D4972: Start drag with press and hold on touch events

davidedmundson added a comment. Kicker uses a normal mouse area for drag starting, then a different plasmoid specific C++ helper for mimedata REPOSITORY R296 KDeclarative REVISION DETAIL https://phabricator.kde.org/D4972 To: davidedmundson, #plasma Cc: broulik, plasma-devel,

D4972: Start drag with press and hold on touch events

broulik added a comment. Even Qt's own `Drag` attached property crashes when the item it's attached to is deleted as I had to experience, cf https://phabricator.kde.org/D4976 REPOSITORY R296 KDeclarative REVISION DETAIL https://phabricator.kde.org/D4972 To: davidedmundson, #plasma Cc:

D4972: Start drag with press and hold on touch events

hein added a comment. So do Task Manager and Folder View. Because DragArea is kind of broken and will crash Plasma if it's deleted during the drag (which happens quite often when you remove something by DND). REPOSITORY R296 KDeclarative REVISION DETAIL

D4972: Start drag with press and hold on touch events

broulik added a comment. Kickoff also uses DragArea iirc, I tend to use Qt's own Drag thing, though, but if we have better/proper touchscreen support in there then, might be worth re-considering. REPOSITORY R296 KDeclarative REVISION DETAIL https://phabricator.kde.org/D4972 To:

D4937: Add KFileWidget::setSelectedUrl()

fvogt accepted this revision. This revision is now accepted and ready to land. REPOSITORY R241 KIO BRANCH master REVISION DETAIL https://phabricator.kde.org/D4937 To: dfaure, fvogt Cc: fvogt, #frameworks

D4937: Add KFileWidget::setSelectedUrl()

dfaure closed this revision. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D4937 To: dfaure, fvogt Cc: fvogt, #frameworks

Jenkins-kde-ci: kio master stable-kf5-qt5 » Linux,gcc - Build # 462 - Fixed!

GENERAL INFO BUILD SUCCESS Build URL: https://build.kde.org/job/kio%20master%20stable-kf5-qt5/PLATFORM=Linux,compiler=gcc/462/ Project: PLATFORM=Linux,compiler=gcc Date of build: Wed, 08 Mar 2017 12:11:44 + Build duration: 5 min 23 sec CHANGE SET Revision

Jenkins-kde-ci: kio master stable-kf5-qt5 » Linux,gcc - Build # 462 - Fixed!

GENERAL INFO BUILD SUCCESS Build URL: https://build.kde.org/job/kio%20master%20stable-kf5-qt5/PLATFORM=Linux,compiler=gcc/462/ Project: PLATFORM=Linux,compiler=gcc Date of build: Wed, 08 Mar 2017 12:11:44 + Build duration: 5 min 23 sec CHANGE SET Revision

D4937: Add KFileWidget::setSelectedUrl()

dfaure added a comment. BTW don't port the callers to this new method yet, we'll have to wait until the KF5 minimum requirement is raised in kdialog and plasma-integration. I made a TODO for June ;) REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D4937 To: dfaure,

D4975: Prevent misdetection of EOF on Linux

hindenburg added a comment. Thanks, this needs to get reviewed by the framework team that handles KPty. REPOSITORY R291 KPty REVISION DETAIL https://phabricator.kde.org/D4975 To: Lekensteyn, #konsole, hindenburg Cc: #frameworks

D4847: KAuth integration in document saving

martinkostolny updated this revision to Diff 12309. martinkostolny added a comment. Good point! I was about to say that the target folder is in most cases non-writable without elevated privileges. But we can actually use KAuth action twice, for 2 simple jobs: 1. Create temporary file

D4799: Delay notifications until desktop session has loaded

vpilo updated this revision to Diff 12310. vpilo added a comment. New version, using the idea suggested by @davidedmundson . The KNotifications manager was clearing queued notifications also when the fd.o Notifications service initially registers; and while I get that it's done on

D4847: KAuth integration in document saving

dfaure added a comment. I don't know the full architecture of this kauth stuff, but isn't it possible to save to .tmp rather than /tmp, ~/.cache or anywhere else? Then instead of "more likely" we are *sure* it's on the same device as the destination. REPOSITORY R39 KTextEditor REVISION

Re: Review Request 129983: [RFC] PoC patch for polkit support in kio.

> On March 6, 2017, 11:38 p.m., Aleix Pol Gonzalez wrote: > > src/ioslaves/file/file.cpp, line 1378 > > > > > > I'm not convinced, what's the point of requestroot? Why don't you let > > it call the action

D4968: Mac OS (X) has a POSIX-compliant /bin/sh too

rjvbb added a comment. In https://phabricator.kde.org/D4968#93535, @dfaure wrote: > Can you use arc diff to upload changes? Then context would be available in the diffs (I think). This *was* uploaded with arc diff (through KDevelop's new export-to-Phabricator plugin)... With

D4968: Mac OS (X) has a POSIX-compliant /bin/sh too

dfaure accepted this revision. dfaure added a comment. This revision is now accepted and ready to land. Can you use arc diff to upload changes? Then context would be available in the diffs (I think). REPOSITORY R244 KCoreAddons REVISION DETAIL https://phabricator.kde.org/D4968 To:

D4937: Add KFileWidget::setSelectedUrl()

dfaure updated this revision to Diff 12280. dfaure added a comment. Add test for relative URL and document the proper way to do it REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D4937?vs=12255=12280 BRANCH master REVISION DETAIL

D4968: Mac OS (X) has a POSIX-compliant /bin/sh too

bcooksley added a comment. Something is definitely odd about this revision - it doesn't appear to have been submitted using Arcanist. The branch, along with local commit information is missing, and Arcanist always uploads context. How does this KDevelop integration work? If it does

D4968: Mac OS (X) has a POSIX-compliant /bin/sh too

rjvbb added a comment. In https://phabricator.kde.org/D4968#93542, @bcooksley wrote: > Something is definitely odd about this revision - it doesn't appear to have been submitted using Arcanist. > The branch, along with local commit information is missing, and Arcanist always

D4975: Prevent misdetection of EOF on Linux

Lekensteyn created this revision. Restricted Application added a project: Frameworks. Restricted Application added a subscriber: Frameworks. REVISION SUMMARY On Linux, even if slave activity is detected on a master fd, FIONREAD can still return zero available bytes in the buffer. Previously