[okular] [Bug 408849] okular crashes scrolling through certain pdf

2019-07-03 Thread Albert Astals Cid
https://bugs.kde.org/show_bug.cgi?id=408849

Albert Astals Cid  changed:

   What|Removed |Added

 CC||davidmen...@gmail.com

--- Comment #7 from Albert Astals Cid  ---
*** Bug 409462 has been marked as a duplicate of this bug. ***

-- 
You are receiving this mail because:
You are the assignee for the bug.

[okular] [Bug 408849] okular crashes scrolling through certain pdf

2019-06-22 Thread Albert Astals Cid
https://bugs.kde.org/show_bug.cgi?id=408849

Albert Astals Cid  changed:

   What|Removed |Added

  Latest Commit||https://invent.kde.org/kde/
   ||okular/commit/f3f7ca79fcf1a
   ||e3521ed9b7d8b586f3d45b8a487
 Status|CONFIRMED   |RESOLVED
 Resolution|--- |FIXED

--- Comment #5 from Albert Astals Cid  ---
Git commit f3f7ca79fcf1ae3521ed9b7d8b586f3d45b8a487 by Albert Astals Cid.
Committed on 22/06/2019 at 10:35.
Pushed by aacid into branch 'Applications/19.04'.

Fix crash for rects that don't have an action

It shouldn't really happen, but in some cases it does, so be protective

M  +6-1ui/pageview.cpp

https://invent.kde.org/kde/okular/commit/f3f7ca79fcf1ae3521ed9b7d8b586f3d45b8a487

-- 
You are receiving this mail because:
You are the assignee for the bug.

[okular] [Bug 408849] okular crashes scrolling through certain pdf

2019-06-22 Thread Albert Astals Cid
https://bugs.kde.org/show_bug.cgi?id=408849

--- Comment #4 from Albert Astals Cid  ---
So i had a brain malfunction and when implementing this in 2016 i forgot to
commit half the implementation :?

Commiting in a moment

-- 
You are receiving this mail because:
You are the assignee for the bug.

[okular] [Bug 408849] okular crashes scrolling through certain pdf

2019-06-22 Thread Albert Astals Cid
https://bugs.kde.org/show_bug.cgi?id=408849

Albert Astals Cid  changed:

   What|Removed |Added

 CC||aa...@kde.org

--- Comment #3 from Albert Astals Cid  ---
(In reply to Tobias Deiminger from comment #2)
> The problem is that Okular has bogus handling for Poppler::Link::OCGState.
> If a document contains such a link, we meet the case arm
> 
> case Poppler::Link::OCGState:
> Q_UNREACHABLE();
> 
> This aborts Okular immediately in debug builds when scrolling to the page,
> or leads to construction of an Okular::ObjectRect with m_object = 0x0 in
> release builds, causing the crash to happen later (that's what happened in
> my trial from comment 2).
> 
> We either have to handle the "not implemented yet" situation more
> gracefully. Or create a suitable Okular::Action object to expose the feature
> if Okular is already capable of handling OCGState links. Didn't check what's
> the current state here, does anybody know?

That is bad, we have support for OCGState links. Let me try to see what
happened.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[okular] [Bug 408849] okular crashes scrolling through certain pdf

2019-06-18 Thread Tobias Deiminger
https://bugs.kde.org/show_bug.cgi?id=408849

--- Comment #2 from Tobias Deiminger  ---
The problem is that Okular has bogus handling for Poppler::Link::OCGState. If a
document contains such a link, we meet the case arm

case Poppler::Link::OCGState:
Q_UNREACHABLE();

This aborts Okular immediately in debug builds when scrolling to the page, or
leads to construction of an Okular::ObjectRect with m_object = 0x0 in release
builds, causing the crash to happen later (that's what happened in my trial
from comment 2).

We either have to handle the "not implemented yet" situation more gracefully.
Or create a suitable Okular::Action object to expose the feature if Okular is
already capable of handling OCGState links. Didn't check what's the current
state here, does anybody know?

-- 
You are receiving this mail because:
You are the assignee for the bug.

[okular] [Bug 408849] okular crashes scrolling through certain pdf

2019-06-17 Thread Tobias Deiminger
https://bugs.kde.org/show_bug.cgi?id=408849

Tobias Deiminger  changed:

   What|Removed |Added

 CC||haxti...@posteo.de
 Ever confirmed|0   |1
 Status|REPORTED|CONFIRMED

--- Comment #1 from Tobias Deiminger  ---
I can't cause a crash by simply scrolling. However, a crash happens
reproducible when right-clicking the "Protrusion" link on left margin of page
4, using latest okular and poppler master, with backtrace as follows:

Thread 1 "okular" received signal SIGSEGV, Segmentation fault.
0x7fffe6a99989 in PageView::createProcessLinkMenu(PageViewItem*, QPoint
const&) () from ./okularpart.so
(gdb) bt
#0  0x7fffe6a99989 in PageView::createProcessLinkMenu(PageViewItem*, QPoint
const&) () from ./okularpart.so
#1  0x7fffe6a8eaee in PageView::mouseReleaseEvent(QMouseEvent*) () from
./okularpart.so
#2  0x757a8f98 in QWidget::event(QEvent*) () from
/home/tobias/kde/qt5/lib/libQt5Widgets.so.5
#3  0x7584aaee in QFrame::event(QEvent*) () from
/home/tobias/kde/qt5/lib/libQt5Widgets.so.5
#4  0x7fffe6a93a43 in PageView::viewportEvent(QEvent*) () from
./okularpart.so
#5  0x74cfa2dd in
QCoreApplicationPrivate::sendThroughObjectEventFilters(QObject*, QEvent*) ()
from /home/tobias/kde/qt5/lib/libQt5Core.so.5
#6  0x7576a3d5 in QApplicationPrivate::notify_helper(QObject*, QEvent*)
() from /home/tobias/kde/qt5/lib/libQt5Widgets.so.5
#7  0x75772567 in QApplication::notify(QObject*, QEvent*) () from
/home/tobias/kde/qt5/lib/libQt5Widgets.so.5
#8  0x74cfa558 in QCoreApplication::notifyInternal2(QObject*, QEvent*)
() from /home/tobias/kde/qt5/lib/libQt5Core.so.5
#9  0x75770eb2 in QApplicationPrivate::sendMouseEvent(QWidget*,
QMouseEvent*, QWidget*, QWidget*, QWidget**, QPointer&, bool) ()
   from /home/tobias/kde/qt5/lib/libQt5Widgets.so.5
#10 0x757c4113 in ?? () from
/home/tobias/kde/qt5/lib/libQt5Widgets.so.5
#11 0x757c6729 in ?? () from
/home/tobias/kde/qt5/lib/libQt5Widgets.so.5
#12 0x7576a3fc in QApplicationPrivate::notify_helper(QObject*, QEvent*)
() from /home/tobias/kde/qt5/lib/libQt5Widgets.so.5
#13 0x757719cf in QApplication::notify(QObject*, QEvent*) () from
/home/tobias/kde/qt5/lib/libQt5Widgets.so.5
#14 0x74cfa558 in QCoreApplication::notifyInternal2(QObject*, QEvent*)
() from /home/tobias/kde/qt5/lib/libQt5Core.so.5
#15 0x751033db in
QGuiApplicationPrivate::processMouseEvent(QWindowSystemInterfacePrivate::MouseEvent*)
()
   from /home/tobias/kde/qt5/lib/libQt5Gui.so.5
#16 0x75104f55 in
QGuiApplicationPrivate::processWindowSystemEvent(QWindowSystemInterfacePrivate::WindowSystemEvent*)
()
   from /home/tobias/kde/qt5/lib/libQt5Gui.so.5
#17 0x750ddc3b in
QWindowSystemInterface::sendWindowSystemEvents(QFlags)
()
   from /home/tobias/kde/qt5/lib/libQt5Gui.so.5
#18 0x7fffe90a88eb in ?? ()
   from
/media/tobias/edf72459-168b-43b6-b35a-f2b198488c25/tobias/Dev/OSScontrib/kde/qt5/plugins/platforms/../../lib/libQt5XcbQpa.so.5
#19 0x74cf888a in
QEventLoop::exec(QFlags) () from
/home/tobias/kde/qt5/lib/libQt5Core.so.5
#20 0x74d01a40 in QCoreApplication::exec() () from
/home/tobias/kde/qt5/lib/libQt5Core.so.5
#21 0x5556197c in main ()

I'm setting this to confirmed, but note that in worst case you and me saw two
unrelated issues with the same document.

-- 
You are receiving this mail because:
You are the assignee for the bug.