[okular] [Bug 317346] Okular overwrites PDF form data without asking

2017-09-09 Thread Andrew Chen 
https://bugs.kde.org/show_bug.cgi?id=317346

Andrew Chen  changed:

   What|Removed |Added

 CC||andrew.chuanye.chen@gmail.c
   ||om

--- Comment #4 from Andrew Chen  ---
Related bugs:
https://bugs.kde.org/show_bug.cgi?id=362996
https://bugs.kde.org/show_bug.cgi?id=267350

In my opinion okular should save form data out of band only for recovery
purposes, and when exiting prompt the user to either save the in the pdf itself
or discard the data.

I'm happy to work on this if no one else is.

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

[okular] [Bug 229552] "Go to page" should allow relative jumps

2017-09-09 Thread doug strain 
https://bugs.kde.org/show_bug.cgi?id=229552

doug strain  changed:

   What|Removed |Added

 CC||t...@digimeas.com

--- Comment #4 from doug strain  ---
Hi!  
Came here to request this feature. 

I'm looking at a 2191 page scanned (that is, I can't search it) pdf whose TOC
lists chapters and pages. E.g. "Integrated Control Unit" is listed as "23-156".
It DOESN'T say that it's on page 1628.

It could be as simple as evaluating "+169", or accepting +/- after the page
number, e.g. "1473+169".  Either/both of those seem a trivial fix (from the
outside).

As for "other software", Okular could be a trend-setter.

Binary searches with a calculator are so 20th century :-)

Thanks
Doug

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

[okular] [Bug 315552] Allow save annotations directly to original pdf file

2017-09-09 Thread Arthur Țițeică 
https://bugs.kde.org/show_bug.cgi?id=315552

Arthur Țițeică  changed:

   What|Removed |Added

 CC||arthur+...@cloud.titeica.ro

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

[okular] [Bug 286825] Okular should generate PDFs when printing instead of GhostScript

2017-09-09 Thread Arthur Țițeică 
https://bugs.kde.org/show_bug.cgi?id=286825

Arthur Țițeică  changed:

   What|Removed |Added

 CC||arthur+...@cloud.titeica.ro

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

[okular] [Bug 376990] Okular randomly crashes when launched from Emacs

2017-09-09 Thread Tristan Miller 
https://bugs.kde.org/show_bug.cgi?id=376990

Tristan Miller  changed:

   What|Removed |Added

   See Also||https://bugs.kde.org/show_b
   ||ug.cgi?id=384540

--- Comment #14 from Tristan Miller  ---
(In reply to Albert Astals Cid from comment #13)
> Defenitely not an okular bug, i guess a Plasma bug makes sense since you say
> that you can only reproduce this inside a Plasma session.

OK, reported as Bug 384540.

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

D7714: Show dialog to ask when closing when more than tab open

2017-09-09 Thread Henrik Fehlauer 
rkflx added a subscriber: colomar.
rkflx added a comment.


  > What part of "Are you sure you want to quit" Yes/No you find ambiguous?
  
  KDE's HIG says
  
Label command buttons with an imperative verb.
  
  and also
  
Use descriptive button labels instead of standard Yes/No or OK/Cancel 
buttons. For example, if the user must choose to continue or stop an action, 
provide the buttons "Continue" and "Cancel".
  
  and IIRC there was even a time when a lot of KDE dialogs were converted from 
Yes/No style to a more action oriented style.
  
  > This has nothing to do with "this window", it's about the application 
itself being closed.
  
  Try opening multiple Okular windows with tabs and quit. Only one window (from 
the perspective of the user) will be closed.
  
  > Why is warning better?
  
  The "i" in the icon stands for "information", but in reality you are asking 
the user to pick between two buttons (the "warning" icons signals "attention, 
decide between two things!" with the imperative literally depicted by "!") Of 
course, a question mark would also work (maybe even better).
  
  > That's why firefox wording that i copied is much better
  
  Fair enough. It would be a nice touch if you could also change it in Dolphin 
then, though.
  
  > that doesn't have anything to do with this, we have multiple "don't ask me 
again" already, so this can't be a blocker.
  
  I thought it was a relevant question, so I brought it up. I'm sorry you feel 
attacked.
  
  ---
  
  For the rest, let's see if @colomar has any tips.

REPOSITORY
  R223 Okular

REVISION DETAIL
  https://phabricator.kde.org/D7714

To: aacid
Cc: colomar, rkflx, #okular, aacid

D7671: Fix automatic reload of files saved with QSaveFile

2017-09-09 Thread Albert Astals Cid 
aacid added a comment.


  In https://phabricator.kde.org/D7671#143898, @rkflx wrote:
  
  > After further analysis, I can draw these conclusions (contradicting some 
assumptions from above, confirming others):
  >
  > 1. KDirWatch correctly emits at least one "dirty" each for directory 
watches when files are created, modified, deleted or moved_to (i.e., KDirWatch 
is not the reason for the bug).
  
  
  "Correctly" is an overstatement in my opinion, it should emit as many dirty 
as things that make it dirty, i.e. i disagree  KDirWatch is not a reason for 
this bug (if we ignore possible races)
  
  > 2. The directory code path in Okular only works for deleted files, but 
never works for moved files (that's the real reason for the bug).
  > 3. The file code path in Okular would work, if KDirWatch emitted dirty for 
a file on move_to (but KDirWatch does not and should not).
  
  Again i disagree here, it's emitting a removed/added signal, so why wouldn't 
it emit dirty?

REPOSITORY
  R223 Okular

REVISION DETAIL
  https://phabricator.kde.org/D7671

To: progwolff, aacid
Cc: sander, rkflx, #okular, aacid

[okular] [Bug 376990] Okular randomly crashes when launched from Emacs

2017-09-09 Thread Albert Astals Cid 
https://bugs.kde.org/show_bug.cgi?id=376990

--- Comment #13 from Albert Astals Cid  ---
Defenitely not an okular bug, i guess a Plasma bug makes sense since you say
that you can only reproduce this inside a Plasma session.

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

D7714: Show dialog to ask when closing when more than tab open

2017-09-09 Thread Albert Astals Cid 
aacid added a comment.


  In https://phabricator.kde.org/D7714#144118, @rkflx wrote:
  
  > Thanks for picking up old reviewboard requests.
  
  
  It's what being the defacto maitainer means
  
  > I like the feature in general, it is standard in a lot of applications 
today.
  > 
  > But to make it easier to use, users should not relearn the UI they are 
already familiar with. This is what Dolphin currently provides vs. the patch:
  > 
  > | F3903699: dolphin-tab-quit.png  | 
F3903701: okular-tab-quit.png  |
  > |
  > 
  > Besides the relearning aspect, I find the dialog in Dolphin preferable 
because of the following properties:
  > 
  > - Explicit actions on the buttons instead of ambiguous Yes/No which 
requires reading of the explanation.
  
  What part of "Are you sure you want to quit" Yes/No you find ambiguous?
  
  > - Slightly better wording ("you have...open in this window" vs. "you have 
tabs.").
  
  This has nothing to do with "this window", it's about the application itself 
being closed.
  
  > - Warning icon instead of information icon.
  
  Why is warning better? IMO Warning is for "you better pay attention answering 
here because if not something will be very wrong", nothing will happen if you 
answer wrong that question, worse case scenario you have to reopen some files.
  
  > - Standard do-not-ask-again message instead of custom message (also, "when 
closing more than one tab" does not correspond directly to the "quit" action 
the user initiated and thus might be confusing, as you cannot close e.g. 4 out 
of 5 tabs).
  
  Do not ask again is not what you want. Because if you say "No" and "Do not 
ask again" suddently you're fucked and you can't quit. That's why firefox 
wording that i copied is much better
  
  > - Third button with the correct action for those of us hitting [Ctrl] + [Q] 
while we meant [Ctrl] + [W].
  
  You press esc and then Ctrl+W ;)
  
  > - (Okular's dialog title is better, though.)
  > 
  >   Thus, I would prefer something akin to the Dolphin dialog being 
implemented in Okular.
  
  Sorry i don't see any major benefit in Dolphin's dialog.
  
  > Lastly, Dolphin has a way to reset the warning status in the settings, is 
there any way to reset it in Okular too?
  
  No, but that doesn't have anything to do with this, we have multiple "don't 
ask me again" already, so this can't be a blocker. You're welcome to implement 
that.

REPOSITORY
  R223 Okular

REVISION DETAIL
  https://phabricator.kde.org/D7714

To: aacid
Cc: rkflx, #okular, aacid

[okular] [Bug 376990] Okular randomly crashes when launched from Emacs

2017-09-09 Thread Tristan Miller 
https://bugs.kde.org/show_bug.cgi?id=376990

--- Comment #12 from Tristan Miller  ---
(In reply to Luigi Toscano from comment #11)
> If you can reproduce with even demo/simple Qt5 applications, and maybe from
> an environment which is not Plasma, this may be a bug or a quirk in Qt; you
> may try to report it to Qt.

I can reproduce the problem reported by jap...@linux01.gwdg.de (i.e., KDE
applications launched from Emacs crash when their pull-down menus are
activated).  The problem occurs only when running Plasma, and only with KDE
applications.  (That is, the problem does not occur with other desktop
environments, such as LXQt, and the problem does not occur with non-KDE Qt
applications, such as Clementine and Skype 4.3.0.37.)

Given this, where would be the best place to report this bug?

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