22.-26. March 2019, Leipzig - KDE Privacy Sprint

[Sandro Knauß]

Hi,

The Privacy sprint is coming  we're happy to announce the preliminary
date as 22.-26. March 2019 in Leipzig, Germany and we're looking forward to 
seeing you there. So please give us feedback, if you want to come. You can 
also get reimbursement via https://reimbursements.kde.org

We're working hard on making this sprint amazing and look forward to all
the constructive fun we're going to have.

Best, the privacy goal sprint team 
Florian, Louise and Sandro

--

On Mittwoch, 28. November 2018 13:56:54 CET Sandro Knauß wrote:
> Hey,
> 
> We volunteered to organize a KDE Privacy Sprint. We wanted to make it happen
> in March. We think at a sprint of five days. The main goal for the sprint
> is to locate privacy issues and after locating them hopefully fix them.
> Let's take the advantage to speak in person with multiple people of
> different part in KDE to find solutions. Currently the date is not fixed,
> so we doodle for one:
> 
> https://bitpoll.mafiasi.de/poll/8BYKJA5B/
> 
> Please add yourself to the doodle fast, so that we can have a fixed date
> within the next weeks.
> 
> Explicitly we would like to invite people from following teams:
> Plasma
> Plasma mobile
> Vault
> Falkon
> Choqok
> KDE Pim
> Frameworks
> 
> Please feel also invited, if you are not in such a team an be simply
> interested in that topic!
> 
> The sprint will either be located in Leipzig(Germany) or Les Magnas (France)
> http://auxsaisons.free.fr/. At the moment both options are still open, but
> soon we need to decide. What is your preference of place?
> 
> Some issues that came into my mind:
> 
> *  currently when clicking on a link in an email the url is accessed within
> Plasma and than forward this to Firefox. If the user wants to use TorBrowser
> as default browser it is a Privacy Breach.
>  *  having an interface to torify/vpn applications
> 
> Wifi:
> 
> * detect wiki login pages - do not access internet before, e.g. you are
> using inet of a hotel
> *  have a way to allow/dissallow applications interact with inet for
> specific wifis, e.g. you are using inet of a hotel
> 
> TLS:
> 
> * detecting bad SSL standards aka only TLS 1.0...
> * more helpful SSL problem dialog (properly if you have a wifi login page)
> 
> Mails:
> 
> * Encrypting headers of mails to give less metadata to surveillance T742.
> * Make it possible to search in encrypted mails, as until now, the encrypted
> data are a binary blob for the search. The search important to find mails
> again T8447.
> * Add GnuPG TOFU (Trust On First Use) trust model support for KMail. With
> TOFU you will get more security without exchanging the gpg fingerprints. To
> make it clear, exchanging fingerprints gives better security, but is
> exhausting.
> * Make Akregator webviewer protect against ad trackers T7528.
> 
> For more information look at https://phabricator.kde.org/T8622.
> 
> Share your thoughts and see you!
> 
> louise and hefee

D18253: Move the Android API level checks to runtime

[Aleix Pol Gonzalez]

apol accepted this revision.
apol added a comment.
This revision is now accepted and ready to land.


  Maybe we could print something for adb logcat that says that notifications 
won't be working?

REPOSITORY
  R289 KNotifications

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D18253

To: vkrause, apol
Cc: apol, kde-frameworks-devel, michaelh, ngraham, bruns

D17797: Include applets/22

[Noah Davis]

This revision was automatically updated to reflect the committed changes.
Closed by commit R266:d5a2b47c2d37: Include applets/22 (authored by 
nicolasfella, committed by ndavis).

REPOSITORY
  R266 Breeze Icons

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D17797?vs=49294=49495

REVISION DETAIL
  https://phabricator.kde.org/D17797

AFFECTED FILES
  icons-dark/index.theme
  icons/index.theme

To: nicolasfella, #breeze, ndavis
Cc: ndavis, kde-frameworks-devel, trickyricky26, michaelh, ngraham, bruns

D17816: Support for xattrs on kio copy/move

[Cochise César]

cochise updated this revision to Diff 49487.
cochise marked 3 inline comments as done.
cochise added a comment.


  Use subjobs on file_copy and other small fixes
  
  Not using subjobs in KIO::copy, as it breakes the tests.

REPOSITORY
  R241 KIO

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D17816?vs=48960=49487

BRANCH
  xattr-copy-support (branched from master)

REVISION DETAIL
  https://phabricator.kde.org/D17816

AFFECTED FILES
  autotests/jobtest.cpp
  autotests/jobtest.h
  src/core/CMakeLists.txt
  src/core/ConfigureChecks.cmake
  src/core/config-kiocore.h.cmake
  src/core/copyjob.cpp
  src/core/copyxattrjob.cpp
  src/core/copyxattrjob.h
  src/core/filecopyjob.cpp

To: cochise, dfaure
Cc: cfeck, bruns, phidrho, dhaumann, funkybomber, abika, pino, davidedmundson, 
ngraham, atha.kane, spoorun, nicolasfella, kde-frameworks-devel, michaelh

D18253: Move the Android API level checks to runtime

[Volker Krause]

vkrause created this revision.
Herald added a project: Frameworks.
Herald added a subscriber: kde-frameworks-devel.
vkrause requested review of this revision.

REVISION SUMMARY
  This doesn't make notifications on API level < 23 work yet, but it allows
  us to build against lower API levels and have something that still works
  when run on higher API levels.
  
  Tested on API level 21 and 27.
  
  This should also enable building the Android support on the CI.

REPOSITORY
  R289 KNotifications

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D18253

AFFECTED FILES
  src/CMakeLists.txt
  src/android/org/kde/knotifications/KNotification.java
  src/android/org/kde/knotifications/NotifyByAndroid.java
  src/knotificationmanager.cpp
  src/notifybyandroid.cpp

To: vkrause
Cc: kde-frameworks-devel, michaelh, ngraham, bruns

D18249: [datamodel] Rework items insert/remove

[David Edmundson]

davidedmundson added inline comments.

INLINE COMMENTS

> datamodel.cpp:55-61
> +const QHash roles = sourceModel()->roleNames();
> +m_roleIds.reserve(roles.count());
> +for (auto i = roles.constBegin(); i != roles.constEnd(); ++i) {
>  m_roleIds[QString::fromUtf8(i.value())] = i.key();
>  }
>  
> +setRoleNames(roles);

This part, ship it! but I don't think it'll make a huge difference to anything

> datamodel.cpp:77
>  
> +beginResetModel();
> +

This shouldn't be needed

setSourceModel will will do a reset internally

> datamodel.cpp:379
> +
> +if (oldLength > 0) {
> +beginRemoveRows(QModelIndex(), sourceIndex, sourceIndex + oldLength 
> - 1);

I don't understand what you're changing here, can you provide a bit more detail 
on the exact problem.

> datamodel.cpp:444
>  beginRemoveRows(QModelIndex(), sourceIndex, sourceIndex + 
> count - 1);
> -}
> -m_items.remove(sourceName);
> -if (count > 0) {
> +m_items.remove(sourceName);
>  endRemoveRows();

this isn't equivalent.

I could have an entry with an empty count

Now this gets left behind

REPOSITORY
  R242 Plasma Framework (Library)

REVISION DETAIL
  https://phabricator.kde.org/D18249

To: anthonyfieroni, davidedmundson, broulik, ngraham, mart, #plasma
Cc: kde-frameworks-devel, michaelh, ngraham, bruns

D18249: [datamodel] Rework items insert/remove

[Anthony Fieroni]

anthonyfieroni created this revision.
anthonyfieroni added reviewers: davidedmundson, broulik, ngraham, mart, Plasma.
Herald added a project: Frameworks.
Herald added a subscriber: kde-frameworks-devel.
anthonyfieroni requested review of this revision.

REVISION SUMMARY
  I give a try to rework insertion / deletion of items in datamodel which is 
used in systemtray. This continues approach from D16890 
, what the problem is
  
  1. Right click systemtray -> configure
  2. Click on Entries
  3. Click Ok
  
  Even only touching model dismissing dialog made index after invalid, now 
clicking on any entry in systray result in invalid index
  
  Old approach tries to make optimization but that's not a right place since 
QML is time-to-time rewritten itself, also it sets role names incorrect to me.

TEST PLAN
  Still not complete tested

REPOSITORY
  R242 Plasma Framework (Library)

REVISION DETAIL
  https://phabricator.kde.org/D18249

AFFECTED FILES
  src/declarativeimports/core/datamodel.cpp

To: anthonyfieroni, davidedmundson, broulik, ngraham, mart, #plasma
Cc: kde-frameworks-devel, michaelh, ngraham, bruns

D18191: Rebuild AAR when Java sources change

[Volker Krause]

This revision was automatically updated to reflect the committed changes.
Closed by commit R289:400ce7bc5f93: Rebuild AAR when Java sources change 
(authored by vkrause).

REPOSITORY
  R289 KNotifications

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D18191?vs=49265=49470

REVISION DETAIL
  https://phabricator.kde.org/D18191

AFFECTED FILES
  cmake/modules/FindGradle.cmake

To: vkrause, apol
Cc: kde-frameworks-devel, michaelh, ngraham, bruns

D18192: Remove unused forward declaration

[Volker Krause]

This revision was automatically updated to reflect the committed changes.
Closed by commit R289:b2c0b2748447: Remove unused forward declaration (authored 
by vkrause).

REPOSITORY
  R289 KNotifications

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D18192?vs=49266=49471

REVISION DETAIL
  https://phabricator.kde.org/D18192

AFFECTED FILES
  src/knotification.h

To: vkrause, dhaumann
Cc: kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Fabian Vogt]

fvogt added inline comments.

INLINE COMMENTS

> fvogt wrote in suprocess.h:75
> Just use 0 instead - the enum isn't meant for that apparently.

This is still there, so not done yet. The enum SuErrors is (apparently) not 
meant to be used by callers of exec as they have different accessibility.

REPOSITORY
  R299 KDESu

REVISION DETAIL
  https://phabricator.kde.org/D10716

To: jriddell, sitter, fvogt
Cc: starbuck, fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D18248: Guard resource deletion in OuptutConfiguration::sendApplied

[David Edmundson]

This revision was automatically updated to reflect the committed changes.
Closed by commit R127:9b0a3dcfeae0: Guard resource deletion in 
OuptutConfiguration::sendApplied (authored by davidedmundson).

REPOSITORY
  R127 KWayland

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D18248?vs=49455=49465

REVISION DETAIL
  https://phabricator.kde.org/D18248

AFFECTED FILES
  src/server/outputconfiguration_interface.cpp

To: davidedmundson, #kwin, zzag
Cc: zzag, kde-frameworks-devel, michaelh, ngraham, bruns

KDE CI: Frameworks » kwayland » kf5-qt5 SUSEQt5.11 - Build # 9 - Still Unstable!

[CI System]

BUILD UNSTABLE
 Build URL
https://build.kde.org/job/Frameworks/job/kwayland/job/kf5-qt5%20SUSEQt5.11/9/
 Project:
kf5-qt5 SUSEQt5.11
 Date of build:
Mon, 14 Jan 2019 17:02:47 +
 Build duration:
18 min and counting
   BUILD ARTIFACTS
  abi-compatibility-results.yamlcompat_reports/KF5Wayland_compat_report.htmllogs/KF5Wayland/5.54.0/log.txt
   JUnit Tests
  Name: (root) Failed: 0 test(s), Passed: 1 test(s), Skipped: 0 test(s), Total: 1 test(s)Name: projectroot.autotests Failed: 4 test(s), Passed: 42 test(s), Skipped: 0 test(s), Total: 46 test(s)Failed: projectroot.autotests.client.kwayland_testRemoteAccessFailed: projectroot.autotests.client.kwayland_testTextInputFailed: projectroot.autotests.client.kwayland_testWaylandSeatFailed: projectroot.autotests.client.kwayland_testWaylandSurface
   Cobertura Report
  
   Project Coverage Summary
  
   Name
  PackagesFilesClassesLinesConditionalsCobertura Coverage Report63%
(5/8)92%
(239/260)92%
(239/260)82%
(25649/31293)51%
(10138/20063)Coverage Breakdown by Package
Name
   FilesClassesLinesConditionalsautotests.client98%
(42/43)98%
(42/43)92%
(11401/12363)46%
(5966/12849)autotests.server100%
(5/5)100%
(5/5)99%
(355/358)49%
(169/344)src.client99%
(73/74)99%
(73/74)84%
(6185/7359)64%
(1787/2791)src.compat100%
(2/2)100%
(2/2)100%
(81/81)100%
(0/0)src.server100%
(117/117)100%
(117/117)85%
(7627/8940)64%
(2216/3450)src.tools0%
(0/2)0%
(0/2)0%
(0/785)0%
(0/302)src.tools.testserver0%
(0/3)0%
(0/3)0%
(0/120)0%
(0/14)tests0%
(0/14)0%
(0/14)0%
(0/1287)0%
(0/313)

KDE CI: Frameworks » kwayland » kf5-qt5 SUSEQt5.10 - Build # 3 - Still Unstable!

[CI System]

BUILD UNSTABLE
 Build URL
https://build.kde.org/job/Frameworks/job/kwayland/job/kf5-qt5%20SUSEQt5.10/3/
 Project:
kf5-qt5 SUSEQt5.10
 Date of build:
Mon, 14 Jan 2019 17:02:47 +
 Build duration:
13 min and counting
   BUILD ARTIFACTS
  abi-compatibility-results.yamlcompat_reports/KF5Wayland_compat_report.htmllogs/KF5Wayland/5.54.0/log.txt
   JUnit Tests
  Name: (root) Failed: 0 test(s), Passed: 1 test(s), Skipped: 0 test(s), Total: 1 test(s)Name: projectroot.autotests Failed: 1 test(s), Passed: 45 test(s), Skipped: 0 test(s), Total: 46 test(s)Failed: projectroot.autotests.client.kwayland_testWaylandSeat
   Cobertura Report
  
   Project Coverage Summary
  
   Name
  PackagesFilesClassesLinesConditionalsCobertura Coverage Report63%
(5/8)92%
(240/260)92%
(240/260)85%
(26637/31314)53%
(10630/20063)Coverage Breakdown by Package
Name
   FilesClassesLinesConditionalsautotests.client100%
(43/43)100%
(43/43)99%
(12195/12365)50%
(6390/12849)autotests.server100%
(5/5)100%
(5/5)99%
(355/358)49%
(169/344)src.client99%
(73/74)99%
(73/74)85%
(6261/7368)65%
(1802/2791)src.compat100%
(2/2)100%
(2/2)100%
(81/81)100%
(0/0)src.server100%
(117/117)100%
(117/117)87%
(7745/8950)66%
(2269/3450)src.tools0%
(0/2)0%
(0/2)0%
(0/785)0%
(0/302)src.tools.testserver0%
(0/3)0%
(0/3)0%
(0/120)0%
(0/14)tests0%
(0/14)0%
(0/14)0%
(0/1287)0%
(0/313)

KDE CI: Frameworks » kwayland » kf5-qt5 FreeBSDQt5.12 - Build # 4 - Still Unstable!

[CI System]

BUILD UNSTABLE
 Build URL
https://build.kde.org/job/Frameworks/job/kwayland/job/kf5-qt5%20FreeBSDQt5.12/4/
 Project:
kf5-qt5 FreeBSDQt5.12
 Date of build:
Mon, 14 Jan 2019 17:02:47 +
 Build duration:
6 min 18 sec and counting
   JUnit Tests
  Name: projectroot.autotests Failed: 14 test(s), Passed: 28 test(s), Skipped: 0 test(s), Total: 42 test(s)Failed: projectroot.autotests.client.kwayland_testCompositorFailed: projectroot.autotests.client.kwayland_testDataDeviceFailed: projectroot.autotests.client.kwayland_testDataSourceFailed: projectroot.autotests.client.kwayland_testRegionFailed: projectroot.autotests.client.kwayland_testShmPoolFailed: projectroot.autotests.client.kwayland_testSubCompositorFailed: projectroot.autotests.client.kwayland_testSubSurfaceFailed: projectroot.autotests.client.kwayland_testWaylandConnectionThreadFailed: projectroot.autotests.client.kwayland_testWaylandRegistryFailed: projectroot.autotests.client.kwayland_testWaylandShellFailed: projectroot.autotests.client.kwayland_testWaylandSurfaceFailed: projectroot.autotests.client.kwayland_testXdgShellStableFailed: projectroot.autotests.client.kwayland_testXdgShellV6Failed: projectroot.autotests.server.kwayland_testWaylandServerDisplay

D18248: Guard resource deletion in OuptutConfiguration::sendApplied

[Vlad Zagorodniy]

zzag accepted this revision.
zzag added a comment.
This revision is now accepted and ready to land.


  OK, makes sense.

REPOSITORY
  R127 KWayland

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D18248

To: davidedmundson, #kwin, zzag
Cc: zzag, kde-frameworks-devel, michaelh, ngraham, bruns

D18248: Guard resource deletion in OuptutConfiguration::sendApplied

[David Edmundson]

davidedmundson added a comment.


  Yep

REPOSITORY
  R127 KWayland

REVISION DETAIL
  https://phabricator.kde.org/D18248

To: davidedmundson, #kwin
Cc: zzag, kde-frameworks-devel, michaelh, ngraham, bruns

D18248: Guard resource deletion in OuptutConfiguration::sendApplied

[Vlad Zagorodniy]

zzag added a comment.


  If I understand it correctly, KWin tries to notify whether the changes have 
been applied successfully after the interface was unbound, but before 
OutputConfigurationInterface object is deleted, is it correct? (sorry if that's 
a stupid question)

REPOSITORY
  R127 KWayland

REVISION DETAIL
  https://phabricator.kde.org/D18248

To: davidedmundson, #kwin
Cc: zzag, kde-frameworks-devel, michaelh, ngraham, bruns

D11235: [DrKonqi] Request change of ptrace scope from KCrash

[Oswald Buddenhagen]

ossi requested changes to this revision.
ossi added inline comments.
This revision now requires changes to proceed.

INLINE COMMENTS

> debuggerlaunchers.cpp:57
>  if ( pid > 0 ) {
> +queryPtrace(pid);
>  m_monitor->startMonitoring(pid);

according to my reading of the documentation 
(https://www.kernel.org/doc/Documentation/security/Yama.txt) and the kernel 
source, this is unnecessary in this branch, as descendants of the assigned 
ptracer have the right as well (that's why the "normal" backtrace creation 
works without your patch), and not even detaching breaks the chain.

> queryptrace.cpp:67
> +
> +if (select(sfd + 1, , NULL, NULL, ) > 0 &&
> +read(sfd, msg, 3) > 0 && strcmp(msg, "OK") == 0) {

use poll() here, too.

> queryptrace.cpp:68
> +if (select(sfd + 1, , NULL, NULL, ) > 0 &&
> +read(sfd, msg, 3) > 0 && strcmp(msg, "OK") == 0) {
> +qCInfo(DRKONQI_LOG) << "ptrace granted by debugged process";

the current version of the complementary patch just echoes back the pid, so 
this cannot possibly work.

> croick wrote in queryptrace.cpp:26
> QTemporaryDir will not work, since the location is set in KCrash.

the location is still bogus, though. see comment on the other diff.

> queryptrace.h:22
> +
> +/** In Linux try to make the process attach to the debugger */
> +void queryPtrace(qint64 pid);

no. "on linux, tell the process to allow the debugger to attach to it".

> queryptrace.h:23
> +/** In Linux try to make the process attach to the debugger */
> +void queryPtrace(qint64 pid);
> +

that's a rather misleading function name. setPtracer() would reflect the 
purpose and upstream naming.

> croick wrote in crashtest.cpp:138
> That reasoning does not seem to be true. The output can still be read when 
> started "normally".

only if kdeinit was started from the same console, which you cannot assume.

also, this change wouldn't belong into this patch anyway.

REPOSITORY
  R871 DrKonqi

REVISION DETAIL
  https://phabricator.kde.org/D11235

To: croick, #plasma_workspaces, #frameworks, ossi
Cc: ossi, lepagevalleeemmanuel, maximilianocuria, adridg, plasma-devel, 
sukalyanbanga, jraleigh, GB_2, ragreen, Pitel, ZrenBot, lesliezhai, 
ali-mohamed, jensreuterberg, abetts, sebas, apol, mart

D11236: [KCrash] Establish socket to allow change of ptrace scope

[Oswald Buddenhagen]

ossi requested changes to this revision.
ossi added a comment.
This revision now requires changes to proceed.


  yay, i finally have "some" time for this. ^^
  
  it took me a while to get a coherent picture of the problem and solution, 
because your description unnecessarily dwells on the irrelevant cases of ptrace 
restrictions, but omits the crucial fact that this is about tracers that are 
not descendant processes of drkonqi, specifically kdevelop launched via 
kdeinit. see also comment on DefaultDebuggerLauncher in the complementary 
change. please make sure to point that out in the commit messages.
  
  the socket code seemed like an unnecessary complication at first, but then i 
realized that launching via kdeinit makes it impossible to just talk to the 
child via stdout/stderr or at least pass a file descriptor of a socketpair end. 
that's something that could be actually addressed in the klauncher api, because 
it's possible to pass open fds to other processes over sockets.

INLINE COMMENTS

> kcrash.cpp:655
>  //if the process was started directly, use waitpid(), as it's a 
> child...
>  while (waitpid(-1, nullptr, 0) != pid) {}
>  } else {

you need to cover that branch as well.

> kcrash.cpp:668
> +// create socket path to transfer ptrace scope and open 
> connection
> +const QByteArray socketpath = 
> QStringLiteral("%1/kcrash_%2").arg(QDir::tempPath()).arg(pid).toLocal8Bit();
> +int sockfd = openDrKonqiSocket(socketpath);

use RuntimeLocation, as the kdeinit-related code (at start of 
setCrashHandler()) does.

> kcrash.cpp:870
> +
> +if (bind(sockfd, (struct sockaddr *)_server, 
> sizeof(drkonqi_server)) < 0) {
> +perror("Warning: bind() for communication with DrKonqi failed");

you need to unlink first, otherwise stale sockets will throw you off.

> kcrash.cpp:884
> +{
> +fd_set set;
> +FD_ZERO();

why don't you use poll()? this code is linux-specific anyway, so you can rely 
on posix functions from the previous decade.

> kcrash.cpp:904
> +clsockfd = accept(sockfd, (struct sockaddr *)_client, 
> );
> +} while (clsockfd == -1 && (errno == EINTR || errno == EAGAIN));
> +if (clsockfd < 0)

EAGAIN is a workaround for some unspecified broken non-linux systems, so you 
can drop that here.

> kcrash.cpp:915
> +if (ucred.pid != pid) {
> +perror("Warning: peer pid does not match DrKonqi pid");
> +return -1;

that's bogus use of perror()

REPOSITORY
  R285 KCrash

REVISION DETAIL
  https://phabricator.kde.org/D11236

To: croick, #frameworks, ossi
Cc: dfaure, lepagevalleeemmanuel, kde-frameworks-devel, sitter, michaelh, 
ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Jonathan Riddell]

jriddell marked an inline comment as done.

REPOSITORY
  R299 KDESu

REVISION DETAIL
  https://phabricator.kde.org/D10716

To: jriddell, sitter, fvogt
Cc: starbuck, fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Jonathan Riddell]

jriddell marked 3 inline comments as done.

REPOSITORY
  R299 KDESu

REVISION DETAIL
  https://phabricator.kde.org/D10716

To: jriddell, sitter, fvogt
Cc: starbuck, fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Fabian Vogt]

fvogt requested changes to this revision.
This revision now requires changes to proceed.

REPOSITORY
  R299 KDESu

REVISION DETAIL
  https://phabricator.kde.org/D10716

To: jriddell, sitter, fvogt
Cc: starbuck, fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Jonathan Riddell]

jriddell updated this revision to Diff 49459.
jriddell added a comment.


  - make test lgpl, link configcore instead of service

REPOSITORY
  R299 KDESu

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D10716?vs=49457=49459

BRANCH
  arcpatch-D10716

REVISION DETAIL
  https://phabricator.kde.org/D10716

AFFECTED FILES
  CMakeLists.txt
  autotests/CMakeLists.txt
  autotests/config-kdesutest.h.cmake
  autotests/kdesutest.cpp
  autotests/su
  autotests/sudo
  src/suprocess.cpp
  src/suprocess.h

To: jriddell, sitter, fvogt
Cc: starbuck, fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Jonathan Riddell]

jriddell updated this revision to Diff 49457.
jriddell added a comment.


  - style fixes whitespace

REPOSITORY
  R299 KDESu

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D10716?vs=49456=49457

BRANCH
  arcpatch-D10716

REVISION DETAIL
  https://phabricator.kde.org/D10716

AFFECTED FILES
  CMakeLists.txt
  autotests/CMakeLists.txt
  autotests/config-kdesutest.h.cmake
  autotests/kdesutest.cpp
  autotests/su
  autotests/sudo
  src/suprocess.cpp
  src/suprocess.h

To: jriddell, sitter, fvogt
Cc: starbuck, fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Jonathan Riddell]

jriddell updated this revision to Diff 49456.
jriddell added a comment.


  - make string const, follow normal code style of pointer * being with 
variable name not the type

REPOSITORY
  R299 KDESu

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D10716?vs=49454=49456

BRANCH
  arcpatch-D10716

REVISION DETAIL
  https://phabricator.kde.org/D10716

AFFECTED FILES
  CMakeLists.txt
  autotests/CMakeLists.txt
  autotests/config-kdesutest.h.cmake
  autotests/kdesutest.cpp
  autotests/su
  autotests/sudo
  src/suprocess.cpp
  src/suprocess.h

To: jriddell, sitter, fvogt
Cc: starbuck, fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Harald Sitter]

sitter accepted this revision.
sitter added a comment.


  lgtm

REPOSITORY
  R299 KDESu

REVISION DETAIL
  https://phabricator.kde.org/D10716

To: jriddell, sitter, fvogt
Cc: starbuck, fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D18248: Guard resource deletion in OuptutConfiguration::sendApplied

[David Edmundson]

davidedmundson created this revision.
davidedmundson added a reviewer: KWin.
Herald added a project: Frameworks.
Herald added a subscriber: kde-frameworks-devel.
davidedmundson requested review of this revision.

REVISION SUMMARY
  A client (kscreen-doctor especially) can disconnect whilst kwin still
  has an instance.

TEST PLAN
  Had reproducible crash on kscreen-doctor.
  Now don't.

REPOSITORY
  R127 KWayland

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D18248

AFFECTED FILES
  src/server/outputconfiguration_interface.cpp

To: davidedmundson, #kwin
Cc: kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Jonathan Riddell]

jriddell updated this revision to Diff 49454.
jriddell added a comment.


  - revert enum, make strings const, add licence header

REPOSITORY
  R299 KDESu

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D10716?vs=49442=49454

BRANCH
  arcpatch-D10716

REVISION DETAIL
  https://phabricator.kde.org/D10716

AFFECTED FILES
  CMakeLists.txt
  autotests/CMakeLists.txt
  autotests/config-kdesutest.h.cmake
  autotests/kdesutest.cpp
  autotests/su
  autotests/sudo
  src/suprocess.cpp
  src/suprocess.h

To: jriddell, sitter, fvogt
Cc: starbuck, fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D17693: DocumentPrivate: Treat some chars also as "auto bracket" only when we have a selection

[loh tar]

loh.tar updated this revision to Diff 49452.
loh.tar retitled this revision from "DocumentPrivate: Treat angle bracket < and 
backtick ` also as "auto bracket" when we have a selection" to 
"DocumentPrivate: Treat some chars also as "auto bracket" only when we have a 
selection".
loh.tar edited the summary of this revision.
loh.tar added a comment.


  - Add chars ´ ` _ . : | # @ ~ * ! ? $ % / \ = to special handling
  - Avoid chars[0]

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D17693?vs=47857=49452

REVISION DETAIL
  https://phabricator.kde.org/D17693

AFFECTED FILES
  src/document/katedocument.cpp

To: loh.tar, #ktexteditor
Cc: cullmann, sars, kwrite-devel, kde-frameworks-devel, #ktexteditor, hase, 
michaelh, ngraham, bruns, demsking, dhaumann

D10716: handle wrong password when using sudo which asks for another password

[Nathaniel Graham]

ngraham added a subscriber: starbuck.

REPOSITORY
  R299 KDESu

REVISION DETAIL
  https://phabricator.kde.org/D10716

To: jriddell, sitter, fvogt
Cc: starbuck, fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D18243: Fix qmake module generation for Qt 5.12.1

[Volker Krause]

This revision was automatically updated to reflect the committed changes.
Closed by commit R240:79bf23008f66: Fix qmake module generation for Qt 5.12.1 
(authored by vkrause).

REPOSITORY
  R240 Extra CMake Modules

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D18243?vs=49447=49449

REVISION DETAIL
  https://phabricator.kde.org/D18243

AFFECTED FILES
  modules/ECMGeneratePriFile.cmake

To: vkrause, apol
Cc: kde-frameworks-devel, kde-buildsystem, michaelh, ngraham, bruns

D18243: Fix qmake module generation for Qt 5.12.1

[Aleix Pol Gonzalez]

apol accepted this revision.
This revision is now accepted and ready to land.

REPOSITORY
  R240 Extra CMake Modules

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D18243

To: vkrause, apol
Cc: kde-frameworks-devel, kde-buildsystem, michaelh, ngraham, bruns

D18243: Fix qmake module generation for Qt 5.12.1

[Volker Krause]

vkrause created this revision.
Herald added projects: Frameworks, Build System.
Herald added subscribers: kde-buildsystem, kde-frameworks-devel.
vkrause requested review of this revision.

REVISION SUMMARY
  Patch by Eike Ziller from https://github.com/KDAB/GammaRay/pull/526 in
  response to Qt change https://codereview.qt-project.org/#/c/248684/.

REPOSITORY
  R240 Extra CMake Modules

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D18243

AFFECTED FILES
  modules/ECMGeneratePriFile.cmake

To: vkrause
Cc: kde-frameworks-devel, kde-buildsystem, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Jonathan Riddell]

jriddell updated this revision to Diff 49442.
jriddell marked 2 inline comments as done.
jriddell added a comment.


  - Just use 0 instead - the enum isnt meant for that apparently.

REPOSITORY
  R299 KDESu

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D10716?vs=49440=49442

BRANCH
  arcpatch-D10716

REVISION DETAIL
  https://phabricator.kde.org/D10716

AFFECTED FILES
  CMakeLists.txt
  autotests/CMakeLists.txt
  autotests/config-kdesutest.h.cmake
  autotests/kdesutest.cpp
  autotests/su
  autotests/sudo
  src/suprocess.cpp
  src/suprocess.h

To: jriddell, sitter, fvogt
Cc: fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Jonathan Riddell]

jriddell marked an inline comment as done.

REPOSITORY
  R299 KDESu

REVISION DETAIL
  https://phabricator.kde.org/D10716

To: jriddell, sitter, fvogt
Cc: fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D17528: Refactor SlaveInterface::calcSpeed

[Chinmoy Ranjan Pradhan]

chinmoyr added inline comments.

INLINE COMMENTS

> fvogt wrote in slaveinterface.cpp:111
> This is undefined behaviour if the vector is empty.

Initial item is set to {0,0} before calcSpeed is called.

REPOSITORY
  R241 KIO

REVISION DETAIL
  https://phabricator.kde.org/D17528

To: chinmoyr, dfaure, fvogt
Cc: fvogt, davidedmundson, broulik, bruns, kde-frameworks-devel, michaelh, 
ngraham

D17691: Add rows info to the plasma virtual desktop protocol

[Vlad Zagorodniy]

zzag added inline comments.

INLINE COMMENTS

> plasmavirtualdesktop_interface.h:55
> + * Sets how many rows the virtual desktops should be laid into
> + * @since 5.53
> + */

Wrong version.

> davidedmundson wrote in plasmavirtualdesktop_interface.h:51
> Is this even implemented?? I can't find it
> 
> If so it's technically safe to remove from the header, but it's a pretty 
> rubbish situation to begin with.

Yeah, given that it was introduced quite recently and it's not implemented, 
maybe it would be better to delete the method. In either case, if you try to 
use the method, you'll probably get linker errors.

Also, maybe we don't need 
PlasmaVirtualDesktopManagementInterface::Private::{rows,columns}.

REPOSITORY
  R127 KWayland

REVISION DETAIL
  https://phabricator.kde.org/D17691

To: mart, #plasma, #kwin
Cc: zzag, davidedmundson, hein, kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Fabian Vogt]

fvogt added inline comments.

INLINE COMMENTS

> jriddell wrote in suprocess.h:75
> yes
> autotests/kdesutest.cpp:40:45: error: ‘KDESu::SuProcess::SuErrors ok’ is 
> private within this context

Just use 0 instead - the enum isn't meant for that apparently.

REPOSITORY
  R299 KDESu

REVISION DETAIL
  https://phabricator.kde.org/D10716

To: jriddell, sitter, fvogt
Cc: fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Jonathan Riddell]

jriddell updated this revision to Diff 49440.
jriddell added a comment.


  - only build tests if BUILD_TESTING is set
  - use QStringLiteral

REPOSITORY
  R299 KDESu

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D10716?vs=49244=49440

BRANCH
  arcpatch-D10716

REVISION DETAIL
  https://phabricator.kde.org/D10716

AFFECTED FILES
  CMakeLists.txt
  autotests/CMakeLists.txt
  autotests/config-kdesutest.h.cmake
  autotests/kdesutest.cpp
  autotests/su
  autotests/sudo
  src/suprocess.cpp
  src/suprocess.h

To: jriddell, sitter, fvogt
Cc: fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Jonathan Riddell]

jriddell marked an inline comment as done.
jriddell added inline comments.

INLINE COMMENTS

> fvogt wrote in suprocess.h:75
> Still necessary?

yes
autotests/kdesutest.cpp:40:45: error: ‘KDESu::SuProcess::SuErrors ok’ is 
private within this context

REPOSITORY
  R299 KDESu

REVISION DETAIL
  https://phabricator.kde.org/D10716

To: jriddell, sitter, fvogt
Cc: fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D18191: Rebuild AAR when Java sources change

[Aleix Pol Gonzalez]

apol accepted this revision.
This revision is now accepted and ready to land.

REPOSITORY
  R289 KNotifications

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D18191

To: vkrause, apol
Cc: kde-frameworks-devel, michaelh, ngraham, bruns

D17528: Refactor SlaveInterface::calcSpeed

[Stefan Brüns]

bruns added inline comments.

INLINE COMMENTS

> fvogt wrote in slaveinterface.cpp:114
> What does this actually test for?

This results in a fast ramp up when the transfer has stalled for more than ~8 
seconds, like at the begin of a transfer. Likely needs a comment.

> fvogt wrote in slaveinterface_p.h:62
> Is there any reason to use a qvector instead of a std::array/c array?
> It'll only introduce more overhead.

std::array can not be resized. At first, there is only one element.

This requirement could be avoided  around by filling the the array with {0,0} 
elements, which results in identical behavior.

We would also need a method for shifting the entries. (.removeFirst(), 
append()).

REPOSITORY
  R241 KIO

REVISION DETAIL
  https://phabricator.kde.org/D17528

To: chinmoyr, dfaure, fvogt
Cc: fvogt, davidedmundson, broulik, bruns, kde-frameworks-devel, michaelh, 
ngraham

KDE CI: Frameworks » kio » kf5-qt5 FreeBSDQt5.12 - Build # 15 - Still Unstable!

[CI System]

BUILD UNSTABLE
 Build URL
https://build.kde.org/job/Frameworks/job/kio/job/kf5-qt5%20FreeBSDQt5.12/15/
 Project:
kf5-qt5 FreeBSDQt5.12
 Date of build:
Mon, 14 Jan 2019 12:39:35 +
 Build duration:
6 min 42 sec and counting
   JUnit Tests
  Name: projectroot Failed: 4 test(s), Passed: 48 test(s), Skipped: 0 test(s), Total: 52 test(s)Failed: projectroot.autotests.kiocore_kmountpointtestFailed: projectroot.autotests.kiowidgets_dropjobtestFailed: projectroot.autotests.kiowidgets_kdirlistertestFailed: projectroot.autotests.kiowidgets_kdirmodeltestName: projectroot.autotests Failed: 0 test(s), Passed: 6 test(s), Skipped: 0 test(s), Total: 6 test(s)Name: projectroot.src.ioslaves.trash Failed: 1 test(s), Passed: 0 test(s), Skipped: 0 test(s), Total: 1 test(s)Failed: projectroot.src.ioslaves.trash.tests.testtrashName: projectroot.src.kpasswdserver Failed: 0 test(s), Passed: 1 test(s), Skipped: 0 test(s), Total: 1 test(s)

D17528: Refactor SlaveInterface::calcSpeed

[Fabian Vogt]

fvogt requested changes to this revision.
fvogt added inline comments.
This revision now requires changes to proceed.

INLINE COMMENTS

> slaveinterface.cpp:102
> +// Note for future reference: A list is maintained for sizes and times.
> +// Minimum list size is 1 and maximum list size is 8. Delta is calculated
> +// using first and last item from the list.

This comment should refer to `max_count` instead of hardcoding the number.

> slaveinterface.cpp:111
>  
> +const SlaveInterfacePrivate::TransferInfo first = 
> d->transfer_details.first();
> +const SlaveInterfacePrivate::TransferInfo last = {elapsed_time, 
> (d->filesize - d->offset)};

This is undefined behaviour if the vector is empty.

> chinmoyr wrote in slaveinterface.cpp:113
> I doubt it will be zero here. `elapsed_time` stores elapsed time since the 
> timer started so it will continue to grow (?) and the difference will always 
> be > 0.

Doubt is not enough - it is absolutely necessary to special case that. A div / 
0 is an instant crash on x86_64.

> slaveinterface.cpp:114
> +KIO::filesize_t lspeed = 1000 * (last.size - first.size) / 
> (last.time - first.time);
>  if (!lspeed) {
> +d->transfer_details.clear();

What does this actually test for?

> slaveinterface_p.h:62
> +};
> +QVector transfer_details;
> +QElapsedTimer elapsed_timer;

Is there any reason to use a qvector instead of a std::array/c array?
It'll only introduce more overhead.

> dfaure wrote in slaveinterface_p.h:39
> global namespace pollution, better keep this within 
> KIO::SlaveInterfacePrivate.

I absolutely agree here.

There is no reason to have this as a global variable, especially not a "static" 
one in a header.

Please move it into the class.

REPOSITORY
  R241 KIO

REVISION DETAIL
  https://phabricator.kde.org/D17528

To: chinmoyr, dfaure, fvogt
Cc: fvogt, davidedmundson, broulik, bruns, kde-frameworks-devel, michaelh, 
ngraham

D17691: Add rows info to the plasma virtual desktop protocol

[David Edmundson]

davidedmundson added inline comments.

INLINE COMMENTS

> davidedmundson wrote in plasma-virtual-desktop.xml:56
> This needs to be in a version 2 and done at the end with appropriate bumps 
> everywhere.

This isn't done.

REPOSITORY
  R127 KWayland

REVISION DETAIL
  https://phabricator.kde.org/D17691

To: mart, #plasma, #kwin
Cc: zzag, davidedmundson, hein, kde-frameworks-devel, michaelh, ngraham, bruns

D18178: fix environment setup of klanguagename test

[Harald Sitter]

This revision was automatically updated to reflect the committed changes.
Closed by commit R265:5ba6283e4ec1: fix environment setup of klanguagename test 
(authored by sitter).

REPOSITORY
  R265 KConfigWidgets

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D18178?vs=49224=49433

REVISION DETAIL
  https://phabricator.kde.org/D18178

AFFECTED FILES
  autotests/klanguagenametest.cpp

To: sitter, aacid, rikmills
Cc: kde-frameworks-devel, michaelh, ngraham, bruns

D18178: fix environment setup of klanguagename test

[Harald Sitter]

sitter edited the summary of this revision.

REPOSITORY
  R265 KConfigWidgets

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D18178

To: sitter, aacid, rikmills
Cc: kde-frameworks-devel, michaelh, ngraham, bruns

D10716: handle wrong password when using sudo which asks for another password

[Harald Sitter]

sitter added a comment.


  BUILD_TESTING easy fix: 
https://phabricator.kde.org/source/knotifyconfig/browse/master/CMakeLists.txt$60

REPOSITORY
  R299 KDESu

REVISION DETAIL
  https://phabricator.kde.org/D10716

To: jriddell, sitter, fvogt
Cc: fvogt, kde-frameworks-devel, michaelh, ngraham, bruns

D17691: Add rows info to the plasma virtual desktop protocol

[Marco Martin]

mart updated this revision to Diff 49425.
mart added a comment.


  - verify the signal has been emitted

REPOSITORY
  R127 KWayland

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D17691?vs=47854=49425

BRANCH
  phab/virtualdesktoprows

REVISION DETAIL
  https://phabricator.kde.org/D17691

AFFECTED FILES
  autotests/client/test_plasma_virtual_desktop.cpp
  src/client/plasmavirtualdesktop.cpp
  src/client/plasmavirtualdesktop.h
  src/client/protocols/plasma-virtual-desktop.xml
  src/server/plasmavirtualdesktop_interface.cpp
  src/server/plasmavirtualdesktop_interface.h

To: mart, #plasma, #kwin
Cc: zzag, davidedmundson, hein, kde-frameworks-devel, michaelh, ngraham, bruns

D17691: Add rows info to the plasma virtual desktop protocol

[Vlad Zagorodniy]

zzag added inline comments.

INLINE COMMENTS

> mart wrote in test_plasma_virtual_desktop.cpp:280
> why? we want to compare the number of rows, no?

What if rowsChanged is never emitted? In KWin, we usually do

  QVERIFY(spy.wait());

REPOSITORY
  R127 KWayland

REVISION DETAIL
  https://phabricator.kde.org/D17691

To: mart, #plasma, #kwin
Cc: zzag, davidedmundson, hein, kde-frameworks-devel, michaelh, ngraham, bruns

D17691: Add rows info to the plasma virtual desktop protocol

[Marco Martin]

mart added inline comments.

INLINE COMMENTS

> zzag wrote in test_plasma_virtual_desktop.cpp:280
> QVERIFY

why? we want to compare the number of rows, no?

REPOSITORY
  R127 KWayland

REVISION DETAIL
  https://phabricator.kde.org/D17691

To: mart, #plasma, #kwin
Cc: zzag, davidedmundson, hein, kde-frameworks-devel, michaelh, ngraham, bruns