dev (>= 4:4.11)
kbreakout: libkdegames-dev (>= 4:4.11)
kdiamond: libkdegames-dev (>= 4:4.11)
kfourinline: libkdegames-dev (>= 4:4.11)
kgoldrunner: libkdegames-dev (>= 4:14.12.2)
kigo: libkdegames-dev (>= 4:14.12.0)
killbots: libkdegames-dev (>= 4:4.11)
kiriki: libkdegames-dev (>= 4:4.11)
kjumpingcube: libkdegames-dev (>= 4:4.11)
klickety: libkdegames-dev (>= 4:4.11)
klines: libkdegames-dev (>= 4:4.11)
kmahjongg: libkdegames-dev (>= 4:14.12.2)
kmines: libkdegames-dev (>= 4:4.11)
knavalbattle: libkdegames-dev (>= 4:4.11)
knetwalk: libkdegames-dev (>= 4:4.11)
knights: libkdegames-dev (> 4:4.9)
kolf: libkdegames-dev (>= 4:14.12.2)
kollision: libkdegames-dev (>= 4:4.11)
konquest: libkdegames-dev (>= 4:14.12.2)
kpat: libkdegames-dev (>= 4:4.11)
kreversi: libkdegames-dev (>= 4:14.12.2)
kshisen: libkdegames-dev (>= 4:4.11)
ksirk: libkdegames-dev (>= 4:4.11)
ksnakeduel: libkdegames-dev (>= 4.9.0~)
kspaceduel: libkdegames-dev (>= 4:4.11)
ksquares: libkdegames-dev (>= 4:4.11)
ksudoku: libkdegames-dev (>= 4:14.12.2)
ktuberling: libkdegames-dev (>= 4:4.11)
kubrick: libkdegames-dev (>= 4:4.11)
libkmahjongg: libkdegames-dev (>= 4:14.12.2)
lskat: libkdegames-dev (>= 4:14.12.2)
palapeli: libkdegames-dev (>= 4:14.12.2)
picmi: libkdegames-dev (>= 4:4.11)
tagua: libkdegames-dev (>= 4:4.10.0)

I find it curious that there are more significantly more broken 
build-depends than broken depends, maybe some of those build-depends are 
bogus.


   

As a deriviative maintainer I'd like to know what the rough plans and
timescales are here so I can consider the best way to handle this
downstream.
 

Do you have a preferred solution in mind?
   
IMO if it is likely that some packages will take significant time to 
migrate to the new kdegames kf5 then introducing a new

libkdegames-kde4 source package is probablly the way to go.

ongoing libkdegames transition.

[peter green]

There appears to be a transition going on with the libkdegames source 
package which seems to require sourceful changes in the rdeps. A 
substantial number of rdeps have transitioned but a substantial number 
have not and no bug reports seem to have been filed on said rdeps, nor 
does there seem to be a transition tracker.


The new kdegames package has migrated to testing, presumablly under 
"smooth updates" leaving out of data packages in testing.



As a deriviative maintainer I'd like to know what the rough plans and 
timescales are here so I can consider the best way to handle this 
downstream.

Bug#801512: kblackbox build-depends on libkdegames-dev (>= 4:14.12.2) which is no longer built.

[peter green]

Package: kblackbox
Version: 4:15.08.0-1
Severity: serious
Tags: patch

kblackbox build-depends on libkdegames-dev (>= 4:14.12.2) . 
libkdegames-dev is no longer build by the kdegames source package.


In raspbian stretch simply changing the build-dep resulted in a 
successful build, I assume it will in Debian too.

Bug#801542: kbreakout build-depends on libkdegames-dev (>= 4:14.12.2) which is no longer built.

[peter green]

Package: kbreakout
Version: 4:15.08.0-1
Severity: serious
Tags: patch

kbreakout build-depends on libkdegames-dev (>= 4:14.12.2) . libkdegames-dev is 
no longer build by the kdegames source package.

In raspbian stretch simply changing the build-dep to libkf5kdegames-dev 
resulted in a successful build, I assume it will in Debian too.

Bug#797389: calligra: FTBFS, missing build-dep on libsoprano-dev

[peter green]

We patched this issue in raspbian stretch. Unfortunately there now seems 
to be a new issue.


http://buildd.raspbian.org/status/fetch.php?pkg=calligra=armhf=1%3A2.8.5%2Bdfsg-1.2%2Brpi1%2Bb2=1443503934

make[1]: Entering directory '/«BUILDDIR»/calligra-2.8.5+dfsg'
dh_install --list-missing
cp: cannot stat 'debian/tmp/usr/lib/kde4/kformdesigner_mapbrowser.so': No such 
file or directory
dh_install: cp --reflink=auto -a 
debian/tmp/usr/lib/kde4/kformdesigner_mapbrowser.so 
debian/kexi-map-form-widget//usr/lib/kde4/ returned exit code 1
make[1]: *** [override_dh_install] Error 2
debian/rules:34: recipe for target 'override_dh_install' failed
make[1]: Leaving directory '/«BUILDDIR»/calligra-2.8.5+dfsg'
make: *** [binary-arch] Error 2
debian/rules:12: recipe for target 'binary-arch' failed

I see the following in our +rpi1+b2 build log but not our +rpi1+b1 build log

* Marble
 KDE4 World Globe Widget library
 Required by RDF, Kexi Forms and Reports to show locations on a map

Further investigation shows the following change in the 
build-dependencies installed


"libmarble-dev_4:4.14.2-1 libmarblewidget19_4:4.14.2-1" -->  
"libmarble-dev_4:15.08.1+dfsg-2 libmarblewidget-qt5-22_4:15.08.1+dfsg-2"

On further investigation it appears that libmarble has moved completely 
to qt5.

Bug#791865: kactivities-kf5 FTBFS on armel, symbols files issues.

[peter green]

Package: kactivities-kf5
Version: 5.11.0-1
Severity: serious

Your package fails to build on armel with a symbols file error.

dpkg-gensymbols: warning: some symbols or patterns disappeared in the symbols 
file: see diff output below
dpkg-gensymbols: warning: debian/libkf5activities5/DEBIAN/symbols doesn't match 
completely debian/libkf5activities5.symbols
--- debian/libkf5activities5.symbols (libkf5activities5_5.11.0-1_armel)
+++ dpkg-gensymbols53zPO5   2015-07-07 04:58:25.351194353 +
@@ -88,21 +88,29 @@
  _ZNK11KActivities8Consumer10metaObjectEv@Base 4.96.0
  _ZNK11KActivities8Consumer15currentActivityEv@Base 4.96.0
  (optional=templinst)_ZNKSt8functionIFvvEEclEv@Base 4.97.0
- 
(optional=templinst)_ZNSt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE2EE10_M_destroyEv@Base
 4.98.0
+ _ZNSt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE1EE10_M_destroyEv@Base 
5.11.0-1
+#MISSING: 5.11.0-1# 
(optional=templinst)_ZNSt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE2EE10_M_destroyEv@Base
 4.98.0
  _ZTIN11KActivities10ControllerE@Base 4.96.0
  _ZTIN11KActivities16ResourceInstanceE@Base 4.96.0
  _ZTIN11KActivities4InfoE@Base 4.96.0
  _ZTIN11KActivities8ConsumerE@Base 4.96.0
- _ZTISt11_Mutex_baseILN9__gnu_cxx12_Lock_policyE2EE@Base 4.98.0
- _ZTISt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE2EE@Base 4.98.0
+ _ZTIN9__gnu_cxx7__mutexE@Base 5.11.0-1
+ _ZTISt11_Mutex_baseILN9__gnu_cxx12_Lock_policyE1EE@Base 5.11.0-1
+#MISSING: 5.11.0-1# _ZTISt11_Mutex_baseILN9__gnu_cxx12_Lock_policyE2EE@Base 
4.98.0
+ _ZTISt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE1EE@Base 5.11.0-1
+#MISSING: 5.11.0-1# 
_ZTISt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE2EE@Base 4.98.0
  _ZTSN11KActivities10ControllerE@Base 4.96.0
  _ZTSN11KActivities16ResourceInstanceE@Base 4.96.0
  _ZTSN11KActivities4InfoE@Base 4.96.0
  _ZTSN11KActivities8ConsumerE@Base 4.96.0
- _ZTSSt11_Mutex_baseILN9__gnu_cxx12_Lock_policyE2EE@Base 4.98.0
- _ZTSSt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE2EE@Base 4.98.0
+ _ZTSN9__gnu_cxx7__mutexE@Base 5.11.0-1
+ _ZTSSt11_Mutex_baseILN9__gnu_cxx12_Lock_policyE1EE@Base 5.11.0-1
+#MISSING: 5.11.0-1# _ZTSSt11_Mutex_baseILN9__gnu_cxx12_Lock_policyE2EE@Base 
4.98.0
+ _ZTSSt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE1EE@Base 5.11.0-1
+#MISSING: 5.11.0-1# 
_ZTSSt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE2EE@Base 4.98.0
  _ZTVN11KActivities10ControllerE@Base 4.96.0
  _ZTVN11KActivities16ResourceInstanceE@Base 4.96.0
  _ZTVN11KActivities4InfoE@Base 4.96.0
  _ZTVN11KActivities8ConsumerE@Base 4.96.0
- _ZTVSt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE2EE@Base 4.98.0
+ _ZTVSt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE1EE@Base 5.11.0-1
+#MISSING: 5.11.0-1# 
_ZTVSt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE2EE@Base 4.98.0
dh_makeshlibs: failing due to earlier errors



I imagine this should be a fairly easy fix.

While kactivities-kf5 has never been built on armel kdeplasma-addons has 
and now build-depends (indirectly) on this package.



--
To UNSUBSCRIBE, email to debian-qt-kde-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/559dcb9f.6020...@p10link.net

qt5 symbols in raspbian jessie.

[peter green]

I meant to bring this up earlier but didn't get arround to it.

In raspbian jessie we have encountered some symbols differences from 
Debian armhf, so-far i've been marking the symbols in question as 
optional but i'm not positive if this is the right thing to do.


You can see our debdiffs at

http://debdiffs.raspbian.org/main/q/qtdeclarative-opensource-src/qtdeclarative-opensource-src_5.3.2-4%2brpi1.debdiff
http://debdiffs.raspbian.org/main/q/qtwebkit-opensource-src/qtwebkit-opensource-src_5.3.2%2bdfsg-3%2brpi1.debdiff

Are these symbols just template insanciations that can be safely marked 
as optional or is there more to it than that?



--
To UNSUBSCRIBE, email to debian-qt-kde-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/5525c2d3@manchester.ac.uk

Bug#762243: kde-workspace FTBFS on arm64 dpkg-shlibdeps issues

[peter green]

Package: kde-workspace
Version: $:4.11.11-2
Severity: important
User: debian-...@lists.debian.org
Usertags: arm64

kde-workspace seems to be failing to build on arm64 with dpkg-shlibdeps 
issues related to libkwinglesutils.so.


https://buildd.debian.org/status/fetch.php?pkg=kde-workspacearch=arm64ver=4%3A4.11.11-2stamp=143959

-- Installing: /«PKGBUILDDIR»/debian/tmp/usr/lib/libkwinglesutils.so.1.0.0
-- Installing: /«PKGBUILDDIR»/debian/tmp/usr/lib/libkwinglesutils.so.1
-- Installing: /«PKGBUILDDIR»/debian/tmp/usr/lib/libkwinglesutils.so
-- Removed runtime path from 
/«PKGBUILDDIR»/debian/tmp/usr/lib/libkwinglesutils.so.1.0.0
--snip--
dpkg-shlibdeps: warning: can't extract name and version from library name 
'libkdeinit4_kwin_rules_dialog.so'
dpkg-shlibdeps: error: couldn't find library libkwinglesutils.so.1 needed by 
debian/kde-window-manager/usr/lib/kde4/libkdeinit/libkdeinit4_kwin_gles.so (ELF 
format: 'elf64-littleaarch64'; RPATH: '')
dpkg-shlibdeps: error: couldn't find library libkwinglesutils.so.1 needed by 
debian/kde-window-manager/usr/lib/kde4/kwin4_effect_gles_builtins.so (ELF 
format: 'elf64-littleaarch64'; RPATH: '')
dpkg-shlibdeps: warning: can't extract name and version from library name 
'libkdeinit4_kwin.so'
--snip--
dpkg-shlibdeps: error: cannot continue due to the errors listed above
Note: libraries are not searched in other binary packages that do not have any 
shlibs or symbols file.
To help dpkg-shlibdeps find private libraries, you might need to use -l.
dh_shlibdeps: dpkg-shlibdeps -Tdebian/kde-window-manager.substvars 
debian/kde-window-manager/usr/lib/kde4/kcm_kwin4_effect_builtins.so 
debian/kde-window-manager/usr/lib/kde4/kcm_kwin4_genericscripted.so 
debian/kde-window-manager/usr/lib/kde4/kcm_kwin_scripts.so 
debian/kde-window-manager/usr/lib/kde4/kcm_kwincompositing.so 
debian/kde-window-manager/usr/lib/kde4/kcm_kwindecoration.so 
debian/kde-window-manager/usr/lib/kde4/kcm_kwindesktop.so 
debian/kde-window-manager/usr/lib/kde4/kcm_kwinoptions.so 
debian/kde-window-manager/usr/lib/kde4/kcm_kwinrules.so 
debian/kde-window-manager/usr/lib/kde4/kcm_kwinscreenedges.so 
debian/kde-window-manager/usr/lib/kde4/kcm_kwintabbox.so 
debian/kde-window-manager/usr/lib/kde4/kwin3_aurorae.so 
debian/kde-window-manager/usr/lib/kde4/kwin3_b2.so 
debian/kde-window-manager/usr/lib/kde4/kwin3_laptop.so 
debian/kde-window-manager/usr/lib/kde4/kwin3_oxygen.so 
debian/kde-window-manager/usr/lib/kde4/kwin4_effect_builtins.so 
debian/kde-window-manager/usr/lib/kde4/kwin4_effect_gles_builtins.so 
debian/kde-window-manager/usr/lib/kde4/kwin_b2_config.so 
debian/kde-window-manager/usr/lib/kde4/kwin_oxygen_config.so 
debian/kde-window-manager/usr/lib/kde4/libkdeinit/libkdeinit4_kwin_rules_dialog.so
 debian/kde-window-manager/usr/lib/kde4/libkdeinit/libkdeinit4_kwin_gles.so 
debian/kde-window-manager/usr/lib/kde4/libkdeinit/libkdeinit4_kwin.so 
debian/kde-window-manager/usr/lib/kde4/libexec/kwin_killer_helper 
debian/kde-window-manager/usr/lib/kde4/libexec/kwin_opengl_test 
debian/kde-window-manager/usr/lib/kde4/libexec/kwin_rules_dialog 
debian/kde-window-manager/usr/lib/kde4/imports/org/kde/kwin/decorations/plastik/libplastikplugin.so
 
debian/kde-window-manager/usr/lib/kde4/imports/org/kde/kwin/decoration/libdecorationplugin.so
 debian/kde-window-manager/usr/lib/kconf_update_bin/kwin_update_default_rules 
debian/kde-window-manager/usr/lib/kconf_update_bin/kwin_update_settings_410 
debian/kde-window-manager/usr/lib/kconf_update_bin/kwin_update_settings_411 
debian/kde-window-manager/usr/lib/kconf_update_bin/kwin_update_settings_49 
debian/kde-window-manager/usr/lib/kconf_update_bin/kwin_update_tabbox_qml_settings
 debian/kde-window-manager/usr/lib/kconf_update_bin/kwin_update_tabbox_settings 
debian/kde-window-manager/usr/bin/kwin 
debian/kde-window-manager/usr/bin/kwin_gles returned exit code 2


It appears the file is being installed so i'm not sure why 
dpkg-shlibdeps is not finding it. I have not tried to reproduce or debug 
this issue locally.



--
To UNSUBSCRIBE, email to debian-qt-kde-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/541cc93d.30...@p10link.net

Strange link failure with qtbase-opensource-src on arm64.

[peter green]

Qtbase-opensource-src 5.3.1+dfsg-6 failed to build on arm64 (both 
debian-ports and debian official) with the following error


g++ -Wl,-z,relro -Wl,--as-needed -Wl,--no-undefined -Wl,-e,qt_core_boilerplate -Wl,-O1 -shared -Wl,-Bsymbolic-functions -Wl,--dynamic-list,/«BUILDDIR»/qtbase-opensource-src-5.3.1+dfsg/src/corelib/QtCore.dynlist -Wl,-soname,libQt5Core.so.5 -o libQt5Core.so.5.3.1 .obj/qstring_compat.o .obj/qabstractanimation.o .obj/qvariantanimation.o .obj/qpropertyanimation.o .obj/qanimationgroup.o .obj/qsequentialanimationgroup.o .obj/qparallelanimationgroup.o .obj/qpauseanimation.o .obj/qatomic_unix.o .obj/qglobal.o .obj/qglobalstatic.o .obj/qlibraryinfo.o .obj/qmalloc.o .obj/qnumeric.o .obj/qlogging.o .obj/qatomic.o .obj/qexception.o .obj/qresultstore.o .obj/qfutureinterface.o .obj/qfuturewatcher.o .obj/qmutex.o .obj/qreadwritelock.o .obj/qrunnable.o .obj/qmutexpool.o .obj/qsemaphore.o .obj/qthread.o .obj/qthreadpool.o .obj/qthreadstorage.o .obj/qthread_unix.o .obj/qwaitcondition_unix.o .obj/qarraydata.o .obj/qbitarray.o .obj/qbytearray.o .obj/qbytearraymatcher.o .obj/qcollator.o .obj/qcommandlineoption.o .obj/qcommandlineparser.o .obj/qcryptographichash.o .obj/qdatetime.o .obj/qdatetimeparser.o .obj/qeasingcurve.o .obj/qelapsedtimer.o .obj/qfreelist.o .obj/qhash.o .obj/qline.o .obj/qlinkedlist.o .obj/qlist.o .obj/qlocale.o .obj/qlocale_tools.o .obj/qpoint.o .obj/qmap.o .obj/qmargins.o .obj/qmessageauthenticationcode.o .obj/qcontiguouscache.o .obj/qrect.o .obj/qregexp.o .obj/qregularexpression.o .obj/qrefcount.o .obj/qshareddata.o .obj/qsharedpointer.o .obj/qsimd.o .obj/qsize.o .obj/qstring.o .obj/qstringbuilder.o .obj/qstringlist.o .obj/qtextboundaryfinder.o .obj/qtimeline.o .obj/qtimezone.o .obj/qtimezoneprivate.o .obj/qunicodetools.o .obj/qvector.o .obj/qvsnprintf.o .obj/qelapsedtimer_unix.o .obj/qlocale_unix.o .obj/qtimezoneprivate_tz.o .obj/qlocale_icu.o .obj/qcollator_icu.o .obj/qtimezoneprivate_icu.o .obj/pcre16_byte_order.o .obj/pcre16_chartables.o .obj/pcre16_compile.o .obj/pcre16_config.o .obj/pcre16_dfa_exec.o .obj/pcre16_exec.o .obj/pcre16_fullinfo.o .obj/pcre16_get.o .obj/pcre16_globals.o .obj/pcre16_jit_compile.o .obj/pcre16_maketables.o .obj/pcre16_newline.o .obj/pcre16_ord2utf16.o .obj/pcre16_refcount.o .obj/pcre16_string_utils.o .obj/pcre16_study.o .obj/pcre16_tables.o .obj/pcre16_ucd.o .obj/pcre16_utf16_utils.o .obj/pcre16_valid_utf16.o .obj/pcre16_version.o .obj/pcre16_xclass.o .obj/harfbuzz-buffer.o .obj/harfbuzz-gdef.o .obj/harfbuzz-gsub.o .obj/harfbuzz-gpos.o .obj/harfbuzz-impl.o .obj/harfbuzz-open.o .obj/harfbuzz-stream.o .obj/harfbuzz-shaper-all.o .obj/qharfbuzz.o .obj/qabstractfileengine.o .obj/qbuffer.o .obj/qdatastream.o .obj/qdataurl.o .obj/qtldurl.o .obj/qdebug.o .obj/qdir.o .obj/qdiriterator.o .obj/qfile.o .obj/qfiledevice.o .obj/qfileinfo.o .obj/qipaddress.o .obj/qiodevice.o .obj/qlockfile.o .obj/qnoncontiguousbytedevice.o .obj/qprocess.o .obj/qtextstream.o .obj/qtemporarydir.o .obj/qtemporaryfile.o .obj/qresource.o .obj/qresource_iterator.o .obj/qsavefile.o .obj/qstandardpaths.o .obj/qurl.o .obj/qurlidna.o .obj/qurlquery.o .obj/qurlrecode.o .obj/qsettings.o .obj/qfsfileengine.o .obj/qfsfileengine_iterator.o .obj/qfilesystemwatcher.o .obj/qfilesystemwatcher_polling.o .obj/qfilesystementry.o .obj/qfilesystemengine.o .obj/qfileselector.o .obj/qloggingcategory.o .obj/qloggingregistry.o .obj/qfsfileengine_unix.o .obj/qfilesystemengine_unix.o .obj/qlockfile_unix.o .obj/qprocess_unix.o .obj/qfilesystemiterator_unix.o .obj/qstandardpaths_unix.o .obj/qfilesystemwatcher_inotify.o .obj/qabstractitemmodel.o .obj/qabstractproxymodel.o .obj/qitemselectionmodel.o .obj/qidentityproxymodel.o .obj/qsortfilterproxymodel.o .obj/qstringlistmodel.o .obj/qjson.o .obj/qjsondocument.o .obj/qjsonobject.o .obj/qjsonarray.o .obj/qjsonvalue.o .obj/qjsonwriter.o .obj/qjsonparser.o .obj/qpluginloader.o .obj/qfactoryloader.o .obj/quuid.o .obj/qlibrary.o .obj/qelfparser_p.o .obj/qmachparser.o .obj/qlibrary_unix.o .obj/qabstracteventdispatcher.o .obj/qabstractnativeeventfilter.o .obj/qbasictimer.o .obj/qeventloop.o .obj/qcoreapplication.o .obj/qcoreevent.o .obj/qmetaobject.o .obj/qmetatype.o .obj/qmetaobjectbuilder.o .obj/qmimedata.o .obj/qobject.o .obj/qobjectcleanuphandler.o .obj/qsignalmapper.o .obj/qsocketnotifier.o .obj/qtimer.o .obj/qtranslator.o .obj/qvariant.o .obj/qcoreglobaldata.o .obj/qsharedmemory.o .obj/qsystemsemaphore.o .obj/qpointer.o .obj/qmath.o .obj/qsystemerror.o .obj/qcore_unix.o .obj/qcrashhandler.o .obj/qeventdispatcher_unix.o .obj/qtimerinfo_unix.o .obj/qeventdispatcher_glib.o .obj/qsharedmemory_unix.o .obj/qsystemsemaphore_unix.o .obj/qisciicodec.o .obj/qlatincodec.o .obj/qtextcodec.o .obj/qtsciicodec.o .obj/qutfcodec.o .obj/qicucodec.o .obj/qstatemachine.o .obj/qabstractstate.o .obj/qstate.o .obj/qfinalstate.o .obj/qhistorystate.o .obj/qabstracttransition.o .obj/qsignaltransition.o .obj/qeventtransition.o .obj/qmimedatabase.o .obj/qmimetype.o .obj/qmimemagicrulematcher.o 

Bug#730157: kdepimlibs FTBFS on ia64, linker segfaults.

[peter green]

Stephan Schreiber wrote:

block 730157 by 718047
thanks

This is likely bug#718047. Removing the --as-needed switch from the ld 
invoke is the tentative workaround for that.
I just tried to reproduce the issue on merulo and the package failed 
with a ld segfault but on a different file, I guess whether a given 
binary succeeds or fails to link is somewhat sensitive to the 
environment the compiler is running in.


Anyway I was able to disable --as-needed and get a successful build. 
Patch is attatched, no intent to NMU.
diff -ur kdepimlibs-4.11.3/debian/rules kdepimlibs-4.11.3.new/debian/rules
--- kdepimlibs-4.11.3/debian/rules	2013-11-06 17:42:55.0 +
+++ kdepimlibs-4.11.3.new/debian/rules	2013-11-23 23:18:16.0 +
@@ -1,5 +1,10 @@
 #!/usr/bin/make -f
 
+DEB_HOST_ARCH:=$(shell dpkg-architecture -qDEB_HOST_ARCH)
+ifeq ($(DEB_HOST_ARCH),ia64)
+  dqk_link_with_as_needed := no
+endif
+
 include /usr/share/pkg-kde-tools/qt-kde-team/2/debian-qt-kde.mk
 libpkgs_addsubst_allLibraries = kdepimlibs5-dev kdepimlibs-dbg
 libpkgs_gen_strict_local_shlibs = $(libpkgs_all_packages)

Bug#730157: kdepimlibs FTBFS on ia64, linker segfaults.

[peter green]

Package: kdepimlibs
Severity: serious
Version: 4:4.11.3-1
X-debbugs-cc: debian-i...@lists.debian.org

/usr/bin/c++  -fPIC -g -O2 -Wformat -Werror=format-security -D_FORTIFY_SOURCE=2 -D_FORTIFY_SOURCE=2  -Wnon-virtual-dtor -Wno-long-long -Wundef -Wcast-align -Wchar-subscripts -Wall -W -Wpointer-arith -Wformat-security -fno-exceptions -DQT_NO_EXCEPTIONS -fno-check-new -fno-common -Woverloaded-virtual -fno-threadsafe-statics -fvisibility=hidden -Werror=return-type -fvisibility-inlines-hidden -fexceptions -UQT_NO_EXCEPTIONS -DNDEBUG -DQT_NO_DEBUG -Wl,--enable-new-dtags -Wl,--fatal-warnings -Wl,--no-undefined -lc   -Wl,--as-needed -Wl,--no-undefined -shared -Wl,-soname,libgpgme++-pthread.so.2 -o ../lib/libgpgme++-pthread.so.2.8.0 CMakeFiles/gpgmepp-pthread.dir/gpgmepp-pthread_automoc.o CMakeFiles/gpgmepp-pthread.dir/exception.o CMakeFiles/gpgmepp-pthread.dir/context.o CMakeFiles/gpgmepp-pthread.dir/key.o CMakeFiles/gpgmepp-pthread.dir/trustitem.o CMakeFiles/gpgmepp-pthread.dir/data.o CMakeFiles/gpgmepp-pthread.dir/callbacks.o CMakeFiles/gpgmepp-pthread.dir/eventloopinteractor.o CMakeFiles/gpgmepp-pthread.dir/editinteractor.o CMakeFiles/gpgmepp-pthread.dir/assuanresult.o CMakeFiles/gpgmepp-pthread.dir/keylistresult.o CMakeFiles/gpgmepp-pthread.dir/keygenerationresult.o CMakeFiles/gpgmepp-pthread.dir/importresult.o CMakeFiles/gpgmepp-pthread.dir/decryptionresult.o CMakeFiles/gpgmepp-pthread.dir/verificationresult.o CMakeFiles/gpgmepp-pthread.dir/signingresult.o CMakeFiles/gpgmepp-pthread.dir/encryptionresult.o CMakeFiles/gpgmepp-pthread.dir/engineinfo.o CMakeFiles/gpgmepp-pthread.dir/gpgsetexpirytimeeditinteractor.o CMakeFiles/gpgmepp-pthread.dir/gpgsetownertrusteditinteractor.o CMakeFiles/gpgmepp-pthread.dir/gpgsignkeyeditinteractor.o CMakeFiles/gpgmepp-pthread.dir/gpgadduserideditinteractor.o CMakeFiles/gpgmepp-pthread.dir/defaultassuantransaction.o CMakeFiles/gpgmepp-pthread.dir/scdgetinfoassuantransaction.o CMakeFiles/gpgmepp-pthread.dir/gpgagentgetinfoassuantransaction.o CMakeFiles/gpgmepp-pthread.dir/vfsmountresult.o CMakeFiles/gpgmepp-pthread.dir/configuration.o CMakeFiles/gpgmepp-pthread.dir/context_vanilla.o -lgpgme-pthread -lassuan -lgpg-error -lpthread 
collect2: ld terminated with signal 11 [Segmentation fault]

/usr/bin/ld: BFD (GNU Binutils for Debian) 2.23.52.20130727 assertion fail 
elf64-ia64.c:2723


This is blocking the decrufting of libnepomukwidgets4 which in turn is 
blocking the migration of nepomuk-widgets to testing which in turn is 
blocking the migration of kde-baseapps to testing which in turn means 
that the build-dependencies of konsole and ark cannot be satisfied in 
testing. So even though this bug is filed against a version that is only 
in sid it needs to be resolved one way or another before release.



--
To UNSUBSCRIBE, email to debian-qt-kde-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/528ea32b.90...@p10link.net

Re: building qt4 for arm

[peter green]

Rieker Flaik wrote:

How do you do that?
Any hints?
  
One option worth considering is running qemu in user mode though 
binfmt_support. This has lower overheads than full hardware emulation 
and will be able to use multiple host CPUs for paralell builds (since 
each process runs in a seperate emulator).


To do this

1: install qemu-user-static and binfmt_support on your host box (I would 
suggest using the version of qemu-arm-static from wheezy whatever 
version of debian you are running, the package has no dependencies so 
installing it on a squeeze box shouldn't be a problem)
2: copy /usr/bin/qemu-arm-static from your host box to the same location 
in your arm rootfs

3: chroot into the arm rootfs.


--
To UNSUBSCRIBE, email to debian-qt-kde-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/511a9b71.4070...@p10link.net

Re: building qt4 for arm

[peter green]

Sander wrote:

I believe the Armbrix Zero sells for $145
It has the same cpu and memory as the Arndale (Cortex-A15 1.7GHz dual
core, 2GB 800MHz DDR3, and sata3), just a little less connectors:

http://howchip.com/shop/item.php?it_id=BRIX5250
It's also on a three month lead time and has had virtually no 
documentation released. I'm not sure if any boards are actually in the 
wild or even if anyone has ever tried booting regular linux on it yet.


I'm considering it as a potential buildd but i'm not yet happy enough 
with the state of the project to actually make a purchase.


Right now i'd suggest the odriod U2 as the least bad build box option if 
you want to go the native route. Downsides are that storage and network 
end up on USB and you have to recompile the kernel to enable swap.



--
To UNSUBSCRIBE, email to debian-qt-kde-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/511a998f.9090...@p10link.net

Bug#676576: calligra FTBFS on armel and armhf error: no matching function for call to 'qMin(double, qreal)'

[peter green]

Package: calligra
Version: 1:2.4.2-1

Severity: serious

callira FTBFS on armel and armhf with the following error

/build/buildd-calligra_2.4.2-1-armhf-0fWQX5/calligra-2.4.2/plugins/pathshapes/rectangle/RectangleShape.cpp:64:85:
 error: no matching function for call to 'qMin(double, qreal)'
/build/buildd-calligra_2.4.2-1-armhf-0fWQX5/calligra-2.4.2/plugins/pathshapes/rectangle/RectangleShape.cpp:64:85:
 note: candidate is:
/usr/include/qt4/QtCore/qglobal.h:1196:34: note: templateclass T const T qMin(const 
T, const T)
/build/buildd-calligra_2.4.2-1-armhf-0fWQX5/calligra-2.4.2/plugins/pathshapes/rectangle/RectangleShape.cpp:65:86:
 error: no matching function for call to 'qMin(double, qreal)'
/build/buildd-calligra_2.4.2-1-armhf-0fWQX5/calligra-2.4.2/plugins/pathshapes/rectangle/RectangleShape.cpp:65:86:
 note: candidate is:

The issue is that on arm architectures qreal is defined as float
and due to the way qMin is defined as a template it needs the two
parameters to be of exactly the same type.

The fix should be a trvial matter of adding some typecasts, 
normally I would test this before submitting a bug report 
but my arm box is busy doing other stuff and it's a very long build.


Untested patch

P.S. the package has never built on armel so the armel failure is
not RC but it HAS built on armhf which is now a release architecture
so the armhf failure is rc.

Description: Fix qmin calls
 Add some typecasts so that both parameters to qMin are of the same type
 To fix arm build failure.
Bug-debian: http://bugs.debian.org/??
Author: Peter Michael Green plugw...@p10link.net

---
The information above should follow the Patch Tagging Guidelines, please
checkout http://dep.debian.net/deps/dep3/ to learn about the format. Here
are templates for supplementary fields that you might want to add:

Origin: vendor|upstream|other, url of original patch
Bug: url in upstream bugtracker
Bug-Debian: http://bugs.debian.org/bugnumber
Bug-Ubuntu: https://launchpad.net/bugs/bugnumber
Forwarded: no|not-needed|url proving that it has been forwarded
Reviewed-By: name and email of someone who approved the patch
Last-Update: -MM-DD

--- calligra-2.4.2.orig/plugins/pathshapes/rectangle/RectangleShape.cpp
+++ calligra-2.4.2/plugins/pathshapes/rectangle/RectangleShape.cpp
@@ -61,8 +61,8 @@ bool RectangleShape::loadOdf(const KoXml
 QString cornerRadius = element.attributeNS(KoXmlNS::draw, corner-radius, );
 if (! cornerRadius.isEmpty()) {
 float radius = KoUnit::parseValue(cornerRadius);
-m_cornerRadiusX = qMin(radius / (0.5 * size().width()) * 100, qreal(100));
-m_cornerRadiusY = qMin(radius / (0.5 * size().height()) * 100, qreal(100));
+m_cornerRadiusX = qMin(qreal(radius / (0.5 * size().width()) * 100), qreal(100));
+m_cornerRadiusY = qMin(qreal(radius / (0.5 * size().height()) * 100), qreal(100));
 }
 }
 

Bug#656056: qtwebkit symbols file bugs

[peter green]

You tagged these bugs as pending about a week ago. We (the armhf 
porters) could really do with getting them fixed as it is leading to an 
out of date package in armhf testing and we are trying to get to release 
architecture status.


Any ETA on an upload?



--
To UNSUBSCRIBE, email to debian-qt-kde-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4f756392.4030...@p10link.net

Bug#664897: qtwebkit FTBFS on armhf, symbol mismatch.

[peter green]

Package: qtwebkit
Severity: important

https://buildd.debian.org/status/fetch.php?pkg=qtwebkitarch=armhfver=2.2.0-3stamp=1331681029

Looks like some tweaks to the symbols file are needed.



--
To UNSUBSCRIBE, email to debian-qt-kde-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4f6a2f3c.6000...@postgrad.manchester.ac.uk

Bug#517566: kdebase - FTBFS: error: redefinition of 'struct ucred'

[peter green]

I have done test builds, the package FTBFS in sid and builds fine in lenny. I 
have not tried squeeze.

A notable difference between the suceeding (non-binnmu) and the failing (+b1) 
build is

original build

checking struct ucred... yes


binnmu build

checking struct ucred... no


in configure.in we see
configure.in:AC_EGREP_HEADER([struct ucred], [sys/socket.h], have_ucred=yes)

I'm not sure why this succeeds in lenny and fails in sid but I suspect some 
form of header dependency cleanup.

I changing the filename in that line to bits/socket.h should fix that issue. 


Unfortunately I can't provide a fix for the FTBFS for two reasons
1: cdbs-edit-patch does not seem to work with your setup so i'm not sure how to 
integate the modification and regeneration of the autotools stuff.
2: when testbuilding with the autoconf change in place I ran into another 
(probablly unrelated) failure






--
To UNSUBSCRIBE, email to debian-qt-kde-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#491510: kdebindings build-depends on java-gcj-compat-dev on alpah which is not availible

[peter green]

package: kdebindings
version: 4:3.5.9-1
severity serious

The package depends on java-gcj-compat-dev on a number of platforms 
including alpha, but that package is no longer available on alpha making 
kdebindings ftbfs.


I don't know if it is safe to just remove it from the list of 
architectures for that build-dependency or if more work is needed. I do 
not have any alpha box myself.


This affects both the version in lenny and the version in sid.
As well as being rc in itself this is preventing the transision of the 
fix for 484191 to testing.






--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#484191: fix for 484191

[peter green]

tags 484191 +patch
thanks

patch is attatched, just add it to the quilt series.
Index: kdebindings-3.5.9/qtruby/rubylib/qtruby/Qt.cpp
===
--- kdebindings-3.5.9.orig/qtruby/rubylib/qtruby/Qt.cpp	2008-06-07 00:48:29.0 +
+++ kdebindings-3.5.9/qtruby/rubylib/qtruby/Qt.cpp	2008-06-07 00:50:23.0 +
@@ -950,7 +950,7 @@
 const char *
 get_VALUEtype(VALUE ruby_value)
 {
-	char * classname = rb_obj_classname(ruby_value);
+	const char * classname = rb_obj_classname(ruby_value);
 const char *r = ;
 if(ruby_value == Qnil)
 	r = u;
@@ -1395,7 +1395,7 @@
 }
 
 static QCString *
-find_cached_selector(int argc, VALUE * argv, VALUE klass, char * methodName)
+find_cached_selector(int argc, VALUE * argv, VALUE klass, const char * methodName)
 {
 // Look in the cache
 static QCString * mcid = 0;
@@ -1432,7 +1432,7 @@
 static VALUE
 method_missing(int argc, VALUE * argv, VALUE self)
 {
-	char * methodName = rb_id2name(SYM2ID(argv[0]));
+	const char * methodName = rb_id2name(SYM2ID(argv[0]));
 VALUE klass = rb_funcall(self, rb_intern(class), 0);
 
 	// Look for 'thing?' methods, and try to match isThing() or hasThing() in the Smoke runtime
@@ -1482,7 +1482,7 @@
 
 			VALUE retval = rb_funcall2(qt_internal_module, rb_intern(do_method_missing), argc+3, temp_stack);
 			if (_current_method == -1) {
-char * op = rb_id2name(SYM2ID(argv[0]));
+const char * op = rb_id2name(SYM2ID(argv[0]));
 if (	qstrcmp(op, -) == 0
 		|| qstrcmp(op, +) == 0
 		|| qstrcmp(op, /) == 0
@@ -1557,7 +1557,7 @@
 class_method_missing(int argc, VALUE * argv, VALUE klass)
 {
 	VALUE result = Qnil;
-	char * methodName = rb_id2name(SYM2ID(argv[0]));
+	const char * methodName = rb_id2name(SYM2ID(argv[0]));
 	VALUE * temp_stack = (VALUE *) calloc(argc+3, sizeof(VALUE));
 temp_stack[0] = rb_str_new2(Qt);
 temp_stack[1] = rb_str_new2(methodName);
@@ -1772,7 +1772,7 @@
 VALUE
 getmetainfo(VALUE self, int offset, int index)
 {
-char * signalname = rb_id2name(rb_frame_last_func());
+const char * signalname = rb_id2name(rb_frame_last_func());
 VALUE metaObject_value = rb_funcall(qt_internal_module, rb_intern(getMetaObject), 1, self);
 
 smokeruby_object *ometa = value_obj_info(metaObject_value);