FrameworksRelease
ActivePublic

Recent Activity

Today

chinmoyr requested review of D10818: Store PolicyKit action which the slave is authorized to perform.
Sun, Feb 25, 9:41 AM · Frameworks
Restricted Application edited projects for D10235: [server] Add support for the frame semantics of Pointer version 5, added: Plasma on Wayland; removed Plasma.
Sun, Feb 25, 9:27 AM · Plasma on Wayland, Frameworks
vkrause added a comment to D7864: Basic support for PHP and Python fenced code blocks in Markdown.

@vkrause Do you already have something in mind, when you say this does not scale? ...

Sun, Feb 25, 9:24 AM · Frameworks
romangg edited projects for D10235: [server] Add support for the frame semantics of Pointer version 5, added: Plasma on Wayland; removed Plasma.
Sun, Feb 25, 9:22 AM · Plasma on Wayland, Frameworks
romangg accepted D10235: [server] Add support for the frame semantics of Pointer version 5.
Sun, Feb 25, 9:21 AM · Plasma on Wayland, Frameworks
rkflx added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

What about users who manually create and mount FUSE-type encrypted directories? Plasma Vault is just a nice interface for this workflow, after all.

Sun, Feb 25, 8:03 AM · Plasma, Frameworks, Baloo
Restricted Application edited projects for D10235: [server] Add support for the frame semantics of Pointer version 5, added: Plasma on Wayland; removed Plasma.

Shouldn't the sendFrame event sent by the compositor explicitly? The way it's currently send after each event compositors are not able to group these.

Sun, Feb 25, 8:00 AM · Plasma on Wayland, Frameworks
michaelh added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

I was thinking in the same direction. When the mountpoint of a vault is changed (or new) a warning message about baloo's activity should be displayed. The message should be displayed regardless of baloo's enabled state. There's plenty of space for that in the dialogs.
The user could then disable indexing with balooctl or via systemsettings.
IMO configuration of this should be done by Plasma Vault anyway. Baloo's job would be to provide the tools to do it. It's framework after all and should make as few decisions as possible.

Sun, Feb 25, 7:10 AM · Plasma, Frameworks, Baloo
julianstirling planned changes to D10719: Highlighting for OpenSCAD.

I am confused. If I look in the commit history of this revision there are two diffs, the first commit, and then the subsequent commit. Can phabricator not handle both of them sequentially?

Sun, Feb 25, 1:18 AM · Frameworks

Yesterday

dhaumann updated the diff for D9211: Iterate over initializer_lists to avoid mem allocs.
  • Add and use contains()
  • Use QLaint1String
Sat, Feb 24, 10:21 PM · Frameworks, Kate
Restricted Application edited projects for D10235: [server] Add support for the frame semantics of Pointer version 5, added: Plasma; removed Plasma on Wayland.

Shouldn't the sendFrame event sent by the compositor explicitly? The way it's currently send after each event compositors are not able to group these.

Sat, Feb 24, 9:58 PM · Plasma on Wayland, Frameworks
dhaumann closed D7864: Basic support for PHP and Python fenced code blocks in Markdown.
Sat, Feb 24, 9:46 PM · Frameworks
dhaumann added a comment to D7864: Basic support for PHP and Python fenced code blocks in Markdown.

@vkrause Do you already have something in mind, when you say this does not scale? ...

Sat, Feb 24, 9:43 PM · Frameworks
dhaumann closed D10646: Support case-insensitive WordDetect.
Sat, Feb 24, 9:33 PM · Frameworks
alexeymin added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

Maybe when a new vault is configured, it should add itself to baloo exceptions.

Sat, Feb 24, 9:32 PM · Plasma, Frameworks, Baloo
dhaumann added a comment to D10646: Support case-insensitive WordDetect.

Ok, good catch. Given the uninitialized variables in the other classes, this is good. Will integrate.

Sat, Feb 24, 9:30 PM · Frameworks
dhaumann closed D10174: Add syntax highlighting for SELinux CIL Policies & File Contexts.
Sat, Feb 24, 9:17 PM · Framework: Syntax Highlighting, Frameworks
dhaumann accepted D10174: Add syntax highlighting for SELinux CIL Policies & File Contexts.

I'll remove the selColor attributes, since these will not work with dark color schemes.
Similar, I will remove all underline, italic, and bold attributes where not needed by default.

Sat, Feb 24, 9:17 PM · Framework: Syntax Highlighting, Frameworks
dhaumann closed D10595: Adding ctp file extension to the PHP syntax highlighting.
Sat, Feb 24, 9:04 PM · Frameworks
dhaumann accepted D10595: Adding ctp file extension to the PHP syntax highlighting.

Sounds good to me.

Sat, Feb 24, 9:03 PM · Frameworks
dhaumann closed D10783: Yacc/Bison: Fix the $ symbol and update syntax for Bison.

Done, see commit https://commits.kde.org/syntax-highlighting/f5a32272d26e3b46d4c4201a5d9531de8f23867a

Sat, Feb 24, 9:02 PM · Framework: Syntax Highlighting, Frameworks
dhaumann accepted D10783: Yacc/Bison: Fix the $ symbol and update syntax for Bison.

Looks good to me, will integrate.

Sat, Feb 24, 9:01 PM · Framework: Syntax Highlighting, Frameworks
dhaumann added a comment to D10719: Highlighting for OpenSCAD.

@julianstirling Can you profile the full xml file again? This patch does not contain the .xml file anymore, so in the current form I cannot commit this as is.

Sat, Feb 24, 8:58 PM · Frameworks
michaelh added a reviewer for D10807: fileindexerconfig: Exclude Vaults by default: Plasma.
Sat, Feb 24, 8:53 PM · Plasma, Frameworks, Baloo
michaelh added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

This is not a viable approach.
Any vaults created after baloo is configured may be indexed (unless they are inside ~/Vaults, of course).
For the near future I see no other way to exclude vaults from being indexed other than this one. :-(

Sat, Feb 24, 8:45 PM · Plasma, Frameworks, Baloo
michaelh added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

I honestly don't know.

@ivan any idea?

Sat, Feb 24, 8:37 PM · Plasma, Frameworks, Baloo
nicolasfella updated subscribers of D10807: fileindexerconfig: Exclude Vaults by default.

I honestly don't know.

Sat, Feb 24, 8:36 PM · Plasma, Frameworks, Baloo
romangg accepted D10142: [server] Properly handle the situation when the DataSource for a drag gets destroyed.
Sat, Feb 24, 8:34 PM · Plasma on Wayland, Frameworks
michaelh added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

You have to hit next a couple of times

I saw it. Ok, now what?
How to find out where the vaults are?

Sat, Feb 24, 8:33 PM · Plasma, Frameworks, Baloo
nicolasfella added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

You have to hit next a couple of times

Sat, Feb 24, 8:31 PM · Plasma, Frameworks, Baloo
michaelh added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

Sat, Feb 24, 8:30 PM · Plasma, Frameworks, Baloo
nicolasfella added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

Sat, Feb 24, 8:29 PM · Plasma, Frameworks, Baloo
nicolasfella added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

When creating a Vault you can choose any location for it. ~/Vaults is just the default

Sat, Feb 24, 8:29 PM · Plasma, Frameworks, Baloo
michaelh added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

Vault can also be mounted in other directories than ~/Vaults

How is that done?

Sat, Feb 24, 8:26 PM · Plasma, Frameworks, Baloo
nicolasfella added a comment to D10807: fileindexerconfig: Exclude Vaults by default.

Vault can also be mounted in other directories than ~/Vaults

Sat, Feb 24, 8:24 PM · Plasma, Frameworks, Baloo
michaelh updated the summary of D10810: fileindexerconfigtest: Add vault test.
Sat, Feb 24, 8:19 PM · Baloo, Frameworks
michaelh added a dependency for D10810: fileindexerconfigtest: Add vault test: D10809: autotests: Refactor fileindexerconfigtest.
Sat, Feb 24, 8:17 PM · Baloo, Frameworks
michaelh added a dependent revision for D10809: autotests: Refactor fileindexerconfigtest: D10810: fileindexerconfigtest: Add vault test.
Sat, Feb 24, 8:17 PM · Baloo, Frameworks
michaelh requested review of D10810: fileindexerconfigtest: Add vault test.
Sat, Feb 24, 8:16 PM · Baloo, Frameworks
romangg accepted D10081: Don't assert if used incorrectly from dbus.
Sat, Feb 24, 8:13 PM · Frameworks
michaelh requested review of D10809: autotests: Refactor fileindexerconfigtest.
Sat, Feb 24, 8:10 PM · Baloo, Frameworks
michaelh requested review of D10807: fileindexerconfig: Exclude Vaults by default.
Sat, Feb 24, 7:59 PM · Plasma, Frameworks, Baloo
kmorwinski added a watcher for Frameworks: kmorwinski.
Sat, Feb 24, 7:08 PM
elvisangelaccio updated the diff for D10312: FileUndoManager: don't delete non-existing local files.
  • Improved unit test
  • Replaced slotUnlock() with slotPop()
Sat, Feb 24, 6:28 PM · Frameworks
astippich added a comment to D10803: handle more tags in taglibextractor.

First of all, please review carefully for ABI compatibility, I do not know what to watch for that.

Sat, Feb 24, 6:19 PM · Frameworks
astippich requested review of D10803: handle more tags in taglibextractor.
Sat, Feb 24, 6:10 PM · Frameworks
markg added a comment to D10702: Always use a job to delete files to avoid freezing process waiting on IO.

While this might give you the expected result, it feels like a workaround.

I'm assuming the fast path is there for a reason and is really substantially faster then going through the job route.
If that is the case then the proper fix would be to make that code part async. That is obviously much more complex (otherwise it would've been done already).
Think of using std::async and a QEventLoop. Sounds difficult, right? It is :) But I've been playing with that kind of stuff lately so i'm happy to share an example that you can use as a starting point.
Here it is: https://p.sc2.nl/BygE-Oiwz

I wanted to paste it inline, but that already got quite big so a link it is.
I've added a bunch of comments in the code to explains what it's doing.
Note that the example does make a "QEventLoop", you should not do that within the if statement, but rather outside the while loop and simply call exec() and quit() every time (not making a new QEventLoop for every delete)

Lastly, please benchmark this fast pats (as it currently is) compared to your KIO version and my async version to see if the fast path really is the fast path. As we just don't know and that kinda influences which route to choose here.

Please correct if I am wrong but kio::file_delete will, in the end, calls FileProtocol::deleteRecursive(const QString &path) which works the same way using QFile::remove as before.
So I have the hypothesis that they should not be much difference in performance if any.

It could be impactful because of the KIO::file_delete call and I don't know precisely how expensive such a kio call is.
Or because they are more jobs class instanciated.
But those KIO::file_* calls are already used in a lot of cases in CopyJobPrivate::copyNextFile for instance to recursively copy files, so I know that at least in some cases it is ok.

Please correct me wherever I am wrong I am new in the KIO codebase. I am learning as I go.

The main drawback in this version of the patch currently, is that we loose progressive status update when a lot of files are being removed (slotReport was called each 300 deletion before).
And since needs to be fixed this at least.

Sat, Feb 24, 4:44 PM · Frameworks
davidk added a comment to D8532: [WIP] Restrict file extractor with Seccomp.

Sorry for the late reply and the slow process in general. Reallife keeps me busy...

A whitelist, even if it is broad, would be desirable to reduce the attack surface of the kernel, and is also the way it has been done for Gnome Tracker. But the concerns about maintenance remain, it probably should be tested regularly. Are there ways this can be automated?

If we want to test this, we would need a directory with files for each extractor (kfilemetadata includes such files for its autotests). Then, we should configure seccomp to kill the process if it calls a prohibited syscall. The test should then index all files in the directory. Unfortunately we can't test some things, e.g. the dbus integration and communication with baloo_file. This would need a test which starts the whole extractor as a child process. But i'm not sure if thats feasible. What do you think?

In case the decision goes in favor of the blacklist, would it be possible to add ptrace, process_vm_readv, process_vm_writev?

That's possible of course.

Sat, Feb 24, 4:37 PM · Frameworks
meven added a comment to D10702: Always use a job to delete files to avoid freezing process waiting on IO.

While this might give you the expected result, it feels like a workaround.

I'm assuming the fast path is there for a reason and is really substantially faster then going through the job route.
If that is the case then the proper fix would be to make that code part async. That is obviously much more complex (otherwise it would've been done already).
Think of using std::async and a QEventLoop. Sounds difficult, right? It is :) But I've been playing with that kind of stuff lately so i'm happy to share an example that you can use as a starting point.
Here it is: https://p.sc2.nl/BygE-Oiwz

I wanted to paste it inline, but that already got quite big so a link it is.
I've added a bunch of comments in the code to explains what it's doing.
Note that the example does make a "QEventLoop", you should not do that within the if statement, but rather outside the while loop and simply call exec() and quit() every time (not making a new QEventLoop for every delete)

Lastly, please benchmark this fast pats (as it currently is) compared to your KIO version and my async version to see if the fast path really is the fast path. As we just don't know and that kinda influences which route to choose here.

Sat, Feb 24, 4:29 PM · Frameworks
ngraham closed D10671: [UDisks] Fix auto-mount regression.
Sat, Feb 24, 2:16 PM · Frameworks