https://bugs.kde.org/show_bug.cgi?id=402017
Sune Vuorela changed:
What|Removed |Added
CC||chronoan...@web.de
--- Comment #78 from Sune
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #77 from Nate Graham ---
Git commit 99600667d44ce0002b1cb1f0dfb3db3c634bc4c6 by Nate Graham.
Committed on 08/06/2023 at 15:21.
Pushed by ngraham into branch 'master'.
Don't stop or warn if the edited file was deleted on disk
If an open
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #76 from Bug Janitor Service ---
A possibly relevant merge request was started @
https://invent.kde.org/graphics/okular/-/merge_requests/752
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=402017
Nate Graham changed:
What|Removed |Added
Latest Commit|https://invent.kde.org/grap |https://invent.kde.org/grap
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #74 from Nate Graham ---
Git commit 89f17155bffa20500d7f3f098b08f61eedcb780d by Nate Graham.
Committed on 02/05/2023 at 20:57.
Pushed by ngraham into branch 'master'.
Offer options when trying to overwrite externally-modified file
When
https://bugs.kde.org/show_bug.cgi?id=402017
Nate Graham changed:
What|Removed |Added
Version Fixed In||23.04.1
Latest Commit|
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #72 from Bug Janitor Service ---
A possibly relevant merge request was started @
https://invent.kde.org/graphics/okular/-/merge_requests/733
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=402017
Bug Janitor Service changed:
What|Removed |Added
Status|CONFIRMED |ASSIGNED
--- Comment #71 from Bug
https://bugs.kde.org/show_bug.cgi?id=402017
Greg B changed:
What|Removed |Added
CC||grg...@gmail.com
--- Comment #70 from Greg B ---
It
https://bugs.kde.org/show_bug.cgi?id=402017
Wing Hong Chan changed:
What|Removed |Added
CC||winghongc...@vivaldi.net
--
You are
https://bugs.kde.org/show_bug.cgi?id=402017
Jonas changed:
What|Removed |Added
CC||jonas...@gmail.com
--- Comment #69 from Jonas ---
I'm
https://bugs.kde.org/show_bug.cgi?id=402017
Kishore Gopalakrishnan changed:
What|Removed |Added
CC||kishor...@gmail.com
--
You are
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #68 from Karl ---
There was a time when creating a disk copy would have been seen as a
'bad-thing®', but today, with modern sizes of RAM and disk-drives, I don't
think it is probably the right answer.
If I look at the typical size of PDFs
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #67 from mateMat ---
Perfectly understood.
I terms of strategy, it seems to happen when we open a PDF on kmail or
elsewhere with Okular. Would it make sense to transfer the right to hold this
file from kmail to okular, this way when okular
https://bugs.kde.org/show_bug.cgi?id=402017
Oliver Sander changed:
What|Removed |Added
CC||oliver.san...@tu-dresden.de
--- Comment #66
https://bugs.kde.org/show_bug.cgi?id=402017
kde-b...@fnx.splvs.net changed:
What|Removed |Added
CC||kde-b...@fnx.splvs.net
--- Comment #65
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #64 from Göktuğ Kayaalp ---
I too have encountered this bug a lot, as recent as this morning where I
almost lost a PDF I had tracked down. I’m not knowledgeable in Qt or
C++, but I’m looking around in the codebase nevertheless, in the
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #63 from Nate Graham ---
Albert, we really need to fix this. Can we come up with a path forward?
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=402017
kingfa...@gmail.com changed:
What|Removed |Added
CC||kingfa...@gmail.com
--- Comment #62 from
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #61 from Karl ---
See post 59 - as this works - I've used it a few times - it is obvious that it
can be fixed within the program.
At the worst, there would be some system call - more likely the library calls
to do this already exist
https://bugs.kde.org/show_bug.cgi?id=402017
Yuri Chornoivan changed:
What|Removed |Added
CC||joerg...@web.de
--- Comment #60 from Yuri
https://bugs.kde.org/show_bug.cgi?id=402017
Karl changed:
What|Removed |Added
CC||k...@lrak.net
--- Comment #59 from Karl ---
This bug
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #58 from mateMat ---
Yes this bug is a pain, I lost a lot of work in annotations and comments
yesterday.
Since I am not a KDE dev and not a software developper in general, I can only
give you a external observation of this nasty bug.
I
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #57 from Nate Graham ---
Can we please put our heads together and come up with a solution for this
issue?
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=402017
Nate Graham changed:
What|Removed |Added
CC||math...@valat.eu
--- Comment #56 from Nate
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #55 from Steve Kelem ---
This bug has been languishing for two years.
Can't you apply the fixes you have?
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=402017
Albert Astals Cid changed:
What|Removed |Added
CC||st...@kelem.net
--- Comment #54 from
https://bugs.kde.org/show_bug.cgi?id=402017
Göktuğ Kayaalp changed:
What|Removed |Added
CC||s...@gkayaalp.com
--
You are receiving this
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #53 from Phil ---
I propose the bug report be titled, "Cannot save PDF when loaded file has been
deleted *or touched by any other process*."
To reproduce:
1. open $FILE in okular.
2. Highlight a word.
3. touch $FILE
Result: "There are
https://bugs.kde.org/show_bug.cgi?id=402017
Phil changed:
What|Removed |Added
CC||pv.bugzi...@gmail.com
--- Comment #52 from Phil ---
0.
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #51 from Albert Astals Cid ---
@Sven thanks for the refresher :)
Honestly the only good solution i see is getting the file keeper to work, and
hopefully after 12 years file watching has improved and the file keeper doesn't
break it.
https://bugs.kde.org/show_bug.cgi?id=402017
Luigi Toscano changed:
What|Removed |Added
CC||luigi.tosc...@tiscali.it
--- Comment #50 from
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #49 from Sven Brauch ---
> On your ext4 home partition it should work, but if
> you use FAT for some reason, or even the thumb
> drive fails, the data is still lost.
Ok, but now we have strayed quite far from the original goal. Those are
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #48 from David Hurka ---
Sven wrote:
> if they do open("filename.tmp"), write(), close(),
> rename("filename.tmp", "filename"), you get the old contents,
> because the new contents are in a *different* file (which now
> happens to have the
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #47 from Sven Brauch ---
Sorry for not checking back, I had a look; for PDF it actually works perfectly
by just removing the error message, but for other file types (e.g. PNG) we
would need to keep the FD open manually. There actually
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #46 from David Hurka ---
This means reverting 559836c3 would make Okular work nicely with Firefox in
many cases, but probably not all cases.
I think we can agree on some priorities now?
1. Work nicely with Firefox in many cases (I. e.
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #45 from Albert Astals Cid ---
This is what the old posix says (or at least i remember/understood from my
university years): "Once you have a file descriptor from a file open, the file
can disappear, change, whatever, you will still read
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #44 from David Hurka ---
> Personally, i would suggest trying to figure out why poppler fails and
> fix it other than doing all strange things that need the user to read
> lots of stuff that has "if this and if that and not that"
I looked
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #43 from Albert Astals Cid ---
(In reply to David Hurka from comment #41)
> > I mean think of this message you would have to show the user "Sorry, we
> > can't open your file because you don't have enough space on disk".
>
> If there is
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #42 from Albert Astals Cid ---
(In reply to David Hurka from comment #40)
> > it moves us away from being able to load as you go PDF linearized files.
> Hmm, ok.
>
> Since this is not closed as INTENDED, I assume there must be another
>
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #41 from David Hurka ---
> I mean think of this message you would have to show the user "Sorry, we
> can't open your file because you don't have enough space on disk".
If there is not enough space to create a temporary copy, how should the
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #40 from David Hurka ---
> it moves us away from being able to load as you go PDF linearized files.
Hmm, ok.
Since this is not closed as INTENDED, I assume there must be another solution.
A) Keep the current swapBackingFile etc.
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #39 from Albert Astals Cid ---
I don't think copying the file before opening is acceptable.
It uses my hard disk more than it should, it wears my hard disk more than
it should and it moves us away from being able to load as you go PDF
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #38 from David Hurka ---
There exists this code in part.cpp:
> if (m_document->canSaveChanges()) {
> [...]
> } else {
> // If the generators doesn't support saving changes, we will
> // just copy the original file.
> if
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #37 from Nate Graham ---
Hmm, that's an idea, yeah.
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #36 from David Hurka ---
If some generators can’t keep their own document file, we should probably
consider every file a remote file, i. e. make a temporary copy of every file
and open that instead. Right?
--
You are receiving this mail
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #35 from David Hurka ---
Below is the commit that made Okular refuse to save if the file is modified.
Looks like Sven Brauch was basically trying to revert this commit, but we
should first figure out what’s the problem with poppler.
commit
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #34 from Albert Astals Cid ---
If you can make it work without breaking any features, i don't see the problem.
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #33 from Sven Brauch ---
There's a bit of work around that place needed to make "Save" work properly
("Save as" does automatically, but not "Save"), so I'd like to hear Albert's
opinion first ;)
--
You are receiving this mail because:
You
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #32 from Nate Graham ---
Perhaps a relic from the past when things did not used to work properly? If you
send a merge request, I can help test it.
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #31 from Sven Brauch ---
I tried to just fix this now, and I am more confused than ever about this
issue. Just removing the error message and early return makes Save As work
flawlessly. Why is the error message even there?
--
You are
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #30 from Sven Brauch ---
Regarding other applications: most applications (e.g. kate) are not affected by
this issue, since they just load the whole file into memory, and save it from
their in-memory buffer if requested. Gwenview also seems
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #29 from Sven Brauch ---
I was repeatedly bitten by this issue in the past as well and I don't quite get
it. Can Okular not simply keep an open file handle for the document, which
results in ext4 not removing the data from the disk, even if
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #28 from Pino Toscano ---
(In reply to Nate Graham from comment #27)
> If Firefox isn't interested in playing ball, then our only realistic option
> is work around it in Okular. This isn't ideal, but seems preferable to
> letting user data
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #27 from Nate Graham ---
That's a shame. Thanks for the history.
If Firefox isn't interested in playing ball, then our only realistic option is
work around it in Okular. This isn't ideal, but seems preferable to letting
user data be
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #26 from Pino Toscano ---
(In reply to Nate Graham from comment #25)
> Is there a technically feasible way for Firefox to do something better?
In the same way XDG desktop environment do it: by handling the XDG base
directory and XDG
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #25 from Nate Graham ---
Is there a technically feasible way for Firefox to do something better?
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #24 from Albert Astals Cid ---
(In reply to David Hurka from comment #23)
> And I just want to add: It’s not a problem in Firefox, but in Okular.
> Firefox says “Hey Okular, please open this file and make the user happy!”,
> and not “Hey
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #23 from David Hurka ---
> But if you grep for saving in the okular poppler generator you'll find
> PDFGenerator::save. You can pull the thread from there.
Thanks. The related error message for annotated documents is “There are unsaved
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #22 from Albert Astals Cid ---
(In reply to David Hurka from comment #21)
> Besides that, how does saving the file after annotating work? There must be
> some function in Poppler to save the loaded and modified file, right?
I don't
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #21 from David Hurka ---
I still don’t completely understand the issue with FileKeeper. Is it:
a) FileKeeper keeps Okulars own file handle opened, so Okular itself will not
notice changes to the file through *this* handle?
b) FileKeeper
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #20 from Albert Astals Cid ---
The other thing is trying to figure out if we can just use poppler's file
handle to save the data (that should still be open), last time i tried it
didn't seem to be working well at all, but i have been known
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #19 from Albert Astals Cid ---
> the handle is kept for the _old_ file, so when some PDF producer (latex,
> etc) creates a new file by rename()ing the new temporary file on the old name
> (which is an atomic operation), you don't notice
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #18 from bark mallard ---
Just came across another variation of this problem. I have many documents
open on many workspaces, managing various inter-related workfklows. It is
inevitable in this scenario that I sometimes open the same
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #17 from Nate Graham ---
(In reply to Pino Toscano from comment #16)
> This won't behave nicely in editing sessions e.g. in kile or in other
> document producers, where you edit the sources and save+recompile to PDF
> often, resulting in a
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #16 from Pino Toscano ---
(In reply to David Hurka from comment #13)
> @Pino: Can you tell more about the problem? Can it be fixed, or do we need
> another approach than creating a file handle?
See bug 163363.
Holding an open descriptor
https://bugs.kde.org/show_bug.cgi?id=402017
Yuri Chornoivan changed:
What|Removed |Added
CC||yurc...@ukr.net
--- Comment #15 from Yuri
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #14 from Nate Graham ---
Temporarily wasting some memory or disk space is a far better alternative than
data loss. :)
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=402017
David Hurka changed:
What|Removed |Added
CC||p...@kde.org
--- Comment #13 from David Hurka
https://bugs.kde.org/show_bug.cgi?id=402017
Nate Graham changed:
What|Removed |Added
Priority|NOR |VHI
--- Comment #12 from Nate Graham ---
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #11 from bark mallard ---
This bug caused Okular to lose pdf files opened from Firefox, using Okular
v.1.3.3, installed from the standard Ubuntu 18.04.4 repository, for a fresh
installation about a month ago.
Okular said it couldn't save
https://bugs.kde.org/show_bug.cgi?id=402017
David Hurka changed:
What|Removed |Added
CC||markjball...@googlemail.com
--- Comment #10 from
https://bugs.kde.org/show_bug.cgi?id=402017
Patrick Silva changed:
What|Removed |Added
CC||bugsefor...@gmx.com
--
You are receiving this
https://bugs.kde.org/show_bug.cgi?id=402017
Nate Graham changed:
What|Removed |Added
Severity|wishlist|grave
--- Comment #9 from Nate Graham ---
More
https://bugs.kde.org/show_bug.cgi?id=402017
Richard Dwight changed:
What|Removed |Added
CC||richard.dwi...@gmail.com
--- Comment #8 from
https://bugs.kde.org/show_bug.cgi?id=402017
Nate Graham changed:
What|Removed |Added
CC||gr...@sub.red
--- Comment #7 from Nate Graham
https://bugs.kde.org/show_bug.cgi?id=402017
Nate Graham changed:
What|Removed |Added
CC||martin.marmso...@gmail.com
--- Comment #6 from
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #5 from David Hurka ---
Just discovered a class FileKeeper, which should do this.
Added in commit 6531398b919e58def8c6ba4711d20ea517147189, eleven years ago.
Just grep for #ifdef OKULAR_KEEP_FILE_OPEN
--
You are receiving this mail
https://bugs.kde.org/show_bug.cgi?id=402017
Albert Astals Cid changed:
What|Removed |Added
CC||m...@svenbrauch.de
--- Comment #4 from
https://bugs.kde.org/show_bug.cgi?id=402017
--- Comment #3 from Jens ---
This is true. I can, however, "Print" to a new PDF file.
This is bad because it loses quality and embedded text and comments /
annotations, but it proves the file is still available in memory and *could* be
saved.
--
You
https://bugs.kde.org/show_bug.cgi?id=402017
Albert Astals Cid changed:
What|Removed |Added
Severity|normal |wishlist
CC|
https://bugs.kde.org/show_bug.cgi?id=402017
Nate Graham changed:
What|Removed |Added
CC||n...@kde.org
Ever confirmed|0
82 matches
Mail list logo
