Tim, Darrell, All,
krusader used to build consistently, but I've now run into an iterator issue
that looks like it might be another gcc 4.7 issue. The error is:
g++ -DHAVE_CONFIG_H -I. -I../.. -I/opt/trinity/include -I/opt/tqt3/include -I.
-include tqt.h -DQT_THREAD_SUPPORT -D_REENTRANT -D_LARGEFILE64_SOURCE
-DKDE_NO_COMPAT -DQT_NO_ASCII_CAST -Wno-long-long -Wundef -ansi
-D_XOPEN_SOURCE=500 -D_BSD_SOURCE -Wcast-align -Wchar-subscripts -Wall -W
-Wpointer-arith -fno-builtin -g3 -fno-inline -march=x86-64 -mtune=generic -O2
-pipe -fstack-protector --param=ssp-buffer-size=4 -D_FORTIFY_SOURCE=2
-fpermissive -Wformat-security -Wmissing-format-attribute -Wno-non-virtual-dtor
-fno-exceptions -fno-check-new -fno-common -I/usr/include/tqt -MT expander.o
-MD -MP -MF .deps/expander.Tpo -c -o expander.o expander.cpp
In file included from /opt/trinity/include/kfileitem.h:31:0,
from ../Panel/listpanel.h:37,
from expander.cpp:20:
/opt/trinity/include/kfilemetainfo.h:1237:34: warning: type qualifiers ignored
on function return type [-Wignored-qualifiers]
In file included from
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/bits/stl_algobase.h:66:0,
from
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/algorithm:62,
from expander.cpp:13:
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/bits/stl_iterator_base_types.h:
In instantiation of 'struct std::iterator_traits<TQValueListConstIterator<KURL> >':
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/bits/stl_iterator_base_types.h:202:5:
required by substitution of 'template<class _Iter> typename
std::iterator_traits::iterator_category std::__iterator_category(const _Iter&)
[with _Iter = TQValueListConstIterator<KURL>]'
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/bits/stl_algo.h:4490:41:
required from '_IIter std::find_if(_IIter, _IIter, _Predicate) [with _IIter =
TQValueListConstIterator<KURL>; _Predicate =
std::unary_negate<std::const_mem_fun_ref_t<bool, KURL> >]'
expander.cpp:674:102: required from here
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/bits/stl_iterator_base_types.h:166:53:
error: no type named 'iterator_category' in 'class TQValueListConstIterator<KURL>'
In file included from
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/algorithm:63:0,
from expander.cpp:13:
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/bits/stl_algo.h:
In instantiation of '_IIter std::find_if(_IIter, _IIter, _Predicate) [with
_IIter = TQValueListConstIterator<KURL>; _Predicate =
std::unary_negate<std::const_mem_fun_ref_t<bool, KURL> >]':
expander.cpp:674:102: required from here
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/bits/stl_algo.h:4490:41:
error: no matching function for call to
'__iterator_category(TQValueListConstIterator<KURL>&)'
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/bits/stl_algo.h:4490:41:
note: candidate is:
In file included from
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/bits/stl_algobase.h:66:0,
from
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/algorithm:62,
from expander.cpp:13:
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/bits/stl_iterator_base_types.h:202:5:
note: template<class _Iter> typename std::iterator_traits::iterator_category
std::__iterator_category(const _Iter&)
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../include/c++/4.7.1/bits/stl_iterator_base_types.h:202:5:
note: substitution of deduced template arguments resulted in errors seen above
Is this a gcc 4.7.1 issue or a renaming issue?
--
David C. Rankin, J.D.,P.E.
Tim, Darrell,
Digikam had been fixed once before for libpng15 issues. The patch
(digikam-libpng15+gcc47_1.diff) contained:
diff -uNrb digikam.orig/digikam/kioslave/digikamthumbnail.cpp
digikam/digikam/kioslave/digikamthumbnail.cpp
--- digikam.orig/digikam/kioslave/digikamthumbnail.cpp 2012-04-19
07:02:43.000000000 -0500
+++ digikam/digikam/kioslave/digikamthumbnail.cpp 2012-04-19 16:44:10.307517462
-0500
@@ -413,16 +413,16 @@
if (color_type == PNG_COLOR_TYPE_PALETTE)
png_set_expand(png_ptr);
- if (info_ptr->color_type == PNG_COLOR_TYPE_RGB_ALPHA)
+ if (color_type == PNG_COLOR_TYPE_RGB_ALPHA)
has_alpha = 1;
- if (info_ptr->color_type == PNG_COLOR_TYPE_GRAY_ALPHA)
+ if (color_type == PNG_COLOR_TYPE_GRAY_ALPHA)
{
has_alpha = 1;
has_grey = 1;
}
- if (info_ptr->color_type == PNG_COLOR_TYPE_GRAY)
+ if (color_type == PNG_COLOR_TYPE_GRAY)
has_grey = 1;
unsigned char **lines;
The first part of the patch is in the GIT tree, however, it looks like the
second part of the patch was either lost or not applied because the code still
contains:
if (info_ptr->color_type == PNG_COLOR_TYPE_RGB_ALPHA)
has_alpha = 1;
if (info_ptr->color_type == PNG_COLOR_TYPE_GRAY_ALPHA)
{
has_alpha = 1;
has_grey = 1;
}
if (info_ptr->color_type == PNG_COLOR_TYPE_GRAY)
has_grey = 1;
unsigned char **lines;
The error I'm seeing is:
digikamthumbnail.cpp:416:17: error: invalid use of incomplete type 'png_info
{aka struct png_info_def}'
In file included from digikamthumbnail.cpp:96:0:
/usr/include/png.h:741:16: error: forward declaration of 'png_info {aka struct
png_info_def}'
digikamthumbnail.cpp:419:17: error: invalid use of incomplete type 'png_info
{aka struct png_info_def}'
In file included from digikamthumbnail.cpp:96:0:
/usr/include/png.h:741:16: error: forward declaration of 'png_info {aka struct
png_info_def}'
digikamthumbnail.cpp:425:17: error: invalid use of incomplete type 'png_info
{aka struct png_info_def}'
In file included from digikamthumbnail.cpp:96:0:
/usr/include/png.h:741:16: error: forward declaration of 'png_info {aka struct
png_info_def}'
You can simply apply the patch with the sed call:
sed -i '/info_ptr->color_type/s|info_ptr->color_type|color_type|'
${srcdir}/digikam/digikam/kioslave/digikamthumbnail.cpp
Then digikam builds against libpng15:
Finished making: tde-digikam 14.0.0_dev-1 (Tue Jun 26 19:03:33 UTC 2012)
--
David C. Rankin, J.D.,P.E.
We have our hands full for the next two months preparing for R14. I'm using Trinity from GIT as my primary desktop, but I'm looking forward to the official release, which means fewer bugs and build issues. :-)
This morning I ran across a web article about light-weight web browsers. Would be nice to see konqueror added to such lists.
I am proposing we create a konqueror wish list as a post R14 goal.
Please share your konqueror wish list and pet peeves.
I will collate the information to etherpad.
Thanks!
Darrell
Tim, Darrel, All,
Hell Yes! The nasty kwrite crash on line-wrap when built with gcc 4.7.0-x
is _fixed_ when built with gcc 4.7.1! Damn I hate gcc transitions! To think
about the time lost because somebody botched the latest and greatest update to
the underpinnings for all software just boggles the mind. Not just on TDE, but
globally, just think about the man-hours lost chasing bugs that ended up being
due to a gcc bug.... Wow! Same thing happened with the update to 4.5.2 during
the spring of 2011. Lesson - don't put out new versions of anything during the
spring when gcc is likely to throw you a curve ball...
Regardless, kwrite is _working fine_! That gets a second *Hell Yes!*
--
David C. Rankin, J.D.,P.E.
The failure:
In file included from /usr/include/ruby-1.9.1/ruby.h:32:0,
from Qt.cpp:49:
/usr/include/ruby-1.9.1/ruby/ruby.h:24:25: fatal error: ruby/config.h: No such file or directory
compilation terminated.
make[5]: *** [Qt.lo] Error 1
The oddity is no such failure on Slackware 13.1 or 13.37, 32 or 64 bit. config.h is installed in the same place on all systems.
Any ideas?
Darrell
The splash images for digikam and showFoto need updating from KDE -> TDE, at the bottom of each image:
GIT locations:
applications/digikam/digikam/data/pics/digikam-splash.png
applications/digikam/digikam/showfoto/pics/showfoto-splash.png
Possibly each respective xcf file too, but I'm thinking we delete the xcf files.
Thanks!
Darrell
Hi trinity-devel,
First off, let me apologize for my recent 8 months of inactivity.
There has been a lot of work that I needed to do, and as such, I took
a number of months off from doing any linux work to make sure I got
what other real-life stuff needed to be done. I know a lot of people
were expecting me to finish my work; sadly, 8 months have passed and I
have missed too much to keep up with it for now.
That being said, here comes the bad news. My server was brought down
due to a faulty hard drive in May, and I've only managed to get it
warrantied and replaced now. As such, my server's mirrors were brought
down and HEAnet cannot mirror from my server at the moment. In
addition, the MeetBot was brought down as well (if you idle on
freenode, you might have noticed [lindaemon] was gone).
Even though my summer internship is giving me a bit more work to do,
and we don't have a MeetBot for now, I would still like to proceed
with the June Meeting (even with how late it is). We have not had a
meeting in a long while (since April, right?) and this would be a good
progress indicator of where we are at R14 and 3.5.13.1. Communication
is key; so let's take that key and use it for success!
The June Meeting is scheduled for next Wednesday, June 27, 2012.
Please let me know if you cannot make it.
The time for the meeting is set for 11 AM, Eastern Daylight Standard
Time (US). As always, if I get enough requests, I will move the
meeting up or down the week.
Please bring up a list of topics that need to be addressed. At the
moment, that would contain the list of fixes for 3.5.13.1 and the
state of the repository for R14.
Thanks, and sorry again,
Robert Xu
Tim, Darrell,
I have run into another failure that I don't understand. This time it is an
error looking for x11 with kchmviewer. I haven't seen this error before:
checking for X... libraries , headers
checking for gethostbyname... yes
checking for connect... yes
checking for remove... yes
checking for shmat... yes
checking for IceConnectionNumber in -lICE... yes
checking for chm_open in -lchm... yes
Please remember to use GNU make, often installed as gmake.
Install prefix : /opt/trinity
KDE support : disabled.
Qt includes :
Qt libraries :
chmlib to link : system
checking if lib should be compiled... yes
checking if po should be compiled... yes
checking if src should be compiled... yes
checking that generated files are newer than configure... done
configure: error: conditional "include_x11" was never defined.
Usually this means the macro was only invoked conditionally.
==> ERROR: A failure occurred in build().
Where do I even begin finding out what caused the:
configure: error: conditional "include_x11" was never defined.
--
David C. Rankin, J.D.,P.E.
All,
The latest amarok failure I experienced was due to tunepimp (an amarok dep)
failing to be available any longer from Arch. Looking further at the
musicbrainz.org site, tunepimp is deprecated and new development should not
occur with the library due to the RDF database lookup no longer being supported.
Further information is here:
http://musicbrainz.org/doc/History:libtunepimp/Download
Attempting to build tunepimp results in failure on gcc 4.7.1:
g++ -DHAVE_CONFIG_H -I. -I. -I../.. -I../../include/tunepimp-0.5
-I../../libltdl -Wall -O2 -march=x86-64 -mtune=generic -O2 -pipe
-fstack-protector --param=ssp-buffer-size=4 -D_FORTIFY_SOURCE=2 -MT mp4.lo -MD
-MP -MF .deps/mp4.Tpo -c mp4.cpp -fPIC -DPIC -o .libs/mp4.o
mp4.cpp:53:1: warning: deprecated conversion from string constant to 'char*'
[-Wwrite-strings]
<snip>
mp4.cpp: In function 'int mp4ReadMetadata(metadata_t*, const char*, int, const
char*)':
mp4.cpp:124:41: error: 'MP4GetMetadataName' was not declared in this scope
<snip many others>
Apparently tunepimp has been replaced with one of the following:
http://musicbrainz.org/doc/Picard_Taggerhttp://musicbrainz.org/doc/Classic_Taggerhttp://musicbrainz.org/doc/CD_Lookup_Tool
I know nada about figuring out which one of the new musicbraiz lookups
would be the one that amarok will need. What say the experts on how to fix amarok?
--
David C. Rankin, J.D.,P.E.
Tim, All,
I just started a full x86_64 build. 1st since name change and I've hit a snag
with tqca-tls. My /etc/profile.d/tqt.sh is executable this time :) The build
can't find TQt:
==> Setting PATH and Trinity Environment variables
==> Patching...
==> Starting configure...
Configuring qca-tls ...
Verifying TQt 3.x Multithreaded (MT) build environment ... fail
There was an error compiling 'conf'. Be sure you have a proper
TQt 3.x Multithreaded (MT) build environment set up.
==> ERROR: A failure occurred in build().
Aborting...
HuH??
Looking at the conf.log it looks like a bad joke on 64 bit boxes concerning
what was generated:
g++ -c -pipe -Wall -W -O2 -D_REENTRANT -DX11_INC='"/usr/X11R6/include"'
-DX11_LIBDIR='"/usr/X11R6/lib64"' -DX11_LIB='"-lXext -lX11 -lm"' -DCC='"gcc"'
-DQT_NO_DEBUG -DQT_THREAD_SUPPORT -I/opt/tqt3/mkspecs/default -I.
-I'/usr/include/tqt' -I/opt/tqt3/include -I/usr/X11R6/include -o conf.o conf.cpp
g++ -Wl,-rpath,/opt/tqt3/lib64 -o conf conf.o -L/opt/tqt3/lib64
-L/usr/X11R6/lib64 -ltqt-mt -lXext -lX11 -lm -lpthread
/usr/bin/ld: cannot find -ltqt-mt
collect2: error: ld returned 1 exit status
make: *** [conf] Error 1
Where is this /opt/tqt3/lib'64' garbage coming from? Arch just 'lib' not any
'lib64' dir names when handling libraries. Where in this package are the 'lib'
or 'lib64' search directories and lib directory names controlled? Like noted
earlier, I can build tqca-tls on i686 all day long, but this is one of the first
i've built on x86_64 and something isn't quite right. Prior to the name change
stuff, this built like clockwork. Where should we look for the culprit here
(there isn't much here).
What say the gurus? Ideas?
--
David C. Rankin, J.D.,P.E.