Minutes monday meeting week 25

[Bhushan Shah]

Hello,

please find the notes for meeting, sorry for delay

fvogt:

15:31 <fvogt> I can't join due to lunch, but my topics would be bug 394534 and https://phabricator.kde.org/D13535 - have fun!
15:31 <sKreamer> KDE bug 394534 in systemsettings (kcm_mouse) "kcminit crash on startup" [crash,] https://bugs.kde.org/show_bug.cgi?id=394534

d_ed:

15:32 <d_ed> Spent last week on  the stable XDGShell , followed by wayland fractional scaling and cursor scaling
15:32 <d_ed> all of which are going fine
15:33 <d_ed> and that's it.. there's a few minor bugs from the release coming in, but surprisingly nothing too major

Riddell:

15:34 <Riddell> 5.13.1 is out tomorrow!
15:34 <Riddell> get your bugfixes in
15:34 <Riddell> do we need a kickoff meeting for 5.14?
15:35 <Riddell> is IRC the best format again?
15:35 <Riddell> I'll send round a doodle poll
15:36 <Riddell> next()

notmart:

 * helping on mouse kcm port
 * fixed some popup menus in kirigami that weren't working on Qt 5.9
 * Kirigami: start of a refactor that would move the toolbarheader into the pagerow: this will solve most resizing and scrolling glitches
 * Kirigami: fixed pull to refresh
 * Wayland virtual desktop: new approach to desktops order D12820, should be pretty much the final form
 * Wrote a little tool to autogenerate videos or screenshots from qml, needed for the work on the HIG, to generate images and screencasts for the visual examples without wasting too much time https://cgit.kde.org/scratch/mart/qmlgrabber.git/

Sho_:

15:39 <Sho_> * [FV] Pressing Shift now swaps the Trash for the Delete action when only one is enabled (now matching Dolphin behavior)
15:39 <Sho_> Aside from this, I spent more time meeting with my GSoC student working on the new input settings KCM
15:39 <Sho_> Here's a short video demo of what it can do now: https://youtu.be/tbgM5GhIsaw
15:40 <Sho_> What you can see there is both a basic XKB-based keyboard layout and an input source backed by fcitx, and it's possible to add/remove/configure these, order them, as well as switch between them using global hotkeys, and the code takes care of e.g. reconfiguring the fcitx daemon to satisfy whatever state dynamically
15:41 <Sho_> When fcitx goes away, it dynamically marks the fcitx-backed ones unavailable (similar to the Language KCM) and switches to the highest-priority still-available one
15:41 <Sho_> We have a few more features to work on, including the XKB options stuff and retiring the old keyboard layout indicator and merging Aleix' work on a plasmoid-based replacement into kimpanel
15:42 <Sho_> Then we'd like to schedule a meeting with e.g. Marco and David and others interested to come up with options / a plan for how to auto-manage kimpanel on, well, the panel, similar to the old tray indicator
15:42 <Sho_> And then we want to put it on Phab for code review and VDG UI design feedback/iteration
15:42 <Sho_> If there's any time left at the end of GSoC after this we'll also add-in another backend for ibus, otherwise probably seperate to GSoC before the 5.14 release this is hopefully good for
15:43 <Sho_> I'll add that what's shown in in the video makes Plasma the first Linux desktop where this stuff works
15:43 <Sho_> Some competitors have some of this in a static kinda sorta way, but not properly dynamic
15:44 <bshah> That's quite cool stuff..
15:44 <Sho_> Now as for X11 vs. Wayland, the work our forebearers did in this area is already quite useful
15:44 <Sho_> There's a DBus wire protocol between the KCM and $a_dbus_service to manage the keyboard layout
15:44 <Sho_> on X11, this Dbus service is registered by a kded module
15:44 <Sho_> on Wayland, it's managed by KWin
15:45 <Sho_> currently we haven't had to change this wire protocol, although we did make a fair amount of changes to the kded module
15:45 <Sho_> for Wayland KWin may need some changes too, it's currenntly all a bit in flux, as are the Wayland protocols used by input method stuff anyway
15:45 <Sho_> we are working directly with Weng (kimpanel/fcitx maintainer) and the people involved are also talking to the Wayland people so
15:45 <Sho_> over time this will hopefully all come together
15:46 <Sho_> other than that I was pretty busy with non-Plasma stuff so there's not much more
15:46 <Sho_> some bugfixes in the pipe
15:47 <bshah> Sho_: is the work done here also communicated with the Purism people? as we're also in feedback look about their input-* stuff
15:47 <bshah> if not, probably it would be good idea ;)
15:48 <Sho_> bshah: both me and Weng have commented on the protocols they're working on. the work done on the KCM is somewhat orthogonal, because it's mostly about talking to e.g. fcitx or ibus, whereas the Wayland protocols are mostly about how fcitx and ibus talk to Kwin and clients
15:49 <Sho_> we'll inform them if there's anything they need to know though

bshah:

- Neon:
     - repairing some breakage with qt 5.11
- Plasma Mobile:
     - Working on image with qt 5.11 and upgraded plasma
     - image build done, needs testing

-- 
Bhushan Shah
http://blog.bshah.in
IRC Nick : bshah on Freenode
GPG key fingerprint : 0AAC 775B B643 7A8D 9AF7 A3AC FE07 8411 7FBC E11D
-------------- next part --------------
15:30 <bshah> meeting!
15:31 <d_ed> yo
15:31 <bshah> notmart: d_ed: Sho_: romangg: Riddell 
15:31 <fvogt> I can't join due to lunch, but my topics would be bug 394534 and https://phabricator.kde.org/D13535 - have fun!
15:31 <sKreamer> KDE bug 394534 in systemsettings (kcm_mouse) "kcminit crash on startup" [crash,] https://bugs.kde.org/show_bug.cgi?id=394534
15:31 <bshah> (and others)
15:31 <bshah> fvogt: thank
15:31 <bshah> *thanks
15:31 <Riddell> meeting!
15:31 <Riddell> https://phabricator.kde.org/D13535
15:31 <sKreamer> Diff 13535 "Do not cancel old clipboard selection if it is same as the new one." [Needs Review] https://phabricator.kde.org/D13535
15:31 <bshah> ?
15:32 <notmart> here
15:32 <Sho_> i'm here but don't have much to report
15:32  *	bshah have d_ed, Riddell, notmart and Sho_ in queue..
15:32 <bshah> d_ed starts
15:32 <d_ed> Spent last week on  the stable XDGShell , followed by wayland fractional scaling and cursor scaling
15:32 <d_ed> all of which are going fine
15:33 <d_ed> and that's it.. there's a few minor bugs from the release coming in, but surprisingly nothing too major
15:34 <d_ed> next()
15:34 <bshah> okay
15:34 <Riddell> 5.13.1 is out tomorrow!
15:34 <bshah> Riddell: your turn! how was your vacation
15:34 <Riddell> get your bugfixes in
15:34 <Riddell> do we need a kickoff meeting for 5.14?
15:34 <Sho_> I think so
15:35 <Riddell> is IRC the best format again?
15:35 <Sho_> there were some things we discussed at the Plasma sprint we should firm up and itemize, e.g. notifications rewrite
15:35 <bshah> yes
15:35 <Sho_> also Wayland stuff
15:35 <Riddell> I'll send round a doodle poll
15:36 <Riddell> next()
15:36 <notmart> yes irc please :)
15:36 <bshah> alright, if no questions for Riddell then notmart goes
15:37 <notmart> ok
15:37 <notmart> * helping on mouse kcm port
15:37 <notmart> * fixed some popup menus in kirigami that weren't working on Qt 5.9
15:37 <notmart> * Kirigami: start of a refactor that would move the toolbarheader into the pagerow: this will solve most resizing and scrolling glitches
15:37 <notmart> * Kirigami: fixed pull to refresh
15:37 <notmart> * Wayland virtual desktop: new approach to desktops order D12820, should be pretty much the final form
15:37 <sKreamer> Diff 12820 "Add KWayland virtual desktop protocol" [Needs Review] https://phabricator.kde.org/D12820
15:37 <notmart> * Wrote a little tool to autogenerate videos or screenshots from qml, needed for the work on the HIG, to generate images and screencasts for the visual examples without wasting too much time https://cgit.kde.org/scratch/mart/qmlgrabber.git/
15:37 <notmart> questions or next
15:38 <bshah> not a question but a thing, there is weirdnes about dialogs openqa in neon has detected and we would need to bug either you or someone to look into it (let's discuss post meeting)
15:39 <bshah> (it's about plasma dialogs)
15:39 <bshah> anyway, Sho_ your turn
15:39 <Sho_> * [FV] Pressing Shift now swaps the Trash for the Delete action when only one is enabled (now matching Dolphin behavior)
15:39 <Sho_> Aside from this, I spent more time meeting with my GSoC student working on the new input settings KCM
15:39 <Sho_> Here's a short video demo of what it can do now: https://youtu.be/tbgM5GhIsaw
15:40 <Sho_> What you can see there is both a basic XKB-based keyboard layout and an input source backed by fcitx, and it's possible to add/remove/configure these, order them, as well as switch between them using global hotkeys, and the code takes care of e.g. reconfiguring the fcitx daemon to satisfy whatever state dynamically
15:41 <Sho_> When fcitx goes away, it dynamically marks the fcitx-backed ones unavailable (similar to the Language KCM) and switches to the highest-priority still-available one
15:41 <Sho_> We have a few more features to work on, including the XKB options stuff and retiring the old keyboard layout indicator and merging Aleix' work on a plasmoid-based replacement into kimpanel
15:42 <Sho_> Then we'd like to schedule a meeting with e.g. Marco and David and others interested to come up with options / a plan for how to auto-manage kimpanel on, well, the panel, similar to the old tray indicator
15:42 <Sho_> And then we want to put it on Phab for code review and VDG UI design feedback/iteration
15:42 <Sho_> If there's any time left at the end of GSoC after this we'll also add-in another backend for ibus, otherwise probably seperate to GSoC before the 5.14 release this is hopefully good for
15:43 <Sho_> I'll add that what's shown in in the video makes Plasma the first Linux desktop where this stuff works
15:43 <Sho_> Some competitors have some of this in a static kinda sorta way, but not properly dynamic
15:44 <bshah> That's quite cool stuff..
15:44 <Sho_> Now as for X11 vs. Wayland, the work our forebearers did in this area is already quite useful
15:44 <Sho_> There's a DBus wire protocol between the KCM and $a_dbus_service to manage the keyboard layout
15:44 <Sho_> on X11, this Dbus service is registered by a kded module
15:44 <Sho_> on Wayland, it's managed by KWin
15:45 <Sho_> currently we haven't had to change this wire protocol, although we did make a fair amount of changes to the kded module
15:45 <Sho_> for Wayland KWin may need some changes too, it's currenntly all a bit in flux, as are the Wayland protocols used by input method stuff anyway
15:45 <Sho_> we are working directly with Weng (kimpanel/fcitx maintainer) and the people involved are also talking to the Wayland people so
15:45 <Sho_> over time this will hopefully all come together
15:46 <Sho_> other than that I was pretty busy with non-Plasma stuff so there's not much more
15:46 <Sho_> some bugfixes in the pipe
15:46 <Sho_> next()
15:47 <bshah> Sho_: is the work done here also communicated with the Purism people? as we're also in feedback look about their input-* stuff
15:47 <bshah> if not, probably it would be good idea ;)
15:48 <Sho_> bshah: both me and Weng have commented on the protocols they're working on. the work done on the KCM is somewhat orthogonal, because it's mostly about talking to e.g. fcitx or ibus, whereas the Wayland protocols are mostly about how fcitx and ibus talk to Kwin and clients
15:49 <Sho_> we'll inform them if there's anything they need to know though
15:49 <bshah> ah okay, sounds good
15:49 <bshah> okay, if no questions, I will close meeting with my (short) notes
15:49 <bshah> Neon:
15:49 <bshah>     - repairing some breakage with qt 5.11
15:49 <bshah> Plasma Mobile:
15:49 <bshah>     - Working on image with qt 5.11 and upgraded plasma
15:49 <bshah>     - image build done, needs testing
15:50 <bshah> questions? or we're done with meeting
15:50 <bshah> Thanks everyone for attending!
15:53 <Sho_> thanks for hosting!
15:55 <notmart> thnks very much bshah as usual :)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: not available
URL: <http://mail.kde.org/pipermail/plasma-devel/attachments/20180621/7cbd3bba/attachment.sig>