Search Results: "lisandro"

27 September 2017

Enrico Zini: Qt cross-architecture development in Debian

Use case: use Debian Stable as an environment to run amd64 development machines to develop Qt applications for Raspberry Pi or other smallish armhf devices. Qt Creator is used as Integrated Development Environment, and it supports cross-compiling, running the built source on the target system, and remote debugging. Debian Stable (vanilla or Raspbian) runs on both the host and the target systems, so libraries can be kept in sync, and both systems have access to a vast amount of libraries, with security support. On top of that, armhf libraries can be installed with multiarch also in the host machine, so cross-builders have access to the exact same libraries as the target system. This sounds like a dream system. But. We're not quite there yet. cross-compile attempts I tried cross compiling a few packages:
$ sudo debootstrap stretch cross
$ echo "strech_cross"   sudo tee cross/etc/debian_chroot
$ sudo systemd-nspawn -D cross
# dpkg --add-architecture armhf
# echo "deb-src http://deb.debian.org/debian stretch main" >> /etc/apt/sources.list
# apt update
# apt install --no-install-recommends build-essential crossbuild-essential-armhf
Some packages work:
# apt source bc
# cd bc-1.06.95/
# apt-get build-dep -a armhf .
# dpkg-buildpackage -aarmhf -j2 -b
 
dh_auto_configure -- --prefix=/usr --with-readline
        ./configure --build=x86_64-linux-gnu --prefix=/usr --includedir=\$ prefix /include --mandir=\$ prefix /share/man --infodir=\$ prefix /share/info --sysconfdir=/etc --localstatedir=/var --disable-silent-rules --libdir=\$ prefix /lib/arm-linux-gnueabihf --libexecdir=\$ prefix /lib/arm-linux-gnueabihf --disable-maintainer-mode --disable-dependency-tracking --host=arm-linux-gnueabihf --prefix=/usr --with-readline
 
dpkg-deb: building package 'dc-dbgsym' in '../dc-dbgsym_1.06.95-9_armhf.deb'.
dpkg-deb: building package 'bc-dbgsym' in '../bc-dbgsym_1.06.95-9_armhf.deb'.
dpkg-deb: building package 'dc' in '../dc_1.06.95-9_armhf.deb'.
dpkg-deb: building package 'bc' in '../bc_1.06.95-9_armhf.deb'.
 dpkg-genbuildinfo --build=binary
 dpkg-genchanges --build=binary >../bc_1.06.95-9_armhf.changes
dpkg-genchanges: info: binary-only upload (no source code included)
 dpkg-source --after-build bc-1.06.95
dpkg-buildpackage: info: binary-only upload (no source included)
With qmake based Qt packages, qmake is not configured for cross-building, probably because it is not currently supported:
# apt source pumpa
# cd pumpa-0.9.3/
# apt-get build-dep -a armhf .
# dpkg-buildpackage -aarmhf -j2 -b
 
        qmake -makefile -nocache "QMAKE_CFLAGS_RELEASE=-g -O2 -fdebug-prefix-map=/root/pumpa-0.9.3=.
          -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2"
          "QMAKE_CFLAGS_DEBUG=-g -O2 -fdebug-prefix-map=/root/pumpa-0.9.3=. -fstack-protector-strong
          -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2"
          "QMAKE_CXXFLAGS_RELEASE=-g -O2 -fdebug-prefix-map=/root/pumpa-0.9.3=. -fstack-protector-strong
          -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2"
          "QMAKE_CXXFLAGS_DEBUG=-g -O2 -fdebug-prefix-map=/root/pumpa-0.9.3=. -fstack-protector-strong
          -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2"
          "QMAKE_LFLAGS_RELEASE=-Wl,-z,relro -Wl,-z,now"
          "QMAKE_LFLAGS_DEBUG=-Wl,-z,relro -Wl,-z,now" QMAKE_STRIP=: PREFIX=/usr
qmake: could not exec '/usr/lib/x86_64-linux-gnu/qt5/bin/qmake': No such file or directory
 
debian/rules:19: recipe for target 'build' failed
make: *** [build] Error 2
dpkg-buildpackage: error: debian/rules build gave error exit status 2
With cmake based Qt packages it goes a little better in that it finds the cross compiler, pkg-config and some multiarch paths, but then it tries to run armhf moc, which fails:
# apt source caneda
# cd caneda-0.3.0/
# apt-get build-dep -a armhf .
# dpkg-buildpackage -aarmhf -j2 -b
 
        cmake .. -DCMAKE_INSTALL_PREFIX=/usr -DCMAKE_VERBOSE_MAKEFILE=ON -DCMAKE_BUILD_TYPE=None
          -DCMAKE_INSTALL_SYSCONFDIR=/etc -DCMAKE_INSTALL_LOCALSTATEDIR=/var -DCMAKE_SYSTEM_NAME=Linux
          -DCMAKE_SYSTEM_PROCESSOR=arm -DCMAKE_C_COMPILER=arm-linux-gnueabihf-gcc
          -DCMAKE_CXX_COMPILER=arm-linux-gnueabihf-g\+\+
          -DPKG_CONFIG_EXECUTABLE=/usr/bin/arm-linux-gnueabihf-pkg-config
          -DCMAKE_INSTALL_LIBDIR=lib/arm-linux-gnueabihf
 
CMake Error at /usr/lib/arm-linux-gnueabihf/cmake/Qt5Core/Qt5CoreConfig.cmake:27 (message):
  The imported target "Qt5::Core" references the file
     "/usr/lib/arm-linux-gnueabihf/qt5/bin/moc"
  but this file does not exist.  Possible reasons include:
  * The file was deleted, renamed, or moved to another location.
  * An install or uninstall procedure did not complete successfully.
  * The installation package was faulty and contained
     "/usr/lib/arm-linux-gnueabihf/cmake/Qt5Core/Qt5CoreConfigExtras.cmake"
  but not all the files it references.
Note: Although I improvised a chroot to be able to fool around with it, I would use pbuilder or sbuild to do the actual builds. Helmut suggests pbuilder --host-arch or sbuild --host. Doing it the non-Debian way This guide in the meantime explains how to set up a cross-compiling Qt toolchain in a rather dirty way, by recompiling Qt pointing it at pieces of the Qt deployed on the Raspberry Pi. Following that guide, replacing the CROSS_COMPILE value with /usr/bin/arm-linux-gnueabihf- gave me a working qtbase, for which it is easy to create a Kit for Qt Creator that works, and supports linking applications with Debian development packages that do not use Qt. However, at that point I need to recompile all dependencies that use Qt myself, and I quickly got stuck at that monster of QtWebEngine, whose sources embed the whole of Chromium. Having a Qt based development environment in which I need to become the maintainer for the whole Qt toolchain is not a product I can offer to a customer. Cross compiling qmake based packages on stretch is not currently supported, so at the moment I had to suggest to postpone all plans for total world domination for at least two years. Cross-building Debian In the meantime, Helmut Grohne has been putting a lot of effort into making Debian packages cross-buildable:
helmut> enrico: yes, cross building is painful. we have ~26000 source packages. of those, ~13000 build arch-dep packages. of those, ~6000 have cross-satisfiable build-depends. of those, I tried cross building ~2300. of those 1300 cross built. so we are at about 10% working. helmut> enrico: plus there are some 607 source packages affected by some 326 bugs with patches. helmut> enrico: gogo nmu them helmut> enrico: I've filed some 1000 bugs (most of them with patches) now. around 600 are fixed :)
He is doing it mostly alone, and I would like people not to be alone when they do a lot of work in Debian, so Join Helmut in the effort of making Debian cross-buildable! Build any Debian package for any device right from the comfort of your own work computer! Have a single development environment seamlessly spanning architecture boundaries, with the power of all that there is in Debian! Join Helmut in the effort of making Debian cross-buildable! Apply here, or join #debian-bootstrap on OFTC! Cross-building Qt in Debian mitya57 summarised the situation on the KDE team side:
mitya57> we have cross-building stuff on our TODO list, but it will likely require a lot of time and neither Lisandro nor I have it currently. mitya57> see https://gobby.debian.org/export/Teams/KDE/qt-cross for a summary of what needs to be done. mitya57> Any help or patches are always welcome :))
qemu-user-static Helmut also suggested to use qemu-user-static to make the host system able to run binaries compiled for the target system, so that even if a non-cross-compiling Qt build tries to run moc and friends in their target architecture version, they would transparently succeed. At that point, it would just be a matter of replacing compiler paths to point to the native cross-compiling gcc, and the build would not be slowed down by much. Fixing bug #781226 would help in making it possible to configure a multiarch version of qmake as the qmake used for cross compiling. I have not had a chance of trying to cross-build in this way yet. In the meantime... Having qtcreator able to work on an amd64 devel machine and deploy/test/debug remotely on an arm target machine, where both machine run debian stable and have libraries in sync, would be a great thing to have even though packages do not cross-build yet. Helmut summarised the situation on IRC:
svuorela and others repeat that Qt upstream is not compatible with Debian's multiarch thinking, in that Qt upstream insists on having one toolchain for each pair of architectures, whereas the Debian way tends to be to make packages generic and split stuff such that it can be mixed and matched. An example being that you need to run qmake (thus you need qmake for the build architecture), but qmake also embeds the relevant paths and you need to query it for them (so you need qmake for the host architecture) Either you run it through qemu, or you have a particular cross qmake for your build/host pair, or you fix qt upstream to stop this madness
Building qmake in Debian for each host-target pair, even just limited to released architectures, would mean building Qt 100 times, and that's not going to scale. I wonder:

15 August 2017

Lisandro Dami n Nicanor P rez Meyer: Qt 4 removal in Debian testing (Buster)/unstable

We have been announcing it: we are going to remove Qt 4 during the Buster cycle.

Or at least that's the best outcome we can expect. Removing a very highly used library is hard, as Qt4's Webkit has proved. Qt 4 is long dead upstream and we have already started to need to patch it with untested patches as in the OpenSSL 1.1 case (will be in experimental in a few hours after this post).

We will try to put as less effort as possible in keeping it alive meaning that from now on if we need to patch it to make it support a newer lib or alike we will simply remove its support if possible. Using the OpenSSL case as an example, if we need to support any version > 1.1 we will simply remove the SSL support. That means things will break.

So, if you depend on FLOSS which is still based on Qt 4 be sure to try to port it. If you depend on a proprietary vendor software which uses Qt 4 then you better start telling them it's really time to update it. Really.

We will soon start filing bugs against packages using Qt 4. I'll update this blog post later to add that info.

For the Qt/KDE team, Lisandro.

24 June 2017

Lisandro Dami n Nicanor P rez Meyer: Qt 5.7 submodules that didn't make it to Stretch but will be in testing

There are two Qt 5.7 submodules that we could not package in time for Strech but are/will be available in their 5.7 versions in testing. This are qtdeclarative-render2d-plugin and qtvirtualkeyboard.

declarative-render2d-plugin makes use of the Raster paint engine instead of OpenGL to render the contents of a scene graph, thus making it useful when Qt Quick2 applications are run in a system without OpenGL 2 enabled hardware. Using it might require tweaking Debian's /etc/X11/Xsession.d/90qt5-opengl. On Qt 5.9 and newer this plugin is merged in Qt GUI so there should be no need to perform any action on the user's behalf.

Debian's VirtualKeyboard currently has a gotcha: we are not building it with the embedded code it ships. Upstream ships 3rd party code but lacks a way to detect and use the system versions of them. See QTBUG-59594, patches are welcomed. Please note that we prefer patches sent directly upstream to the current dev revision, we will be happy to backport patches if necessary.
Yes, this means no hunspell, openwnn, pinyin, tcime nor lipi-toolkit/t9write support.


22 February 2017

Lisandro Dami n Nicanor P rez Meyer: Developing an nrf51822 based embedded device with Qt Creator and Debian

I'm currently developing an nRF51822-based embedded device. Being one the Qt/Qt Creator maintainers in Debian I would of course try to use it for the development. Turns out it works pretty good... with some caveats.

There are already two quite interesting blog posts about using Qt Creator on MAC and on Windows, so I will not repeat the basics, as they are there. Both use qbs, but I managed to use CMake.

Instead I'll add some tips on the stuff that I needed to solve in order to make this happen on current Debian Sid.



Cheers!

17 July 2016

Lisandro Dami n Nicanor P rez Meyer: KDEPIM ready to be more broadly tested

As was posted a couple of weeks ago, the latest version of KDEPIM has been uploaded to unstable.

All packages are now uploaded and built and we believe this version is ready to be more broadly tested.

If you run unstable but have refrained from installing the kdepim packages up to now, we would appreciate it if you go ahead and install them now, reporting any issues that you may find.

Given that this is a big update that includes quite a number of plugins and libraries, it's strongly recommended that you restart your KDE session after updating the packages.

Happy hacking,

The Debian Qt/KDE Team.

Note lun jul 18 08:58:53 ART 2016: Link fixed and s/KDE/KDEPIM/.

3 July 2016

Lisandro Dami n Nicanor P rez Meyer: Upcoming KDEPIM changes in unstable (KMail, Kontact, KOrganizer, etc)

For those who care about kdepim (kmail, kontact, korganizer, etc)

Currently the latest version of kdepim is available in experimental. According to our limited tests it's working way better than kdepim 4.14 (more stable, more performant, less bugs). However migrating from one to the other is not a trivial process (distribution wise, hopefully not for our users).

Among the drawbacks in the new kdepim: knode, ktimetracker and kjots were dropped from the official kdepim components. kjots is now an independent project, not tied to the kdepim release cycle. But more importantly, knode and ktimetracker are not maintained upstream any more, we are temporarily still shipping the old KDE 4 versions, but we will drop them after stretch unless they get new upstream maintainers.

To iron out the wrinkles that are surely still there, we are now planning to start a transition for kdepim, effectively blocking all kdepim related packages in unstable until the transition is complete. This will allow us to keep the current kdepim in testing unchanged until kdepim 16.04 is ready to migrate fully to testing.

If you want to test the new kdepim versions right now, please use the version in experimental, as uploading all the packages to unstable will take some time until there is a working kdepim in unstable (mixing experimental and the unstable uploads should be fine).

If you depend on having a working kdepim, please avoid installing the new kdepim packages that will be landing in unstable in the following days, until all the components are available. We expect this will take a couple of weeks, we will post another entry when the packages are ready in unstable.

A list of the binary packages involved in the transition can be found [here].

If you find issues in the new packages, please let us know either via irc in #debian-kde, the kde mailing list debian-kde@lists.debian.org, or send us a bug report (please make sure that it wasn't reported before).

We'll send an updated note when kdepim is fully uploaded to unstable.

Happy hacking,

The Debian Qt/KDE Team.

28 May 2016

Scarlett Clark: Debian: Outreachy, Debian Reproducible builds Week 1 Progress Report

It has been an exciting first week of my internship. I was able to produce a few patches,
and submitted upstream, as well as into debian packaging. I am hopeful they will get accepted,
preferably upstream so all can benefit! kapptemplate:
https://bugs.kde.org/show_bug.cgi?id=363448
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=825122 choqok:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=825322 However, after speaking with Lisandro ( choqok maintainer ) I decided a better course of action
is to try and fix the actual source of the problem, kconfig_compiler from kde4libs is generating
non utf-8 cpp and header files under certain conditions like an environment that does not have a
locale set. Of course, I have some help with this from wonderful folks in KDE, which is good
because kde4libs codebase is HUGE! So I hope to have two new bugs early next week for choqok. I checked up on my existing kdevplatform bug:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=815962 And noticed it has not received attention, so I created upstream bug:
https://bugs.kde.org/show_bug.cgi?id=363615 And updated the debian bug patch DEP 3 with upstream bug url. I have been working on kdevelop-php without success yet, looks like build-id
( I still need to find the source ) and embedded kernel which I think I found,
though I will reach out to my awesome mentor to get some help on this one. I did not knock out quite as many builds as I wanted, but I picked some hard ones
that are new to me  So in the end it was a very successful first week, because I
have learned several new things that will help me with future reproducible builds. Have a great weekend.

26 May 2016

Lisandro Dami n Nicanor P rez Meyer: Do you want Qt5's QWebEngine in Debian? Do you have library packaging skills? If so, step up!

So far the only missing submodule in Debian's Qt5 stack is QtWebEngine. None of us the current Qt maintainers have the time/will to do the necessary stuff to have it properly packaged.

So if you would like to have QtWebEngine in Debian and:

Then you are the right person for this task. Do not hesitate in pinging me on #debian-kde, irc.oftc.net.

14 February 2016

Lunar: Reproducible builds: week 42 in Stretch cycle

What happened in the reproducible builds effort between February 7th and February 13th 2016:

Toolchain fixes
  • James McCoy uploaded devscripts/2.16.1 which makes dcmd supports .buildinfo files. Original patch by josch.
  • Lisandro Dami n Nicanor P rez Meyer uploaded qt4-x11/4:4.8.7+dfsg-6 which make files created by qch reproducible by using a fixed date instead of the current time. Original patch by Dhole.
Norbert Preining rejected the patch submitted by Reiner Herrmann to make the CreationDate not appear in comments of DVI / PS files produced by TeX. He also mentioned that some timestamps can be replaced by using the -output-comment option and that the next version of pdftex will have patches inspired by reproducible build to mitigate the effects (see SOURCE_DATE_EPOCH patches) .

Packages fixed The following packages have become reproducible due to changes in their build dependencies: abntex, apt-dpkg-ref, arduino, c++-annotations, cfi, chaksem, clif, cppreference-doc, dejagnu, derivations, ecasound, fdutils, gnash, gnu-standards, gnuift, gsequencer, gss, gstreamer0.10, gstreamer1.0, harden-doc, haskell98-report, iproute2, java-policy, libbluray, libmodbus, lizardfs, mclibs, moon-buggy, nurpawiki, php-sasl, shishi, stealth, xmltex, xsom. The following packages became reproducible after getting fixed: Some uploads fixed some reproducibility issues, but not all of them: Patches submitted which have not made their way to the archive yet:
  • #813944 on cvm by Reiner Herrmann: remove gzip headers, fix permissions of some directories and the order of the md5sums.
  • #814019 on latexdiff by Reiner Herrmann: remove the current build date from documentation.
  • #814214 on rocksdb by Chris Lamb: add support for SOURCE_DATE_EPOCH.

reproducible.debian.net A new armhf build node has been added (thanks to Vagrant Cascadian) and integrated into the Jenkins setup for 4 new armhf builder jobs. (h01ger) All packages for Debian testing (Stretch) have been tested on armhf in just 42 days. It took 114 days to get the same point for unstable back when the armhf test infrastructure was much smaller. Package sets have been enabled for testing on armhf. (h01ger) Packages producing architecture-independent ( Arch:all ) binary packages together with architecture dependent packages targeted for specific architectures will now only be tested on matching architectures. (Steven Chamberlain, h01ger) As the Jenkins setup is now made of 252 different jobs, the overview has been split into 11 different smalller views. (h01ger)

Package reviews 222 reviews have been removed, 110 added and 50 updated in the previous week. 35 FTBFS reports were made by Chris Lamb, Danny Edel, and Niko Tyni.

Misc. The recordings of Ludovic Court s' talk at FOSDEM 16 about reproducible builds and GNU Guix is now available. One can also have a look at slides from Fabian Keil's talk about ElecrtroBSD and Baptiste Daroussin's talk about FreeBSD packages.

21 September 2015

Lunar: Reproducible builds: week 21 in Stretch cycle

If you see someone on the Debian ReproducibleBuilds project, buy him/her a beer. This work is awesome. What happened in the reproducible builds effort this week: Media coverage Nathan Willis covered our DebConf15 status update in Linux Weekly News. Access to non-LWN subscribers will be given on Thursday 24th. Linux Journal published a more general piece last Tuesday. Unexpected praise for reproducible builds appeared this week in the form of several iOS applications identified as including spyware. The malware was undetected by Apple screening. This actually happened because application developers had simply downloaded a trojaned version of XCode through an unofficial source. While reproducible builds can't really help users of non-free software, this is exactly the kind of attacks that we are trying to prevent in our systems. Toolchain fixes Niko Tyni wrote and uploaded a better patch for the source order problem in libmodule-build-perl. Tristan Seligmann identified how the code generated by python-cffi could be emitted in random order in some cases. Upstream has already fixed the problem. Packages fixed The following 24 packages became reproducible due to changes in their build dependencies: apache-curator, checkbox-ng, gant, gnome-clocks, hawtjni, jackrabbit, jersey1, libjsr305-java, mathjax-docs, mlpy, moap, octave-geometry, paste, pdf.js, pyinotify, pytango, python-asyncssh, python-mock, python-openid, python-repoze.who, shadow, swift, tcpwatch-httpproxy, transfig. The following packages became reproducible after getting fixed: Some uploads fixed some reproducibility issues but not all of them: Patches submitted which have not made their way to the archive yet: reproducible.debian.net Tests for Coreboot, OpenWrt, NetBSD, and FreeBSD now runs weekly (instead of monthly). diffoscope development Python 3 offers new features (namely yield from and concurrent.futures) that could help implement parallel processing. The clear separation of bytes and unicode strings is also likely to reduce encoding related issues. Mattia Rizolo thus kicked the effort of porting diffoscope to Python 3. tlsh was the only dependency missing a Python 3 module. This got quickly fixed by a new upload. The rest of the code has been moved to the point where only incompatibilities between Python 2.7 and Pyhon 3.4 had to be changed. The commit stream still require some cleanups but all tests are now passing under Python 3. Documentation update The documentation on how to assemble the weekly reports has been updated. (Lunar) The example on how to use SOURCE_DATE_EPOCH with CMake has been improved. (Ben Beockel, Daniel Kahn Gillmor) The solution for timestamps in man pages generated by Sphinx now uses SOURCE_DATE_EPOCH. (Mattia Rizzolo) Package reviews 45 reviews have been removed, 141 added and 62 updated this week. 67 new FTBFS reports have been filled by Chris Lamb, Niko Tyni, and Lisandro Dami n Nicanor P rez Meyer. New issues added this week: randomness_in_r_rdb_rds_databases, python-ply_compiled_parse_tables. Misc. The prebuilder script is now properly testing umask variations again. Santiago Villa started a discussion on debian-devel on how binNMUs would work for reproducible builds.

5 September 2015

Lisandro Damián Nicanor Pérez Meyer: Important Akonadi fix in today's Debian Jessie's update (aka 8.2)

Todays Debian Jessie's update brings a fix in Akonadi that you certainly want in your system.

There was a bug in Akonadi that made it leak files. And if you use Kmail you will certainly want to keep reading: most of us who tested it before pushing it to testing (and now to stable) removed more than 4 GiB of useless data from our homes.

The bug that makes Akonadi leak files gets solved with the latest stable update (and has been in testing for a couple of months already). But you need to purge the leaked files. It's pretty easy: with your normal user account just run:

akonadictl fsck

That's all. After a while you will get back a lot of disk space. Note that you don't need the Akonadi fix in order to run this tool and recover your space. The fix just makes sure this won't happen again.

31 July 2015

Scott Kitterman: Plasma 5 (KDE) In Testing

A few days ago, fellow Qt/KDE team member Lisandro gave an update on the situation with migration to Plasma 5 in Debian Testing (AKA Stretch). It s changed again. All of Plasma 5 is now in Testing. The upgrade probably won t be entirely smooth, which we ll work on that after the gcc5 transition is done, but it will be much better than the half KDE4 SC half Kf5/Plasma 5 situation we ve had for the last several days. The issues with starting kwin should be resolved once users upgrade to Plasma 5. To use the current kwin with KDE SC 4, you will need to add a symlink from /usr/bin/kwin to /usr/bin/kwin_x11. That will be included in the next upload after gcc5. Systemsettings and plasma-nm now work. In my initial testing, I didn t see anything major that was broken. One user reported an issue with sddm starting automatically, but it worked fine for me. During the upgrade you should get a debconf prompt asking if you want to use kdm or sddm. Pick sddm. When I tried to dist-upgrade, apt wanted to remove task-kde-desktop. I let it remove it and some other packages and then in a second step did apt-get install task-kde-desktop. That pulled it back in successfully along with adding and removing a reasonably large stack of packages. Obviously we need to make that work better before Stretch is released, but as long as you don t restart KDE in between those two steps it should be fine. Lastely, I used apt-get autoremove to clear out a lot of no longer needed KDE4 things (when it asks if you want to stop the running kdm, say no). Here are a few notes on terminology and what I understand of the future plans: What used to be called KDE is now three different things (in part because KDE is now the community of people, not the software): KDE Frameworks 5 (Kf5): This is a group of several dozen small libraries that as a group, roughly equate to what used to be kdelibs. Plasma (Workspaces) 5: This is the desktop that we ve just transitioned to. Applications: These are a mix of kdelibs and Kf5 based applications. Currently in Testing there are some of both and this will evolve over time based on upstream development. As an example, the Kf5 based version of konsole is in Unstable and should transition to Testing shortly. Finally, thanks to Maximiliano Curia (maxy on IRC) for doing virtually all of the packaging of Kf5, Plasma 5, and applications. He did the heavy lifting, the rest of us just nibbled around the edges to keep it moving towards testing.

28 July 2015

Lisandro Damián Nicanor Pérez Meyer: Plasma/KF5 : Testing situation

Dear Debian/KDE users,

We are aware that the current situation in testing is very unfortunate, with two main issues:

  1. systemsettings transitioned to testing before the corresponding KDE Control Modules. The result is that systemsettings displays an empty screen. This is tracked in the following bug https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=790703.
  2. plasmoids such as plasma-nm transitioned to testing before plasma-desktop 5. The result is that the plasmoid are no longer displayed in the system tray.

We are working on getting plasma-desktop to transition to testing as soon as possible (hopefully in 2 days time), which will resolve both those issues. We appreciate that the transition to KF5 is much rougher than we would have liked, and apologize to all those impacted.

On behalf of the Qt/KDE team,
Lisandro.

26 May 2015

Lisandro Damián Nicanor Pérez Meyer: The last planned Qt 4 release is here: Qt 4.8.7. Is your app runnning with Qt5?

Qt 4.8.7 has been released today. Quoting from the blog post (emphasis is mine):

Many users have already moved their active projects to Qt 5 and we encourage also others to do so. With a high degree of source compatibility, we have ensured that switching to Qt 5 is smooth and straightforward. It should be noted that Qt 4.8.7 provides only the basic functionality to run Qt based applications on Mac OS X 10.10, full support is in Qt 5.

Qt 4.8.7 is planned to be the last patch release of the Qt 4 series. Standard support is available until December 2015, after which extended support will be available. We recommend all active projects to migrate to Qt 5, as new operating systems and compilers with Qt 4.8 will not be supported. If you have challenges migrating to Qt 5, please contact us or some of our service partners for assistance

Have you started to port your project?

1 May 2015

Lisandro Damián Nicanor Pérez Meyer: Qt4's status and Qt4's webkit removal in Stretch

Hi everyone! As you might know Qt4 has been deprecated (in the sense "you better start to port your code") since Qt5's first release in December 19th 2012. Since that point on Qt4 received only bugfixes. Upstream is about to release the last point release, 4.8.7. This means that only severe bugs like security ones will get a chance to get solved.

Moreover upstream recommended keeping Qt4 until 2017. If we get a Debian release every 2 years that will make Jessie oldstable in 2017 and deprecated in 2018. This means we should really consider starting to port code using Qt4 to Qt5 during Stretch's developing life cycle.

It is important to note that Qt4 depends on a number of dependencies that their maintainers might want to get removed from the archive for similar reasons. In this case we will simply don't hesitate in removing their support as long as Qt4 keeps building. This normally doesn't mean API/ABI breakage but missing plugins that will diminish functionality from your apps, maybe even key ones. As an example let's take the **hypothetical** case in which the libasound2 maintainers are switching to a new libasound3 which is not API-compatible and removing libasound2 in the process. In this case we will have no choice but to remove the dependency and drop the functionality it provides. This is another of the important reasons why you should be switching to Qt5.

Qt4's webkit removalWebkit is definitely not an easy piece of code to maintain. For starters it means having a full copy of the code in the archive for both Qt4 and Qt5. Now add to that the fact that the code evolves quickly and thus having upstream support even for security bugs will be getting harder and harder. So we decided to remove Qt4's webkit from the archive. Of course we still have a lot of KDE stuff using Qt4's webkit, so it won't disappear "soon", but it will at some point.

PortingSome of us where involved in various Qt4 to Qt5 migrations [0] and we know for sure that porting stuff from Qt4 to Qt5 is much much easier and less painful than it was from Qt3 to Qt4.

We also understand that there is still a lot of software still using Qt4. In order to ease the transition time we have provided Wheezy backports for Qt5.

Don't forget to take a look at the C++ API changes page [1] whenever you start porting your application.

[0] http://pkg-kde.alioth.debian.org/packagingqtstuff.html
[1] http://doc.qt.io/qt-5/sourcebreaks.html

Temporarily shipping both Qt4 and Qt5 builds of your libraryIn case you maintain a library chances are that upstream already provides a way to build it using Qt5. Please note there is no point in shipping an application built with both flavours, please use Qt5 whenever possible. This double compilation should be left only for libraries.

You can't mix Qt4 and Qt5 in the same binary, but you may provide libraries compiled against one or the other. For example, your source package foo could provide both libqt4foo1 and libqt5foo1. You need to mangle your debian/rules and/or build system accordingly to achieve this.

A good example both for upstream code allowing both styles of compilation and debian packaging is phonon. Take a look at the CMakeLists.txt files for seeing how a source can be built against both flavours and another to debian/rules to see an example of how to handle the compilation. Just bear in mind that you
need to replace $(overridden_command) with the command itself, that variable substitution comes from internal stuff from our team and you should not be using it without a very good reason. If in doubt, feel free to ask us on IRC [2] or on the mailing list [3].

[2] irc.debian.org #debian-kde
[3] debian-kde@lists.debian.org

TimelineWe plan to start filing wishlist bugs soon. Once we get most of KDE stuff running with Qt5's webkit we will start raising the severities.

6 November 2014

Lisandro Damián Nicanor Pérez Meyer: Early announce: Qt4 removal in Jessie+1

We the Debian Qt/KDE Team want to early-announce [maintainer warning] our decision to remove Qt4 from Jessie+1. This warning is mostly targeted at upstreams.

Qt4 has been deprecated since Qt5's first release on December 19th 2012, that means almost two years ago!

So far we had bugfixes-only releases, but upstream has announced that they will end this support on august 2015. This already means we will have to do a special effort from that point on for Jessie in case RC bugs appears, so having it in Jessie+1 is simply a non-go.

Some of us where involved in various Qt4 to Qt5 migrations [0] and we know for sure that porting stuff from Qt4 to Qt5 is much much easier and less painful than it was from Qt3 to Qt4.

We also understand that there is still a lot of software still using Qt4. In order to easy the transition time we have provided Wheezy backports for Qt5.

Don't forget to take a look at the C++ API change page [1] whenever you start porting your application.

[0] http://perezmeyer.blogspot.com.ar/2014/03/porting-qt-4-apps-to-qt-5-example-with.html
[1] http://qt-project.org/doc/qt-5.0/qtdoc/sourcebreaks.html

[maintainer warning] **Remember the freeze** and do not upload packages ported to Qt5 to unstable. The best thing you can do now is to ask your upstream if the code can be compiled against Qt5 and, why not, try it yourself.

Our first priority now is to release Jessie, and this is why this is an early announce.

9 October 2014

Lisandro Damián Nicanor Pérez Meyer: Qt 5.3.2 in Wheezy-backports: just a few hours away

In more or less 24 hs a few days most of Qt 5.3.2 will be available as a Wheezy backport. That means that if you are using Debian stable you don't need to wait for Jessie: just wait a few hours, add wheezy-backports's repo to your sources.list and get it :)

The rest of Qt 5 will arrive soon.

This is the same version that will be shipped in Jessie, so whatever you develop with it will work with the next Debian stable release :)

Don't forget: you better start porting your Qt4 apps to Qt5!

Note 2014-10-10: uups, it will still take a few days, but it will be there soon :)

Note 2014-10-15: currently building!

30 September 2014

Lisandro Damián Nicanor Pérez Meyer: Qt5 in Jessie: we will release with 5.3.2

Qt 5.3.2 has entered testing a few hours ago. This will be the version of Qt we will release with Debian Jessie, and it happens to be a nice coincidence, because upstream focused in stability for the 5.3 branch.

I'll now focus in fixing as many bugs as possible and in backporting Qt5 to Wheezy.

Let me warn you: if you are an upstream for a Qt4 based project be sure to be ready to switch to Qt5. If you are a maintainer of a Qt4 based project you better start asking your upstream to be ready for it :)

29 May 2014

Enrico Zini: wheezy-for-industrial-development

Wheezy for industrial software development I'm helping with setting up a wheezy-based toolchain for industrial automation. The basic requirements are: live-build, C++11, Qt 5.3, and a frozen internal wheezy mirror. debmirror A good part of a day's work was lost because of #749734 and possibly #628779. Mirror rebuild is still ongoing, and fingers crossed. This is Italy, and you can't simply download 21Gb of debs just to see how it goes. C++11 Stable toolchains for C++11 now exist and have gained fast adoption. It makes sense, since given what is in C++11 it is unthinkable to start a new C++ project with the old standard nowadays. C++11 is supported by g++ 4.8+ or clang 3.3+. None of them is available on wheezy or wheezy-backports. Backports exist of g++ 4.8 only for Ubuntu 12.04, but they are uninstallable on wheezy due at least to a different libc6. I tried rebuilding g++4.8 on wheezy but quickly gave up. clang 3.3 has a build dependency on g++ 4.8. LOL. However, LLVM provides an APT repository with their most recent compiler, and it works, too. C++11 problem solved! Qt 5.3 Qt 5.3 is needed because of the range of platforms it can target. There is no wheezy backport that I can find. I cannot simply get it from Qt's Download page and install it, since we need it packaged, to build live ISOs with it. I'm attempting to backport the packages from experimental to wheezy. Here are its build dependencies: libxcb-1.10 (needed by qt5) Building this is reasonably straightforward. libxkbcommon 0.4.0 (needed by qt5) The version from jessie builds fine on wheezy, provided you remove --fail-missing from the dh_install invocation. libicu 52.1 (needed by harfbuzz) The jessie packages build on wheezy, provided that mentions of clang are deleted from source/configure.ac, since it fails to build with clang 3.5 (the one currently available for wheezy on llvm.org). libharfbuzz-dev Backporting this is a bloodbath: the Debian packages from jessie depend on a forest of gobject hipsterisms of doom, all unavailable on wheezy. I gave up. qt 5.3 qtbase-opensource-src-5.3.0+dfsg can be made to build with an embedded version of harfbuzz, with just this change:
diff -Naur a/debian/control a/debian/control
--- a/debian/control    2014-05-20 18:48:27.000000000 +0200
+++ b/debian/control    2014-05-29 17:45:31.037215786 +0200
@@ -28,7 +28,6 @@
                libgstreamer-plugins-base0.10-dev,
                libgstreamer0.10-dev,
                libgtk2.0-dev,
-               libharfbuzz-dev,
                libicu-dev,
                libjpeg-dev,
                libmysqlclient-dev,
diff -Naur a/debian/rules b/debian/rules
--- a/debian/rules  2014-05-18 01:56:37.000000000 +0200
+++ b/debian/rules  2014-05-29 17:45:25.738634371 +0200
@@ -108,7 +108,6 @@
                -plugin-sql-tds \
                -system-sqlite \
                -platform $(platform_arg) \
-               -system-harfbuzz \
                -system-zlib \
                -system-libpng \
                -system-libjpeg \
(thanks Lisandro Dami n Nicanor P rez Meyer for helping me there!) There are probably going to be further steps in the Qt5 toolchain. Actually, let's try prebuilt binaries The next day with a fresh mind we realised that it is preferable to reduce our tampering with the original wheezy to a minimum. Our current plan is to use wheezy's original Qt and Qt-using packages, and use Qt's prebuilt binaries in /opt for all our custom software. We run Qt's installer, tarred the result, and wrapped it in a Debian package like this:
$ cat debian/rules
#!/usr/bin/make -f
QT_VERSION = 5.3
%:
    dh $@
override_dh_auto_build:
    dh_auto_build
    sed -re 's/@QT_VERSION@/$(QT_VERSION)/g' debian-rules.inc.in > debian-rules.inc
override_dh_auto_install:
    dh_auto_install
    # Download and untar the prebuild Qt5 binaries
    install -d -o root -g root -m 0755 debian/our-qt5-sdk/opt/Qt
    curl http://localserver/Qt$(QT_VERSION).tar.xz   xz -d   tar -C debian/our-qt5-sdk/opt -xf -
    # Move the runtime part to our-qt5
    install -d -o root -g root -m 0755 debian/our-qt5/opt/Qt
    mv debian/our-qt5-sdk/opt/Qt/$(QT_VERSION) debian/our-qt5/opt/Qt/
    # Makes dpkg-shlibdeps work on packages built with Qt from /opt
    # Hack. Don't try this at home. Don't ever do this unless you
    # know what you are doing. This voids your warranty. If you
    # know what you are doing, you won't do this.
    find debian/our-qt5/opt/Qt/$(QT_VERSION)/gcc_64/lib -maxdepth 1 -type f -name "lib*.so*" \
          sed -re 's,^.+/(lib[^.]+)\.so.+$$,\1 5 our-qt5 (>= $(QT_VERSION)),' > debian/our-qt5.shlibs
$ cat debian-rules.inc.in
export PATH := /opt/Qt/@QT_VERSION@/gcc_64/bin:$(PATH)
export QMAKESPEC=/opt/Qt/@QT_VERSION@/gcc_64/mkspecs/linux-clang/
To build one of our packages using Qt5.3 and clang, we just add this to its debian/rules:
include /usr/share/our-qt5/debian-rules.inc
Wrap up We got the dependencies sorted. Hopefully the mirror will rebuild itself tonight and tomorrow we can resume working on our custom live system.

7 May 2014

Mario Lang: Planet bug: empty alt tags for hackergotchis

There is a strange bug in Planet Debian I am seeing since I joined. It is rather minor, but since it is an accessibility bug, I'd like to mention it here. I have written to the Planet Debian maintainers, and was told to figure it out myself. This is a pattern, accessibility is considered wishlist, apparently. And the affected people are supposed to fix it on their own. It is better if I don't say anything more about that attitude.
The Bug On Planet Debian, only some people have an alt tag for their hackergotchi, while all the configured entries look similar. There is no obvious difference in the configuration, but still, only some users here have a proper alt tag for their hackergotchi. Here is a list:
  • Dirk Eddelbuettel
  • Steve Kemp
  • Wouter Verhelst
  • Mehdi (noreply@blogger.com)
  • Andrew Pollock
  • DebConf Organizers
  • Francois Marier
  • The MirOS Project (tg@mirbsd.org)
  • Paul Tagliamonte
  • Lisandro Dami n Nicanor P rez Meyer (noreply@blogger.com)
  • Joey Hess
  • Chris Lamb
  • Mirco Bauer
  • Christine Spang
  • Guido G nther
These people/organisations currently displayed on Planet Debian have a proper alt tag for their hackergotchi. All the other members have none. In Lynx, it looks like the following:
hackergotchi for
And for those where it works, it looks like:
hackergotchi for Dirk Eddelbuettel
Strange, isn't it? If you have any idea why this might be happening, let me know, or even better, tell Planet Debian maintainers how to fix it. P.S.: Package planet-venus says it is a rewrite of Planet, and Planet can be found in Debian as well. I don't see it in unstable, maybe I am blind? Or has it been removed recently? If so, the package description of planet-venus is wrong.

Next.

Previous.