[frameworks-kitemviews] [Bug 487499] New: KWidgetItemDelegate wrongly emits warning about users deleting widgets on itemview destruction
https://bugs.kde.org/show_bug.cgi?id=487499 Bug ID: 487499 Summary: KWidgetItemDelegate wrongly emits warning about users deleting widgets on itemview destruction Classification: Frameworks and Libraries Product: frameworks-kitemviews Version: unspecified Platform: Other OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: general Assignee: k...@davidedmundson.co.uk Reporter: kosse...@kde.org CC: kdelibs-b...@kde.org Target Milestone: --- When subclassing KWidgetItemDelegate and installing it on an itemview, serving widgets by the overridden KWidgetItemDelegate::createItemWidgets(QModelIndex) method, once the itemview is destructed the internal logic wrongly triggers warning for each created widget claiming the user would have deleted that widget: --- 8< --- kf.itemviews: User of KWidgetItemDelegate should not delete widgets created by createItemWidgets! --- 8< --- For now consumers decided to work-around this by preventing the state triggering the warning, by deleting the KWidgetItemDelegate subclass instance explicitly before the itemview is destroyed, e.g.: https://invent.kde.org/unmaintained/kdelibs/-/commit/9aaa24814030041f51405182f03c9760e04bc76f https://invent.kde.org/kdevelop/kdevelop/-/commit/01445fdac0b6d171b7584318a00b5656f3f2db0d The wrong warnings are triggered due to this: the widgets returned by the createItemWidgets method are reparented to the itemview's viewport. When the itemview is destructed, it will delete the viewport as a child widget. And even if the delegate has the itemview as parent, if will always be deleted only after the viewport, as the widget children are deleted at the level of the itemview QWidget destructor, while the other QObject children only afterwards at the level of the itemview QObject destructor. And the viewport again deletes its widget children, to which the created item widgets now belong. And the QEvent::Destroy event then will be seen by the still around delegate's KWidgetItemDelegatePool member's KWidgetItemDelegateEventListener member, as it listens on the created item widgets' events, in KWidgetItemDelegateEventListener::eventFilter(QObject *watched, QEvent *event). Wrongly assuming it was the consumer side deleting the widget. https://invent.kde.org/unmaintained/kdelibs/-/commit/0225d2895a6f260aa918370e66db483c7d85cf7f added that check and warning, motivated by double deletion happening from ktorrent code itself deleting the item widgets as well. For unknown reasons that commit missed out that at the time already this would also trigger the warning for "normal" cases, since the reparenting to the viewport was changed half a year before in https://invent.kde.org/unmaintained/kdelibs/-/commit/cc38fd9ec8991cb8f89219ddea3efc35325725a3 The logic of the check and warning would need to detect whether the item widget destruction is a "proper" one done as part of the itemview destruction, or if it is a bad external one. Besides, the logic is also incomplete as it only cleans up the KWidgetItemDelegatePoolPrivate's widgetInIndex member, but leaves a reference in the usedWidgets one. Dumping analysis here for now. No idea yet how to fix this. -- You are receiving this mail because: You are watching all bug changes.
[kubrick] [Bug 486085] since last update 2 weeks ago, kubrick is rendered transparently and is unplayable
https://bugs.kde.org/show_bug.cgi?id=486085 Friedrich W. H. Kossebau changed: What|Removed |Added CC||kosse...@kde.org Version|unspecified |1.1.24022 -- You are receiving this mail because: You are watching all bug changes.
[kgraphviewer] [Bug 470523] font too small
https://bugs.kde.org/show_bug.cgi?id=470523 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |FIXED Status|REPORTED|RESOLVED Version Fixed In||2.5.0 Latest Commit||https://invent.kde.org/grap ||hics/kgraphviewer/-/commit/ ||acb5fec4e9614324acdbc4cfce4 ||f387df65bd0f4 --- Comment #1 from Friedrich W. H. Kossebau --- Git commit acb5fec4e9614324acdbc4cfce4f387df65bd0f4 by Friedrich W. H. Kossebau. Committed on 16/04/2024 at 09:16. Pushed by kossebau into branch 'master'. Fix bad defaults for fonts, also for colors, shapes & style The refactoring started in 117f22918f594a046e8e541701a3cd863539cdd0 seems to have missed out follow-up work to ensure defaults are used where not set explicitly, as before. e47e5ae6d38c0e16b90714f1887b827a00d4 later only ensured explicit default font size values exist. Dot2QtConsts covers many of the defaults indirectly. For QFont though it was using "" as fort family name, which resulted in random fonts being picked by Qt. The implementation now avoids adding explicit attribute instances, but instead returns default values if not set, for less runtime memory usage. Related: bug 477036 FIXED-IN: 2.5.0 M +59 -11 src/part/graphelement.cpp M +7-28 src/part/graphelement.h https://invent.kde.org/graphics/kgraphviewer/-/commit/acb5fec4e9614324acdbc4cfce4f387df65bd0f4 -- You are receiving this mail because: You are watching all bug changes.
[kgraphviewer] [Bug 388518] Feature Request: Relative paths as argument
https://bugs.kde.org/show_bug.cgi?id=388518 --- Comment #1 from Friedrich W. H. Kossebau --- Git commit ab98d533a99928347cfc36513d330f827c8331dd by Friedrich W. H. Kossebau. Committed on 16/04/2024 at 09:21. Pushed by kossebau into branch 'master'. Fix handling file cmdl arguments with relative path Related: bug 431414, bug 428456 FIXED-IN: 2.5.0 M +2-7src/main.cpp https://invent.kde.org/graphics/kgraphviewer/-/commit/ab98d533a99928347cfc36513d330f827c8331dd -- You are receiving this mail because: You are watching all bug changes.
[kgraphviewer] [Bug 428456] kgraphviewer-2.4.3 fails to show GraphViz graphs when they are supplied in the command line
https://bugs.kde.org/show_bug.cgi?id=428456 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |FIXED Latest Commit||https://invent.kde.org/grap ||hics/kgraphviewer/-/commit/ ||ab98d533a99928347cfc36513d3 ||30f827c8331dd Status|CONFIRMED |RESOLVED Version Fixed In||2.5.0 --- Comment #3 from Friedrich W. H. Kossebau --- Git commit ab98d533a99928347cfc36513d330f827c8331dd by Friedrich W. H. Kossebau. Committed on 16/04/2024 at 09:21. Pushed by kossebau into branch 'master'. Fix handling file cmdl arguments with relative path Related: bug 431414, bug 388518 FIXED-IN: 2.5.0 M +2-7src/main.cpp https://invent.kde.org/graphics/kgraphviewer/-/commit/ab98d533a99928347cfc36513d330f827c8331dd -- You are receiving this mail because: You are watching all bug changes.
[kgraphviewer] [Bug 477036] Graphs are shown with tamil fonts
https://bugs.kde.org/show_bug.cgi?id=477036 Friedrich W. H. Kossebau changed: What|Removed |Added Status|REPORTED|RESOLVED Version Fixed In||2.5.0 Resolution|--- |FIXED Latest Commit||https://invent.kde.org/grap ||hics/kgraphviewer/-/commit/ ||acb5fec4e9614324acdbc4cfce4 ||f387df65bd0f4 --- Comment #2 from Friedrich W. H. Kossebau --- Git commit acb5fec4e9614324acdbc4cfce4f387df65bd0f4 by Friedrich W. H. Kossebau. Committed on 16/04/2024 at 09:16. Pushed by kossebau into branch 'master'. Fix bad defaults for fonts, also for colors, shapes & style The refactoring started in 117f22918f594a046e8e541701a3cd863539cdd0 seems to have missed out follow-up work to ensure defaults are used where not set explicitly, as before. e47e5ae6d38c0e16b90714f1887b827a00d4 later only ensured explicit default font size values exist. Dot2QtConsts covers many of the defaults indirectly. For QFont though it was using "" as fort family name, which resulted in random fonts being picked by Qt. The implementation now avoids adding explicit attribute instances, but instead returns default values if not set, for less runtime memory usage. Related: bug 470523 FIXED-IN: 2.5.0 M +59 -11 src/part/graphelement.cpp M +7-28 src/part/graphelement.h https://invent.kde.org/graphics/kgraphviewer/-/commit/acb5fec4e9614324acdbc4cfce4f387df65bd0f4 -- You are receiving this mail because: You are watching all bug changes.
[kgraphviewer] [Bug 431414] Please open .dot file supplied in the command line
https://bugs.kde.org/show_bug.cgi?id=431414 Friedrich W. H. Kossebau changed: What|Removed |Added Latest Commit||https://invent.kde.org/grap ||hics/kgraphviewer/-/commit/ ||ab98d533a99928347cfc36513d3 ||30f827c8331dd Resolution|--- |FIXED Status|REPORTED|RESOLVED Version Fixed In||2.5.0 --- Comment #3 from Friedrich W. H. Kossebau --- Git commit ab98d533a99928347cfc36513d330f827c8331dd by Friedrich W. H. Kossebau. Committed on 16/04/2024 at 09:21. Pushed by kossebau into branch 'master'. Fix handling file cmdl arguments with relative path Related: bug 428456, bug 388518 FIXED-IN: 2.5.0 M +2-7src/main.cpp https://invent.kde.org/graphics/kgraphviewer/-/commit/ab98d533a99928347cfc36513d330f827c8331dd -- You are receiving this mail because: You are watching all bug changes.
[khelpcenter] [Bug 484176] The KDE help center does not show the documentation for KDE applications.
https://bugs.kde.org/show_bug.cgi?id=484176 --- Comment #12 from Friedrich W. H. Kossebau --- Git commit ef19ffe3c46b385301a7462fdf86d4c32b7b0f34 by Friedrich W. H. Kossebau. Committed on 08/04/2024 at 15:10. Pushed by nicolasfella into branch 'master'. Fix missing URL redirection implementation in web engine feeding from KIO Resulted in broken URLs generated from URL set for pages and the relative URLs used in the pages. M +9-0view.cpp https://invent.kde.org/system/khelpcenter/-/commit/ef19ffe3c46b385301a7462fdf86d4c32b7b0f34 -- You are receiving this mail because: You are watching all bug changes.
[khelpcenter] [Bug 483972] In text search is missing
https://bugs.kde.org/show_bug.cgi?id=483972 Friedrich W. H. Kossebau changed: What|Removed |Added CC||kosse...@kde.org --- Comment #2 from Friedrich W. H. Kossebau --- Merge request was started @ https://invent.kde.org/system/khelpcenter/-/merge_requests/47 -- You are receiving this mail because: You are watching all bug changes.
[khelpcenter] [Bug 483972] In text search is missing
https://bugs.kde.org/show_bug.cgi?id=483972 Friedrich W. H. Kossebau changed: What|Removed |Added CC||mig...@rozsas.eng.br --- Comment #1 from Friedrich W. H. Kossebau --- *** Bug 484545 has been marked as a duplicate of this bug. *** -- You are receiving this mail because: You are watching all bug changes.
[khelpcenter] [Bug 484545] khelpcenter lost the find feature
https://bugs.kde.org/show_bug.cgi?id=484545 Friedrich W. H. Kossebau changed: What|Removed |Added CC||kosse...@kde.org Resolution|--- |DUPLICATE Status|ASSIGNED|RESOLVED --- Comment #2 from Friedrich W. H. Kossebau --- *** This bug has been marked as a duplicate of bug 483972 *** -- You are receiving this mail because: You are watching all bug changes.
[khelpcenter] [Bug 484977] The KHelpcenter takes focus and it is impossible to close
https://bugs.kde.org/show_bug.cgi?id=484977 --- Comment #12 from Friedrich W. H. Kossebau --- (In reply to controloao from comment #11) > It looks like this is it. )) Sounds promising. Thanks for confirming :) -- You are receiving this mail because: You are watching all bug changes.
[khelpcenter] [Bug 484977] The KHelpcenter takes focus and it is impossible to close
https://bugs.kde.org/show_bug.cgi?id=484977 Friedrich W. H. Kossebau changed: What|Removed |Added Latest Commit|https://invent.kde.org/syst |https://invent.kde.org/syst |em/khelpcenter/-/commit/c12 |em/khelpcenter/-/commit/6ae |67491deaf1d29b48d28076e9c61 |4f85ae32a883dd550e8ad438b7f |498785d9b8 |c8ce1a36d8 --- Comment #10 from Friedrich W. H. Kossebau --- Git commit 6ae4f85ae32a883dd550e8ad438b7fc8ce1a36d8 by Friedrich W. H. Kossebau. Committed on 06/04/2024 at 13:51. Pushed by kossebau into branch 'release/24.02'. Unbreak endless invocation loop with "info" pages QWebEngine was handling info: urls as something to invoke an external handler for. Which... is khelpcenter itself. So instead of opening the page, there would be an endless invocation loop. FIXED-IN: 6.0.240202 (cherry picked from commit c1267491deaf1d29b48d28076e9c61498785d9b8) M +8-0view.cpp https://invent.kde.org/system/khelpcenter/-/commit/6ae4f85ae32a883dd550e8ad438b7fc8ce1a36d8 -- You are receiving this mail because: You are watching all bug changes.
[khelpcenter] [Bug 484977] The KHelpcenter takes focus and it is impossible to close
https://bugs.kde.org/show_bug.cgi?id=484977 Friedrich W. H. Kossebau changed: What|Removed |Added Status|REPORTED|RESOLVED Latest Commit||https://invent.kde.org/syst ||em/khelpcenter/-/commit/c12 ||67491deaf1d29b48d28076e9c61 ||498785d9b8 Version Fixed In||6.0.240202 Resolution|--- |FIXED --- Comment #9 from Friedrich W. H. Kossebau --- Git commit c1267491deaf1d29b48d28076e9c61498785d9b8 by Friedrich W. H. Kossebau. Committed on 06/04/2024 at 13:47. Pushed by kossebau into branch 'master'. Unbreak endless invocation loop with "info" pages QWebEngine was handling info: urls as something to invoke an external handler for. Which... is khelpcenter itself. So instead of opening the page, there would be an endless invocation loop. FIXED-IN: 6.0.240202 M +8-0view.cpp https://invent.kde.org/system/khelpcenter/-/commit/c1267491deaf1d29b48d28076e9c61498785d9b8 -- You are receiving this mail because: You are watching all bug changes.
[khelpcenter] [Bug 484977] The KHelpcenter takes focus and it is impossible to close
https://bugs.kde.org/show_bug.cgi?id=484977 --- Comment #6 from Friedrich W. H. Kossebau --- > there kill a bunch of processes generated by KHelpcenter And this killing gave the very crashlog you reported initially, right? In that case then the linked MR/patch should fix things also for what you experienced :) Will land soonish today, also to the current release branch, where the fix should the arrive with end users with upcoming KDE Gear 24.02.2, to be released on April 11, 2024. -- You are receiving this mail because: You are watching all bug changes.
[khelpcenter] [Bug 484977] The KHelpcenter takes focus and it is impossible to close
https://bugs.kde.org/show_bug.cgi?id=484977 Friedrich W. H. Kossebau changed: What|Removed |Added CC||kosse...@kde.org --- Comment #4 from Friedrich W. H. Kossebau --- Hi. How was the crash and thus the given crashlog actually triggered, by killing from the commandline (how exactly?) or waiting some time? Ran into a similar phenomenon of info pages not loading and khelpcenter getting unusable. Found this was due to eternal invocation loop, for which https://invent.kde.org/system/khelpcenter/-/merge_requests/43 has been done to solve this. So might be also the actual bug you experience here? The crash reported though might be some other cause still, thus above questions. -- You are receiving this mail because: You are watching all bug changes.
[kcachegrind] [Bug 465686] Typo in kcachegrind/org.kde.kcachegrind.appdata.xml
https://bugs.kde.org/show_bug.cgi?id=465686 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |FIXED Latest Commit||https://invent.kde.org/sdk/ ||kcachegrind/-/commit/164c5c ||e992d6dd5d8a82bab41bc469155 ||2f16104 Status|REPORTED|RESOLVED --- Comment #1 from Friedrich W. H. Kossebau --- Git commit 164c5ce992d6dd5d8a82bab41bc4691552f16104 by Friedrich W. H. Kossebau. Committed on 17/03/2024 at 13:16. Pushed by kossebau into branch 'master'. appdate: fix typo in "vi_z_ualisations" M +1-1kcachegrind/org.kde.kcachegrind.appdata.xml https://invent.kde.org/sdk/kcachegrind/-/commit/164c5ce992d6dd5d8a82bab41bc4691552f16104 -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 481311] KDirLister freezing when trying to parse a folder with more than 178 files
https://bugs.kde.org/show_bug.cgi?id=481311 Friedrich W. H. Kossebau changed: What|Removed |Added CC|kosse...@kde.org| --- Comment #4 from Friedrich W. H. Kossebau --- Sorry, never really worked on or reasoned in detail about this part of code, only touched it for simple changes, also no time currently. This needs someone active here, with good KIO overview (cmp. https://invent.kde.org/frameworks/kio/-/commits/master for candidates), so have to unsubscribe me again. -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 480289] [wayland] Okteta window starts too tall
https://bugs.kde.org/show_bug.cgi?id=480289 --- Comment #3 from Friedrich W. H. Kossebau --- (In reply to Eamonn Rea from comment #2) > Many apologies, it seems I did not search hard enough, I understand > duplicates of long-standing issues can be frustrating and I did not do > enough due diligence here. Thanks, all fine. And you gave details which I value in itself, deduplication was just some quick clicks for me :) My shame about not having fixed this in decades (I saw Krita meanwhile seems to have solved this, no longer using the hack) overweighs here 8( > Okteta is an invaluable piece of software to me Thanks for the kind words, nice to read once in a while that sharing the work improves things for others. > and this is not a huge issue, Darn, now you decreased the importance to fix it again :P Nah, as I might start to use Wayland myself more in the future, I should also have selfish reasons to tackle this soonish (hope dies last). -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 462703] Starting window size
https://bugs.kde.org/show_bug.cgi?id=462703 --- Comment #4 from Friedrich W. H. Kossebau --- (In reply to Friedrich W. H. Kossebau from comment #3) > https://bugreports.qt.io/browse/QTBUG-120191might perhaps be related here, > though only reported for Qt6. Ah, seems that was for X11, my bad (had wrong memories and too quickly copied the url from my browser history). I might have thought of something else, will have to wait until I rediscover that. -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 462703] Starting window size
https://bugs.kde.org/show_bug.cgi?id=462703 --- Comment #3 from Friedrich W. H. Kossebau --- https://bugreports.qt.io/browse/QTBUG-120191might perhaps be related here, though only reported for Qt6. Myself had recently only shortly looked at the issue, but for now only resorted to update the existing hack (which seems to only work on X11 by what I saw meanwhile) -> https://invent.kde.org/utilities/okteta/-/commit/debc06113cfa8ec4c245c8924855201dbbee3a46 No leading ideas myself where things are broken when it comes to Wayland, also given that quite some things there seem to be developing, so not yet invested there myself. Hoping myself to solve this with a new generic approach to tool views started locally as part of the Qt6 port, sadly currently still stuck in the port away from QtScripts (no longer in Qt6). -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 480289] [wayland] Okteta window starts too tall
https://bugs.kde.org/show_bug.cgi?id=480289 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |DUPLICATE Status|REPORTED|RESOLVED --- Comment #1 from Friedrich W. H. Kossebau --- Thanks for the report. This is sadly a long-standing issue, due to Qt's limited API to control QDockWidgets (used for the embedded tool views). There is some work-around (which I only recently updated), but that only seems to work on X11. *** This bug has been marked as a duplicate of bug 462703 *** -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 462703] Starting window size
https://bugs.kde.org/show_bug.cgi?id=462703 Friedrich W. H. Kossebau changed: What|Removed |Added CC||eamonn...@protonmail.com --- Comment #2 from Friedrich W. H. Kossebau --- *** Bug 480289 has been marked as a duplicate of this bug. *** -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 474838] ELF structure definition e_phoff array should be named "Program headers"
https://bugs.kde.org/show_bug.cgi?id=474838 --- Comment #2 from Friedrich W. H. Kossebau --- Thanks for the report :) -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 474838] ELF structure definition e_phoff array should be named "Program headers"
https://bugs.kde.org/show_bug.cgi?id=474838 Friedrich W. H. Kossebau changed: What|Removed |Added Version Fixed In||0.26.14 Status|REPORTED|RESOLVED Resolution|--- |FIXED Latest Commit||https://invent.kde.org/util ||ities/okteta/-/commit/418e9 ||4ec7774a12671e0742dbcf4f270 ||f26ad83f --- Comment #1 from Friedrich W. H. Kossebau --- Git commit 418e94ec7774a12671e0742dbcf4f270f26ad83f by Friedrich W. H. Kossebau. Committed on 03/12/2023 at 17:33. Pushed by kossebau into branch '0.26'. ELF structure definition: fix wrong e_phoff label FIXED-IN: 0.26.14 M +1-1 kasten/controllers/view/structures/examples/okteta/structures/elf/main.osd https://invent.kde.org/utilities/okteta/-/commit/418e94ec7774a12671e0742dbcf4f270f26ad83f -- You are receiving this mail because: You are watching all bug changes.
[libkcddb] [Bug 476984] Conflicting files in Qt5/Qt6 builds
https://bugs.kde.org/show_bug.cgi?id=476984 --- Comment #4 from Friedrich W. H. Kossebau --- (In reply to Luigi Toscano from comment #3) > I would argue the best solution would be to allow the coinstallability of > the Gear 23.08 Qt5 version and the future Qt6 versions. Which means that the Qt5 version is no longer supported though post 24.02. Which might be inconsistent with e.g. KF5, which might see more maintenance releases when needed by what I understood? -- You are receiving this mail because: You are watching all bug changes.
[libkcddb] [Bug 476984] Conflicting files in Qt5/Qt6 builds
https://bugs.kde.org/show_bug.cgi?id=476984 Friedrich W. H. Kossebau changed: What|Removed |Added CC||kosse...@kde.org --- Comment #2 from Friedrich W. H. Kossebau --- (In reply to Luigi Toscano from comment #1) > The library hasn't be ported to Qt6 - it shouldn't be dual-build, it should > be fully ported. It has been ported to Qt6. But it is not made Qt6-only, because there are consumers which lack people who port those to Qt6. > But we are just in the alpha phase of the "megarelease", so > porting is still possible. I would close this and reopen if it is not done > after the final date for porting (beginning of January) It would be good though to prepare for the worst case scenario (even more given the lack of develpoers caoable of porting complex applications like k3b, with all proper testing beyond "it builds"). As well as allow people to test now against the Qt6 variant of the library, so it needs to be installable & usable next to the Qt5 version. Even more when wanting to test ports for comparing new & old. -- You are receiving this mail because: You are watching all bug changes.
[kblocks] [Bug 472377] Crash on git master while holding down-arrow key
https://bugs.kde.org/show_bug.cgi?id=472377 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |UPSTREAM CC||kosse...@kde.org Status|REPORTED|NEEDSINFO --- Comment #2 from Friedrich W. H. Kossebau --- Saw this crash as well, but seems gone now relative to an update to Qt 6.5.3 on my system, no longer can reproduce. So perhaps was an upstream bug (the backtrace also would default me to that working theory). Can you still reproduce? If not, we could close this as UPSTREAM etc. -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 170925] Support UTF-8 in Okteta
https://bugs.kde.org/show_bug.cgi?id=170925 --- Comment #11 from Friedrich W. H. Kossebau --- *** Bug 473986 has been marked as a duplicate of this bug. *** -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 473986] UTF-8 encoding missing
https://bugs.kde.org/show_bug.cgi?id=473986 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |DUPLICATE Status|REPORTED|RESOLVED --- Comment #1 from Friedrich W. H. Kossebau --- Thanks for the request, I also share the need :) Just, the nature of UTF-8 being a variable width encoding, not a simple 8-Bit one, needs a fundamental change in the rendering logic and there always have been more pressing needs so far. One day... *** This bug has been marked as a duplicate of bug 170925 *** -- You are receiving this mail because: You are watching all bug changes.
[kdevelop] [Bug 473251] Terminal warnings and errors
https://bugs.kde.org/show_bug.cgi?id=473251 Friedrich W. H. Kossebau changed: What|Removed |Added CC|kosse...@kde.org| -- You are receiving this mail because: You are watching all bug changes.
[kdevelop] [Bug 473251] Terminal warnings and errors
https://bugs.kde.org/show_bug.cgi?id=473251 Friedrich W. H. Kossebau changed: What|Removed |Added Component|Analyzer: Clang-Tidy|Language Support: CPP ||(Clang-based) -- You are receiving this mail because: You are watching all bug changes.
[kate] [Bug 472374] File System Browser plugin search doesn't work
https://bugs.kde.org/show_bug.cgi?id=472374 Friedrich W. H. Kossebau changed: What|Removed |Added Product|frameworks-kio |kate Component|general |application Version|5.107.0 |23.04.3 Assignee|kio-bugs-n...@kde.org |kwrite-bugs-n...@kde.org --- Comment #7 from Friedrich W. H. Kossebau --- Moving back to Kate, MR upcoming... -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 472374] File System Browser plugin search doesn't work
https://bugs.kde.org/show_bug.cgi?id=472374 Friedrich W. H. Kossebau changed: What|Removed |Added CC||kosse...@kde.org --- Comment #5 from Friedrich W. H. Kossebau --- As commented on https://invent.kde.org/frameworks/kio/-/merge_requests/1314#note_728978 I would think that Kate actually relied on a behaviour change that was reverted in the mean time as it broke things for filter expressions as used also in file picker dialogs. The current (restored) logic in KIO's KCoreDirLister for name filters matches that of Qt for name filter properties. I.e. name filters here using the wildcard expression syntax. Which is non-partial. The UI pattern used in Kate's File System Browser plugin though is that of narrowing-while-you-type. Where the expression supported usually does not support even wildcards. Rather acts as if every letter is wrapped with wildcards ("sf" matching also "somefile"). So IMHO Kate code would need to convert the expression entered in the field into the syntax used for classic wild card file system expressions, as so far used with class properties of type "name filter". If you agree, please move the bug back to Kate :) -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 472751] Insert Mode overwrites instead of inserting
https://bugs.kde.org/show_bug.cgi?id=472751 --- Comment #4 from Friedrich W. H. Kossebau --- Okay :) > Is Overwrite by default the usual setting for editors? Might depend on the type of data that is edited :) Text? Images? Spreadsheets? Audio? Videos? In case of a hex editor when editing existing files many of those are in formats where data is bound to be at fixed offsets. So inserting is not that typical there, people rather often only twist existing bytes without changing the overall data layout. So the default was set to be Overwrite mode (unless starting with an empty file). Sadly there is no one-default-fits-everyone here? -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 472751] Insert Mode overwrites instead of inserting
https://bugs.kde.org/show_bug.cgi?id=472751 --- Comment #2 from Friedrich W. H. Kossebau --- Also, "Insert Mode" expression in the bug report title might hint that the existing "Overwrite Mode" feature had not yet been noticed, and the editor was in Overwrite mode for you when you experienced the unexpected? -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 472751] Insert Mode overwrites instead of inserting
https://bugs.kde.org/show_bug.cgi?id=472751 --- Comment #1 from Friedrich W. H. Kossebau --- > This behavior may be intentional, but it seems to be at odds with other > programs that use the term "insert". Curious., which other programs shaped your expectations here? -- You are receiving this mail because: You are watching all bug changes.
[kio-extras] [Bug 463614] stat() did not call finished()
https://bugs.kde.org/show_bug.cgi?id=463614 Friedrich W. H. Kossebau changed: What|Removed |Added Version Fixed In||23.12.0 Status|REPORTED|RESOLVED Latest Commit||https://invent.kde.org/netw ||ork/kio-extras/-/commit/943 ||9a2daae4a1758e20f9fd34aef27 ||6e23d33b40 Resolution|--- |FIXED --- Comment #1 from Friedrich W. H. Kossebau --- Git commit 9439a2daae4a1758e20f9fd34aef276e23d33b40 by Friedrich W. H. Kossebau. Committed on 25/07/2023 at 14:27. Pushed by kossebau into branch 'master'. Port kio_fish to WorkerBase As sideeffect closes FIXED-IN: 23.12.0 M +1-1fish/README M +269 -241 fish/fish.cpp M +43 -31 fish/fish.h M +1-1fish/tests/README M +2-2fish/tests/copytester.cpp M +1-1fish/tests/copytester.h M +1-1fish/tests/main.cpp https://invent.kde.org/network/kio-extras/-/commit/9439a2daae4a1758e20f9fd34aef276e23d33b40 -- You are receiving this mail because: You are watching all bug changes.
[konversation] [Bug 433667] Unable to select sound file
https://bugs.kde.org/show_bug.cgi?id=433667 Friedrich W. H. Kossebau changed: What|Removed |Added Latest Commit|https://invent.kde.org/netw |https://invent.kde.org/netw |ork/konversation/-/commit/f |ork/konversation/-/commit/7 |d8907a4b3d2c472b48e262896b3 |d2ff5e47fa2d154f264e6f2fe70 |18e387482322|28b7a09ad1d5 --- Comment #2 from Friedrich W. H. Kossebau --- Git commit 7d2ff5e47fa2d154f264e6f2fe7028b7a09ad1d5 by Friedrich W. H. Kossebau. Committed on 25/06/2023 at 17:51. Pushed by kossebau into branch 'master'. Highlights sounds file picker: fix setting file filter by MIME types "filter" property does not know anything about them, needs explicit method call. FIXED-IN: 1.9.23043 (cherry picked from commit fd8907a4b3d2c472b48e262896b318e387482322) M +1-0src/config/highlight_config.cpp M +0-3src/config/highlight_configui.ui https://invent.kde.org/network/konversation/-/commit/7d2ff5e47fa2d154f264e6f2fe7028b7a09ad1d5 -- You are receiving this mail because: You are watching all bug changes.
[konversation] [Bug 433667] Unable to select sound file
https://bugs.kde.org/show_bug.cgi?id=433667 Friedrich W. H. Kossebau changed: What|Removed |Added Status|REPORTED|RESOLVED Resolution|--- |FIXED Latest Commit||https://invent.kde.org/netw ||ork/konversation/-/commit/f ||d8907a4b3d2c472b48e262896b3 ||18e387482322 Version Fixed In||1.9.23043 --- Comment #1 from Friedrich W. H. Kossebau --- Git commit fd8907a4b3d2c472b48e262896b318e387482322 by Friedrich W. H. Kossebau. Committed on 25/06/2023 at 17:19. Pushed by kossebau into branch 'release/23.04'. Highlights sounds file picker: fix setting file filter by MIME types "filter" property does not know anything about them, needs explicit method call. FIXED-IN: 1.9.23043 M +1-0src/config/highlight_config.cpp M +0-3src/config/highlight_configui.ui https://invent.kde.org/network/konversation/-/commit/fd8907a4b3d2c472b48e262896b318e387482322 -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 369542] Anomaly: KUrlRequester::setFilter() still needs a KDE format filter string
https://bugs.kde.org/show_bug.cgi?id=369542 Friedrich W. H. Kossebau changed: What|Removed |Added Latest Commit|https://invent.kde.org/fram |https://invent.kde.org/fram |eworks/kio/-/commit/ee89c7a |eworks/kio/-/commit/8a1cc09 |9e673f5734cd4f5fb50ebadc691 |8cec9937bb11206f555a9bb978f |5050eb |baf981 --- Comment #10 from Friedrich W. H. Kossebau --- Git commit 8a1cc098cec9937bb11206f555a9bb978fbaf981 by Friedrich W. H. Kossebau. Committed on 23/06/2023 at 14:32. Pushed by kossebau into branch 'master'. Add KUrlRequester::nameFilters, deprecate KUrlRequester::filter For consistency with QFileDialog API and better naming given there are 2 kinds of filters supported, by name and by MIME type. M +109 -0autotests/kurlrequestertest.cpp M +1-1src/widgets/CMakeLists.txt M +62 -7src/widgets/kurlrequester.cpp M +39 -1src/widgets/kurlrequester.h https://invent.kde.org/frameworks/kio/-/commit/8a1cc098cec9937bb11206f555a9bb978fbaf981 -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 369542] Anomaly: KUrlRequester::setFilter() still needs a KDE format filter string
https://bugs.kde.org/show_bug.cgi?id=369542 Friedrich W. H. Kossebau changed: What|Removed |Added Latest Commit||https://invent.kde.org/fram ||eworks/kio/-/commit/ee89c7a ||9e673f5734cd4f5fb50ebadc691 ||5050eb Resolution|--- |FIXED Status|ASSIGNED|RESOLVED --- Comment #9 from Friedrich W. H. Kossebau --- Git commit ee89c7a9e673f5734cd4f5fb50ebadc6915050eb by Friedrich W. H. Kossebau. Committed on 23/06/2023 at 14:29. Pushed by kossebau into branch 'kf5'. Add KUrlRequester::nameFilters, deprecate KUrlRequester::filter For consistency with QFileDialog API and better naming given there are 2 kinds of filters supported, by name and by MIME type. M +109 -0autotests/kurlrequestertest.cpp M +1-1src/widgets/CMakeLists.txt M +62 -7src/widgets/kurlrequester.cpp M +35 -1src/widgets/kurlrequester.h https://invent.kde.org/frameworks/kio/-/commit/ee89c7a9e673f5734cd4f5fb50ebadc6915050eb -- You are receiving this mail because: You are watching all bug changes.
[ktorrent] [Bug 464350] Import Torrent "select torrent" filter shows up incorrectly
https://bugs.kde.org/show_bug.cgi?id=464350 --- Comment #5 from Friedrich W. H. Kossebau --- Git commit a7e865aa41acdb83dd625c68dd8f0d16e7f6dc0d by Friedrich W. H. Kossebau. Committed on 23/06/2023 at 14:26. Pushed by kossebau into branch 'master'. KUrlRequester::setFilter: note special filter syntax, allow unnamed filters The API dox referred to QFileDialog::setNameFilters for the syntax to use which is "Comment (file pattern)". The internal logic though expects the old KDE style syntax "file pattern|Comment". For more compatibility with Qt also allow name/comment-less filters by handling of non-existing "|" hit in the filter parsing. Related: bug 369542 (cherry picked from commit 6e945ed1dc01ffbaf30455341cff0ba945877255) M +6-4src/widgets/kurlrequester.cpp M +8-2src/widgets/kurlrequester.h https://invent.kde.org/frameworks/kio/-/commit/a7e865aa41acdb83dd625c68dd8f0d16e7f6dc0d -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 369542] Anomaly: KUrlRequester::setFilter() still needs a KDE format filter string
https://bugs.kde.org/show_bug.cgi?id=369542 --- Comment #8 from Friedrich W. H. Kossebau --- Git commit a7e865aa41acdb83dd625c68dd8f0d16e7f6dc0d by Friedrich W. H. Kossebau. Committed on 23/06/2023 at 14:26. Pushed by kossebau into branch 'master'. KUrlRequester::setFilter: note special filter syntax, allow unnamed filters The API dox referred to QFileDialog::setNameFilters for the syntax to use which is "Comment (file pattern)". The internal logic though expects the old KDE style syntax "file pattern|Comment". For more compatibility with Qt also allow name/comment-less filters by handling of non-existing "|" hit in the filter parsing. Related: bug 464350 (cherry picked from commit 6e945ed1dc01ffbaf30455341cff0ba945877255) M +6-4src/widgets/kurlrequester.cpp M +8-2src/widgets/kurlrequester.h https://invent.kde.org/frameworks/kio/-/commit/a7e865aa41acdb83dd625c68dd8f0d16e7f6dc0d -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 369542] Anomaly: KUrlRequester::setFilter() still needs a KDE format filter string
https://bugs.kde.org/show_bug.cgi?id=369542 --- Comment #7 from Friedrich W. H. Kossebau --- Git commit 6e945ed1dc01ffbaf30455341cff0ba945877255 by Friedrich W. H. Kossebau. Committed on 23/06/2023 at 14:04. Pushed by kossebau into branch 'kf5'. KUrlRequester::setFilter: note special filter syntax, allow unnamed filters The API dox referred to QFileDialog::setNameFilters for the syntax to use which is "Comment (file pattern)". The internal logic though expects the old KDE style syntax "file pattern|Comment". For more compatibility with Qt also allow name/comment-less filters by handling of non-existing "|" hit in the filter parsing. Related: bug 464350 M +6-4src/widgets/kurlrequester.cpp M +8-2src/widgets/kurlrequester.h https://invent.kde.org/frameworks/kio/-/commit/6e945ed1dc01ffbaf30455341cff0ba945877255 -- You are receiving this mail because: You are watching all bug changes.
[ktorrent] [Bug 464350] Import Torrent "select torrent" filter shows up incorrectly
https://bugs.kde.org/show_bug.cgi?id=464350 --- Comment #4 from Friedrich W. H. Kossebau --- Git commit 6e945ed1dc01ffbaf30455341cff0ba945877255 by Friedrich W. H. Kossebau. Committed on 23/06/2023 at 14:04. Pushed by kossebau into branch 'kf5'. KUrlRequester::setFilter: note special filter syntax, allow unnamed filters The API dox referred to QFileDialog::setNameFilters for the syntax to use which is "Comment (file pattern)". The internal logic though expects the old KDE style syntax "file pattern|Comment". For more compatibility with Qt also allow name/comment-less filters by handling of non-existing "|" hit in the filter parsing. Related: bug 369542 M +6-4src/widgets/kurlrequester.cpp M +8-2src/widgets/kurlrequester.h https://invent.kde.org/frameworks/kio/-/commit/6e945ed1dc01ffbaf30455341cff0ba945877255 -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 471079] okteta show empty HEX-dump area
https://bugs.kde.org/show_bug.cgi?id=471079 --- Comment #5 from Friedrich W. H. Kossebau --- (In reply to xintrea from comment #4) > I'm using LXDE 11 Not exactly sure which platform integration plugin Qt uses to read the system/platform configuration with that environment, seems people even can select different ones. To see which font actually is used, please see to have qtdiag installed. Then on the command line have the tool print its info into a textfile: $ qtdiag > qtdiag.txt Then open qtdiag.txt in a text editor and search for "Fixed font". What does it have for you? Here for comparison the lines I get --- 8< --- Fonts: General font : "Droid Sans" 10 Fixed font : "Droid Sans Mono" 10 Title font : "Droid Sans" 10 Smallest font: "Droid Sans" 8 --- 8< --- What font is listed there for you with "Fixed font"? To configure another fixed-width/monospace font on your system to try to find if that is the cause, with LXDE you might need to edit the fontconfig settings. No own experience with that, please get support from LXDE community on that. They might also know how to configure things for Qt-based applications and perhaps lend a hand with the fixedwidth/monospace font. Okteta itself simply relies on the Qt platform abstraction and tries to integrate by using whatever the system/environment has configured, as provided by the Qt platform integration plugins. -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 471079] okteta show empty HEX-dump area
https://bugs.kde.org/show_bug.cgi?id=471079 --- Comment #3 from Friedrich W. H. Kossebau --- Notes for self/developer: "QPainter::begin: Paint device returned engine == 0, type: 2" here the type 2 seems to be "Pixmap" from QInternal enum PaintDeviceFlags. Okteta uses a QPixmap for caching the cursor (gui/cursor.cpp). Forcing that QPixmap to be an 0x0 QPixmap results in the same log as reported: QPainter::begin: Paint device returned engine == 0, type: 2 QPainter::setPen: Painter not active QPainter::setBrush: Painter not active QPainter::setFont: Painter not active QPainter::setPen: Painter not active QPainter::end: Painter not active, aborted The size of the pixmap is calculated from args PixelX Width, PixelY Height, qreal devicePixelRatio, where the width & height are taken (indirectly from the QFontMetrics of the fixed size font. And devicePixelRatio is taken from the widget (Qt). The latter being 0 would fail elsewhere I assume, so a working theory here would be that either of or both the maxWidth() and height() have a value of 0, perhaps due to broken font metadata. So the font having some (unexpected) 0 values might explain the experienced display. Testing with other fonts for the system fixed size/width font might help to confirm that theory. -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 471079] okteta show empty HEX-dump area
https://bugs.kde.org/show_bug.cgi?id=471079 --- Comment #2 from Friedrich W. H. Kossebau --- One thing to try perhaps is another system font for fixed size fonts. Okteta uses the font as Qt reads from the platform plugin. In case of Plasma, that would be configured in the Systemsettings > Appearance > Fonts. What desktop environment are you using? -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 471079] okteta show empty HEX-dump area
https://bugs.kde.org/show_bug.cgi?id=471079 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |WAITINGFORINFO Status|REPORTED|NEEDSINFO --- Comment #1 from Friedrich W. H. Kossebau --- Thanks for the report. Though 0.26.5 is an outdated version, the latest is 0.26.11 just released last week, which should work with the same requirements as 0.26.5., the latest on the bug fix releases for 0.26. So please first try to update to 0.26.11 and check for the bugs you experienced again. "Paint device returned engine == 0" hints that something is broken in the Qt setup with your system, nothing that Okteta does control (and should control). That might also explain why no content is rendered. Never seen that log output before. Please compare other Qt-based programs. "kf.xmlgui: Index 10 is not within range (0 - 9 )" should be gone since 0.26.9, BTW :) -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 369542] Anomaly: KUrlRequester::setFilter() still needs a KDE format filter string
https://bugs.kde.org/show_bug.cgi?id=369542 Friedrich W. H. Kossebau changed: What|Removed |Added CC||kosse...@kde.org --- Comment #5 from Friedrich W. H. Kossebau --- https://invent.kde.org/frameworks/kio/-/merge_requests/1321 is a MR to approach this issue. -- You are receiving this mail because: You are watching all bug changes.
[ktorrent] [Bug 464350] Import Torrent "select torrent" filter shows up incorrectly
https://bugs.kde.org/show_bug.cgi?id=464350 Friedrich W. H. Kossebau changed: What|Removed |Added Status|ASSIGNED|RESOLVED Latest Commit||https://invent.kde.org/netw ||ork/ktorrent/-/commit/c7caa ||4735d4f8d024029e18ebdc5cdb3 ||cbd87348 Resolution|--- |FIXED Version Fixed In||23.04.3 --- Comment #3 from Friedrich W. H. Kossebau --- Git commit c7caa4735d4f8d024029e18ebdc5cdb3cbd87348 by Friedrich W. H. Kossebau. Committed on 14/06/2023 at 16:14. Pushed by kossebau into branch 'release/23.04'. Import Torrent dialog: fix file filter KUrlRequester::setFilter expected name filters separated by \n, not ;;, other than QFileDialog. And it also has broken support for the Qt syntax "Comment (pattern)", only works with old-style kde "pattern|Comment". FIXED-IN: 23.04.3 M +1-1ktorrent/dialogs/importdialog.cpp M +7-1libktcore/interfaces/functions.cpp M +5-2libktcore/interfaces/functions.h https://invent.kde.org/network/ktorrent/-/commit/c7caa4735d4f8d024029e18ebdc5cdb3cbd87348 -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 441925] Okteta search behavior keeps selecting "selected bytes" when it's not wanted--doesn't remember settings
https://bugs.kde.org/show_bug.cgi?id=441925 Friedrich W. H. Kossebau changed: What|Removed |Added Status|REPORTED|RESOLVED Version Fixed In||0.26.11 Resolution|--- |FIXED Latest Commit||https://invent.kde.org/util ||ities/okteta/-/commit/48e3f ||9820009abd7198653590a3787e2 ||1c45907c --- Comment #1 from Friedrich W. H. Kossebau --- Git commit 48e3f9820009abd7198653590a3787e21c45907c by Friedrich W. H. Kossebau. Committed on 02/06/2023 at 19:29. Pushed by kossebau into branch '0.26'. Search/replace tools: disable automagic for "In selection" checkbox Conditions used not the silver bullet behaviour after all. FIXED-IN: 0.26.11 M +3-1kasten/controllers/view/replace/replacedialog.cpp M +3-1kasten/controllers/view/search/searchdialog.cpp https://invent.kde.org/utilities/okteta/-/commit/48e3f9820009abd7198653590a3787e21c45907c -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 467541] 0.26.10: Test suite is failing
https://bugs.kde.org/show_bug.cgi?id=467541 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |FIXED Status|REOPENED|RESOLVED Latest Commit||https://invent.kde.org/util ||ities/okteta/commit/f8ea84b ||e37fa5440cb3a2c61858efe520f ||957cc2 Version Fixed In||0.26.11 --- Comment #7 from Friedrich W. H. Kossebau --- Git commit f8ea84be37fa5440cb3a2c61858efe520f957cc2 by Friedrich W. H. Kossebau. Committed on 04/05/2023 at 14:04. Pushed by kossebau into branch '0.26'. Fix accidentally shared unit test working directories Runs risk in parallel test runs to have the dirs cleaned up while being used by a parallel test. FIXED-IN: 0.26.11 M +1-1kasten/core/io/tests/bytearrayrawfilesynchronizerfactorytest.cpp M +1-1 libs/kasten/core/tests/testdocumentfilesynchronizerfactorytest.cpp https://invent.kde.org/utilities/okteta/commit/f8ea84be37fa5440cb3a2c61858efe520f957cc2 -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 467541] 0.26.10: Test suite is failing
https://bugs.kde.org/show_bug.cgi?id=467541 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |WORKSFORME Status|REPORTED|NEEDSINFO --- Comment #1 from Friedrich W. H. Kossebau --- Thanks for the report. Cannot reproduce though, and has been passing for me locally (openSUSE TW) and on KDE CI all the time. See e.g. https://invent.kde.org/utilities/okteta/-/pipelines/302655/test_report for the last pipeline on the tag itself. What Qt version are you using? The following looks suspicious: QWARN : TestDocumentFileSynchronizerTest::testLoadReloadFile() QIODevice::write (QFile, "/home/tkloczko//.kde-unit-test//testdocumentfile1synchronizertest//test1.data"): device not open In case the duplicated path separators are an issue here (bad in any case), I pushed a commit to current 0.26 branch here: https://invent.kde.org/utilities/okteta/-/commit/67b3bb25b9fe8ce6794fa8038347a5adb4466478 Could you try again with that commit? The other question would be: is home/tkloczko/.kde-unit-test/ writable during your test runs? -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 466887] Journal Flood: "QObject::connect: No such slot DesktopProtocol::_k_slotRedirection(KIO::Job *, QUrl)"
https://bugs.kde.org/show_bug.cgi?id=466887 Friedrich W. H. Kossebau changed: What|Removed |Added CC|kosse...@kde.org| --- Comment #2 from Friedrich W. H. Kossebau --- Please do not subscribe me to bug reports without explicit reasons, thanks. -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 336936] Okteta's window is too tall in windowed mode
https://bugs.kde.org/show_bug.cgi?id=336936 Friedrich W. H. Kossebau changed: What|Removed |Added Status|RESOLVED|REPORTED Resolution|WORKSFORME |--- -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 465770] Remember Encoding or Allow a Default to be Set or Accept Command Line Value
https://bugs.kde.org/show_bug.cgi?id=465770 Friedrich W. H. Kossebau changed: What|Removed |Added Ever confirmed|0 |1 Status|REPORTED|CONFIRMED --- Comment #3 from Friedrich W. H. Kossebau --- Made a separate reminder to myself about the handbook by the new bug 465780, as I see how that might have helped. Now to your proposed solutions for your problem: #1 (remembering the last used setting) is something worth to considering next to the view profile system, for those who have not yet discovered it or prefer always the last manual settings over some configured default. After all the tools with parameters in recent versions also remember the last parameters manually set (do not remember though why the search/replace dialogs did not get that also). #2 (some central config dialog) is something I personally try to avoid, instead try to have any config as close as possible to the actiual UI. Even more as Okteta is made of elements which are designed to be embedded into other software, where the configuration also needs to be available (cmp. https://frinring.wordpress.com/2010/02/16/okteta-going-for-kdevelop-and-kate/ still valid a decade later :) ). Having separately another way via a central settings dialog for mass configuration makes sense to me also, but had not yet got to that, due to complexity when embedding/integrating into other applications. #3 (setting view properties as additional args on the commandline) might be interesting. Comes with the challenge that Okteta supports multiple documents (and even multiple views on the same document), so perhaps someone might want to have more fine control which document to show with which view settings. For this I first would like to gather a good set of use cases that need this very solution for other reasons to properly design the support. So after this first round of thinking, would keep this report open with the focus on the feature request for remembering the last manual view settings. Might take some time before I get to it, as currently working on other things. Happy to get more comments on the matter though to consider then at the time of going to implementation. -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 465780] New: Handbook: describe more details about View Profiles
https://bugs.kde.org/show_bug.cgi?id=465780 Bug ID: 465780 Summary: Handbook: describe more details about View Profiles Classification: Applications Product: okteta Version: unspecified Platform: Other OS: Linux Status: REPORTED Severity: wishlist Priority: NOR Component: general Assignee: kosse...@kde.org Reporter: kosse...@kde.org Target Milestone: --- The scope of View Profiles is not yet exactly specified, people might make up their own ideas instead. Also grepping for certain keywords (e.g. charset) would not move this feature into their attention (cmp. bug 465770) -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 465770] Remember Encoding or Allow a Default to be Set or Accept Command Line Value
https://bugs.kde.org/show_bug.cgi?id=465770 --- Comment #1 from Friedrich W. H. Kossebau --- Thanks for the feedback. Curious for now, had you discovered the feature "View Profiles" already and tried to solve your need using that? This feature includes selecting a certain charset. And one can set a certain View Profile as default in the View Profile managing dialog (Settings->Manage View Profiles...), to meet the most typical usecases. The proposed alternative solutions are something worth to consider as well, but before discussing and implementing support for that (and then needing to support that in the future because people rely on it) I prefer if people can get things easily done with the existing features :) -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 332770] Loading files over a certain size leads to undefined behavior
https://bugs.kde.org/show_bug.cgi?id=332770 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|WAITINGFORINFO |--- Status|NEEDSINFO |REPORTED -- You are receiving this mail because: You are watching all bug changes.
[kleopatra] [Bug 464229] mainwindow icon bug
https://bugs.kde.org/show_bug.cgi?id=464229 Friedrich W. H. Kossebau changed: What|Removed |Added CC||kosse...@kde.org --- Comment #1 from Friedrich W. H. Kossebau --- As commented on irc, I suspect the cause is that the window manager used here is an X11 one which only fetches the window icons from the X properties directly, does not map the _KDE_NET_WM_DESKTOP_FILE to a desktop file to fetch metadata like the app icon. For those one still needs to set the window icon directly, e.g. by adding app.setWindowIcon(QIcon::fromTheme(QStringLiteral("kleopatra"), app.windowIcon())); to https://invent.kde.org/pim/kleopatra/-/blob/master/src/main.cpp#L128 While at it, all the app metadata setting is better moved to before the "// Initialize GpgME" section, otherwise e.g. the text in the warning message will not be translated properly I expect as well as the window again missing needed metadata to be nicely shown as Kleopatra window. -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 462845] The default button changed to Cancel at permanent delete
https://bugs.kde.org/show_bug.cgi?id=462845 Friedrich W. H. Kossebau changed: What|Removed |Added CC|kosse...@kde.org| -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 462845] The default button changed to Cancel at permanent delete
https://bugs.kde.org/show_bug.cgi?id=462845 Friedrich W. H. Kossebau changed: What|Removed |Added CC||kosse...@kde.org --- Comment #21 from Friedrich W. H. Kossebau --- (In reply to Friedrich W. H. Kossebau from comment #5) > Related commits changing the behaviour: > * https://invent.kde.org/system/dolphin/-/merge_requests/462 > * https://invent.kde.org/frameworks/kio/-/merge_requests/1005 The latter was a wrong read, actually on the KIO side things were changed in * https://invent.kde.org/frameworks/kio/-/merge_requests/818 -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 462845] The default button changed to Cancel at permanent delete
https://bugs.kde.org/show_bug.cgi?id=462845 Friedrich W. H. Kossebau changed: What|Removed |Added CC|kosse...@kde.org| -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 462845] The default button changed to Cancel at permanent delete
https://bugs.kde.org/show_bug.cgi?id=462845 Friedrich W. H. Kossebau changed: What|Removed |Added Status|REPORTED|CONFIRMED Ever confirmed|0 |1 CC||kosse...@kde.org --- Comment #5 from Friedrich W. H. Kossebau --- Related commits changing the behaviour: * https://invent.kde.org/system/dolphin/-/merge_requests/462 * https://invent.kde.org/frameworks/kio/-/merge_requests/1005 -- You are receiving this mail because: You are watching all bug changes.
[plasmashell] [Bug 461793] Files with custom positions lose their position after rename and Plasma restart
https://bugs.kde.org/show_bug.cgi?id=461793 --- Comment #1 from Friedrich W. H. Kossebau --- Fixes things for me: https://invent.kde.org/plasma/plasma-workspace/-/merge_requests/2320 -- You are receiving this mail because: You are watching all bug changes.
[plasmashell] [Bug 461793] New: Files with custom positions lose their position after rename and Plasma restart
https://bugs.kde.org/show_bug.cgi?id=461793 Bug ID: 461793 Summary: Files with custom positions lose their position after rename and Plasma restart Classification: Plasma Product: plasmashell Version: 5.26.3 Platform: Other OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: Folder Assignee: plasma-b...@kde.org Reporter: kosse...@kde.org CC: h...@kde.org Target Milestone: 1.0 SUMMARY When renaming a file in a folder containment which has a custom position, it will be repositioned after Plasma restart to the first free position in the topleft ordering. STEPS TO REPRODUCE 1. Move a file to a random position, e.g. bottom-right. 2. Rename the file, e.g. via context menu. 3. Restart Plasma OBSERVED RESULT File is moved to first free position in the normal ordering. EXPECTED RESULT File keeps it position. -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kxmlgui] [Bug 461183] kf.xmlgui: Index 10 is not within range (0 - 9 ) when switching between tabs
https://bugs.kde.org/show_bug.cgi?id=461183 Friedrich W. H. Kossebau changed: What|Removed |Added Product|okteta |frameworks-kxmlgui Component|general |general Version|0.26.6 |5.99.0 Assignee|kosse...@kde.org|kdelibs-b...@kde.org --- Comment #1 from Friedrich W. H. Kossebau --- Can be seen in different kxmlgui apps, not just Okteta. -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 441802] Can't open file via menu / Ctrl+O but can via drag-and-drop / recent files
https://bugs.kde.org/show_bug.cgi?id=441802 --- Comment #16 from Friedrich W. H. Kossebau --- *** Bug 461185 has been marked as a duplicate of this bug. *** -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 461185] Cannot open file using file dialog
https://bugs.kde.org/show_bug.cgi?id=461185 Friedrich W. H. Kossebau changed: What|Removed |Added Status|REPORTED|RESOLVED Resolution|--- |DUPLICATE --- Comment #1 from Friedrich W. H. Kossebau --- Reported before, needs fixing upstream. Please see bug 441802 and there the link to https://bugreports.qt.io/browse/QTBUG-96157 *** This bug has been marked as a duplicate of bug 441802 *** -- You are receiving this mail because: You are watching all bug changes.
[Breeze] [Bug 458214] Icon request: KSokoban game app icon
https://bugs.kde.org/show_bug.cgi?id=458214 Friedrich W. H. Kossebau changed: What|Removed |Added Status|CONFIRMED |REPORTED Ever confirmed|1 |0 --- Comment #3 from Friedrich W. H. Kossebau --- Time is flying :) Haivng slept more nights, my first thought given in the last comment still hold. What are your thoughts? :) -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kimageformats] [Bug 460105] libraw is not found on macos even though installed - build fails
https://bugs.kde.org/show_bug.cgi?id=460105 Friedrich W. H. Kossebau changed: What|Removed |Added Latest Commit||https://invent.kde.org/fram ||eworks/kimageformats/commit ||/20f74ce5e69f731ba114db4103 ||11b0989dd88b17 Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #2 from Friedrich W. H. Kossebau --- Git commit 20f74ce5e69f731ba114db410311b0989dd88b17 by Friedrich W. H. Kossebau. Committed on 08/10/2022 at 00:15. Pushed by kossebau into branch 'master'. FindLibRaw: fix include dir, should not contain prefix libraw/ See also the examples at https://www.libraw.org/docs/API-CXX.html M +4-5cmake/find-modules/FindLibRaw.cmake M +0-4src/imageformats/raw.cpp https://invent.kde.org/frameworks/kimageformats/commit/20f74ce5e69f731ba114db410311b0989dd88b17 -- You are receiving this mail because: You are watching all bug changes.
[Breeze] [Bug 458214] Icon request: KSokoban game app icon
https://bugs.kde.org/show_bug.cgi?id=458214 --- Comment #2 from Friedrich W. H. Kossebau --- (In reply to Manuel de la Fuente from comment #1) > Created attachment 151929 [details] > List of proposals for the icon > > I've attached a few icons based on the original game assets, with a take on > the current one too (though that one is a bit more rough). Let me know your > thoughts! Yay, happy to see things happening :) First impression thoughts following First though a quick hint that KSokoban might see a rename, due to trademark issues (or respect to those still around who have done and distribute the original game). Yet to be decided & executed. New name should still be around the "dude reshuffling things in warehouse" theme ideally. Interesting idea to switch to using the goal asset as app icon theme. Good: * might reflect more to the game idea to get things to given new locations Bad: * breaks tradition * small shape expansion might look out of proportion * object type hard to identify for those who do not know KSokoban, might result in mental mismapping/confusion Using a green colored gem misses the direct reference for me, as the current game itself does not have such an object appearance. Using a red gem shaped similar to the one in the game actually picks up the tradition of the initial icon, in case you have not yet researched yourself :) (see https://invent.kde.org/games/ksokoban/-/commit/3f1a2d11336c8e175579a8e6019b7074dc5fe3dc ). with the version slightly turned to the viewer matching the original in that. The version matching the current icon looks good. Though as with the current icon, my mind always sees also a boat steering upon me, with a strange deck ;) It surely is unique and by that serves a purpose, but also feels strange. So out of the selection presented, after a first look I would tend to the shape of the two red gems top-left. Though they might perhaps get a tad more 3D effect, like more contrast in whatever way to the sides on the left & the right? The green variants have that better, at least on my display. -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 182577] Okteta unable to load large files
https://bugs.kde.org/show_bug.cgi?id=182577 Friedrich W. H. Kossebau changed: What|Removed |Added CC||hpf...@psnarf.org --- Comment #11 from Friedrich W. H. Kossebau --- *** Bug 458712 has been marked as a duplicate of this bug. *** -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 458712] A hex editor with a 2Gb file limit is useless
https://bugs.kde.org/show_bug.cgi?id=458712 Friedrich W. H. Kossebau changed: What|Removed |Added Status|REPORTED|RESOLVED Resolution|--- |DUPLICATE --- Comment #1 from Friedrich W. H. Kossebau --- Hi Henry Pfeil, I am sorry Okteta does not serve your personal needs, and you only found out after trying it that it does not meet your criteria for what is "useful". Yet, for those who just need to peek and twist small-sized files even on 64-bit address systems (like myself) it works good enough. And yes, I would have liked to have time to implement support for endlessly sized data since the very start two decades ago (should be documented in code even), but always had too many other interesting things to do in life (the answer to: " why not a 64-bit version"). No reason though to no longer share the fruits of my former work with others under the GPL, which explicitly states that such licensed program is provided > "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, > BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS > FOR A PARTICULAR PURPOSE Would it have helped you if the description of the program (as done in the app metadata) would explicitly hint that there is a limit in supported file sizes? Might consider adding that there, to save people time and frustrations (but only when people actually read that and not just go ahead based on their assumptions :) ). For your needs, you might consider also giving wxhexeditor or ImHex a try (not tested myself). *** This bug has been marked as a duplicate of bug 182577 *** -- You are receiving this mail because: You are watching all bug changes.
[Breeze] [Bug 458214] New: Icon request: KSokoban game app icon
https://bugs.kde.org/show_bug.cgi?id=458214 Bug ID: 458214 Summary: Icon request: KSokoban game app icon Product: Breeze Version: unspecified Platform: Other OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: Icons Assignee: visual-des...@kde.org Reporter: kosse...@kde.org CC: kain...@gmail.com, m...@nueljl.in Target Milestone: --- The game app KSokoban (https://apps.kde.org/ksokoban) currently has no Breeze-style icon. But ideally had one :) Even more as it will be back in release mode soon, after over a decade of sleep (already builds and works also with Qt6/preKF6, so future-prepared). The current app-internal icon was added (14 years ago) by: https://invent.kde.org/games/ksokoban/-/commit/761405f5b5d46b7a61d15d2b805e4880e919880f No icons known in other current themes, also not Oxygen. Icon considerations: * might relate to game logic (see https://en.wikipedia.org/wiki/Sokoban) * pick up gem identity of current app icon/logo, referring to the app's unique choice of a gems instead of boxes to push around (IIRC somewhere the original author wrote he simply found gems better looking than boxes) The current visual game material is created by rendering with povray into pixels images at build time (cached PNG copies in the repository though). Long-term plans are to add runtime theming capability to the game itself, to allow e.g. SVG-based designs. But the current look should stay as classic variant, for tradition :) PS: when trying the app to get some ideas about its spirit, one might consider help from https://ksokoban.online/ for solving the puzzles (search for the level names, many if not all are present there). Yes, I searched for that and learned about it for a reason ;) -- You are receiving this mail because: You are watching all bug changes.
[ksudoku] [Bug 457951] ksudoku commit 6e9d941ce with an existing theme config results in a blank play-space and qpainter engine errors
https://bugs.kde.org/show_bug.cgi?id=457951 Friedrich W. H. Kossebau changed: What|Removed |Added Latest Commit||https://invent.kde.org/game ||s/libkdegames/commit/91e246 ||f2b17c899c1d5a9940d7b6f7d98 ||93e2319 Resolution|--- |FIXED Status|ASSIGNED|RESOLVED --- Comment #16 from Friedrich W. H. Kossebau --- Git commit 91e246f2b17c899c1d5a9940d7b6f7d9893e2319 by Friedrich W. H. Kossebau. Committed on 16/08/2022 at 15:10. Pushed by kossebau into branch 'release/22.08'. KgThemeProvider: fix generating broken full path theme id due to fs symlinks relativeToApplications() was using QFileInfo(file).canonicalFilePath() to compare against non-canonicalized directories returned by QStandardPaths, thus missing out in case the file was in a path with symlinks, resulting in ids with full absolute path. This patch fixes that by calculating the id from the known one subdir name and the desktop file name directly. M +1-13 src/kgthemeprovider.cpp https://invent.kde.org/games/libkdegames/commit/91e246f2b17c899c1d5a9940d7b6f7d9893e2319 -- You are receiving this mail because: You are watching all bug changes.
[ksudoku] [Bug 457951] ksudoku commit 6e9d941ce with an existing theme config results in a blank play-space and qpainter engine errors
https://bugs.kde.org/show_bug.cgi?id=457951 --- Comment #15 from Friedrich W. H. Kossebau --- Git commit fc9adcdd5eb843bc3024aa71787ea5de5950a14d by Friedrich W. H. Kossebau. Committed on 17/08/2022 at 10:30. Pushed by kossebau into branch 'release/22.08'. Fix handling of invalid theme ids, fall back to default instead M +1-1src/gui/views/renderer.cpp https://invent.kde.org/games/ksudoku/commit/fc9adcdd5eb843bc3024aa71787ea5de5950a14d -- You are receiving this mail because: You are watching all bug changes.
[ksudoku] [Bug 457951] ksudoku commit 6e9d941ce with an existing theme config results in a blank play-space and qpainter engine errors
https://bugs.kde.org/show_bug.cgi?id=457951 --- Comment #14 from Friedrich W. H. Kossebau --- Thanks again for testing. (In reply to Duncan from comment #10) > (In reply to Friedrich W. H. Kossebau from comment #9) > > https://invent.kde.org/games/libkdegames/-/merge_requests/20.diff > > Good and bad news: > Good: The relative style ksudokurc theme entry works with that patch. Okay :) > Bad: The full-path style does not. > > If there hasn't been a release with the now-existing code that creates the > full-path entry I imagine the bad news can be ignored (unless we want to > give people hand-editing their ksudokurc files the choice of doing full > paths, too). No release of ksudoko using the new code yet, though sadly 22.08.0 release is currently prepared for tomorrow release and would be the first version of ksudoko with the new code path. libkdegames has seen a quite some releases with the id screwing logic (was added in 2014). I would agree that the most simple thing would be to just fallback to the default theme if the config holds a theme id that cannot be resolved. Most places also already do that, so those games can at least be used for gaming. Though ksudoku needs a fix-up for handling that situation, blame on me here, no idea why I failed to consider bad config entries. Local patch already prepared and tested, landing next. Will see to have the prepared 22.08.0 tarballs updated in time for tomorrow now. -- You are receiving this mail because: You are watching all bug changes.
[ksudoku] [Bug 457951] ksudoku commit 6e9d941ce with an existing theme config results in a blank play-space and qpainter engine errors
https://bugs.kde.org/show_bug.cgi?id=457951 --- Comment #9 from Friedrich W. H. Kossebau --- Thanks for getting and showing that log. Okay, that should back up the working theory about the cause :) If you like to help more, please give the meanwhile created bug fix proposal as reported before by the bot a test. It should also apply to current master (I assume). You can download the patch itself by blue "Code" button (choose "Plain diff" in the popup menu) or using this link directly: https://invent.kde.org/games/libkdegames/-/merge_requests/20.diff -- You are receiving this mail because: You are watching all bug changes.
[ksudoku] [Bug 457951] ksudoku commit 6e9d941ce with an existing theme config results in a blank play-space and qpainter engine errors
https://bugs.kde.org/show_bug.cgi?id=457951 --- Comment #6 from Friedrich W. H. Kossebau --- Please try the attached patch for libkdegames: apply, build, install. then start ksudoku on the commandline. Interesting will be all the log output starting with "THEMELOADING", should be something like (in my case having double installation in system and user self-built location): --- 8< --- THEMELOADING considering locations ([lots of paths], "/usr/share/ksudoku") THEMELOADING considering desktop file "/home/some/path/share/ksudoku/themes/abstraction.desktop" THEMELOADING canonical path "/home/some/path/share/ksudoku/themes/abstraction.desktop" THEMELOADING id "themes/abstraction.desktop" THEMELOADING considering desktop file "/home/some/path/share/ksudoku/themes/default.desktop" THEMELOADING canonical path "/home/some/path/share/ksudoku/themes/default.desktop" THEMELOADING id "themes/default.desktop" THEMELOADING considering desktop file "/home/some/path/share/ksudoku/themes/ksudoku_scrible.desktop" THEMELOADING canonical path "/home/some/path/share/ksudoku/themes/ksudoku_scrible.desktop" THEMELOADING id "themes/ksudoku_scrible.desktop" THEMELOADING considering desktop file "/usr/share/ksudoku/themes/abstraction.desktop" THEMELOADING considering desktop file "/usr/share/ksudoku/themes/default.desktop" THEMELOADING considering desktop file "/usr/share/ksudoku/themes/ksudoku_scrible.desktop" --- 8< --- -- You are receiving this mail because: You are watching all bug changes.
[ksudoku] [Bug 457951] ksudoku commit 6e9d941ce with an existing theme config results in a blank play-space and qpainter engine errors
https://bugs.kde.org/show_bug.cgi?id=457951 --- Comment #5 from Friedrich W. H. Kossebau --- Created attachment 151355 --> https://bugs.kde.org/attachment.cgi?id=151355=edit Debug logging patch for libkdegames master (c5848c0d673cebc67eb280d54081ca69efeb28b0) -- You are receiving this mail because: You are watching all bug changes.
[ksudoku] [Bug 457951] ksudoku commit 6e9d941ce with an existing theme config results in a blank play-space and qpainter engine errors
https://bugs.kde.org/show_bug.cgi?id=457951 --- Comment #3 from Friedrich W. H. Kossebau --- >From a first look for potential culprit code, it seems the issue is how the theme id is generated in the libkdegames classes used in the new code (KgThemeProvider): `KgThemeProvider::rediscoverThemes()` goes over all `QStandardPaths::locateAll(QStandardPaths::AppDataLocation, "themes", QStandardPaths::LocateDirectory)` and collects the full path to the desktop files. Then it generates the theme id by first turning the desktop file path to a canonical version by `QFileInfo(file).canonicalFilePath()` and going once more over all `QStandardPaths::standardLocations(QStandardPaths::AppDataLocation)` and, if any of the latter path is a prefix to the canonical desktop file path, subtracts that prefix from the path to get the id. ((oh dear, how random people porting convoluted that logic :) )). See https://invent.kde.org/games/libkdegames/-/blob/master/src/kgthemeprovider.cpp#L185 and https://invent.kde.org/games/libkdegames/-/blob/master/src/kgthemeprovider.cpp#L174 My theory is that the logic of that code seems to miss out some constellation on your system, where QStandardPaths calls deliver some paths which would be turned to something else by the idea of QFileInfo about what the canonical version is. Are you capable to add some fprint debug lines to libkdegames yourself to get insight into the respective values used there? -- You are receiving this mail because: You are watching all bug changes.
[ksudoku] [Bug 457951] ksudoku commit 6e9d941ce with an existing theme config results in a blank play-space and qpainter engine errors
https://bugs.kde.org/show_bug.cgi?id=457951 --- Comment #2 from Friedrich W. H. Kossebau --- Hi, thanks for reporting the issue. Cannot remember I ran into such errors (though been some time) when I tested the change, and trying locally right now I also only get relative path with the new code when changing themes, like theme=themes/abstraction.desktop Will have a look in the code to see what condition might result in a full path later today. -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 314798] There is no obvious way to copy (for pasting) data from defined structure
https://bugs.kde.org/show_bug.cgi?id=314798 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |FIXED Version Fixed In||0.26.9 Latest Commit||https://invent.kde.org/util ||ities/okteta/commit/92a5d54 ||d3034f375f75d0f78be51ec6429 ||575f49 Status|REPORTED|RESOLVED --- Comment #2 from Friedrich W. H. Kossebau --- Git commit 92a5d54d3034f375f75d0f78be51ec6429575f49 by Friedrich W. H. Kossebau. Committed on 03/06/2022 at 16:27. Pushed by kossebau into branch '0.26'. Structures tool: add context menu to copy data & display string to clipboard FIXED-IN: 0.26.9 M +23 -8kasten/controllers/view/structures/structurestool.cpp M +4-1kasten/controllers/view/structures/structurestool.hpp M +48 -0kasten/controllers/view/structures/structureview.cpp M +2-0kasten/controllers/view/structures/structureview.hpp https://invent.kde.org/utilities/okteta/commit/92a5d54d3034f375f75d0f78be51ec6429575f49 -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 337652] Edit flag and enum fields using literals from a structure definition
https://bugs.kde.org/show_bug.cgi?id=337652 --- Comment #1 from Friedrich W. H. Kossebau --- Git commit 29d529e7cc4be8ca5370bef20d8971a9e79ac941 by Friedrich W. H. Kossebau. Committed on 31/05/2022 at 14:13. Pushed by kossebau into branch '0.26'. Structures tool: use combobox editor for enums with defined values M +21 -0 kasten/controllers/view/structures/datatypes/primitive/basicprimitivedatainformation.hpp M +16 -0 kasten/controllers/view/structures/datatypes/primitive/bitfield/boolbitfielddatainformation.cpp M +4-0 kasten/controllers/view/structures/datatypes/primitive/bitfield/boolbitfielddatainformation.hpp M +16 -0 kasten/controllers/view/structures/datatypes/primitive/bitfield/signedbitfielddatainformation.cpp M +4-0 kasten/controllers/view/structures/datatypes/primitive/bitfield/signedbitfielddatainformation.hpp M +16 -0 kasten/controllers/view/structures/datatypes/primitive/bitfield/unsignedbitfielddatainformation.cpp M +4-0 kasten/controllers/view/structures/datatypes/primitive/bitfield/unsignedbitfielddatainformation.hpp M +7-0 kasten/controllers/view/structures/datatypes/primitive/booldatainformation.hpp M +6-0 kasten/controllers/view/structures/datatypes/primitive/chardatainformation.hpp M +6-0 kasten/controllers/view/structures/datatypes/primitive/doubledatainformation.hpp M +40 -0 kasten/controllers/view/structures/datatypes/primitive/enumdatainformation.cpp M +4-0 kasten/controllers/view/structures/datatypes/primitive/enumdatainformation.hpp M +6-0 kasten/controllers/view/structures/datatypes/primitive/floatdatainformation.hpp M +15 -0 kasten/controllers/view/structures/datatypes/primitive/primitivedatainformation.cpp M +9-0 kasten/controllers/view/structures/datatypes/primitive/primitivedatainformation.hpp M +8-0 kasten/controllers/view/structures/datatypes/primitive/sintdatainformation.hpp M +8-0 kasten/controllers/view/structures/datatypes/primitive/uintdatainformation.hpp https://invent.kde.org/utilities/okteta/commit/29d529e7cc4be8ca5370bef20d8971a9e79ac941 -- You are receiving this mail because: You are watching all bug changes.
[frameworks-syntax-highlighting] [Bug 453759] KSyntaxHighlighting does not build with a C++20 compiler on Windows
https://bugs.kde.org/show_bug.cgi?id=453759 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |FIXED Latest Commit||https://invent.kde.org/fram ||eworks/syntax-highlighting/ ||commit/8664e6eb1d3b45805abf ||fef91fe212d253750bc0 Version Fixed In||5.95.0 Status|ASSIGNED|RESOLVED --- Comment #14 from Friedrich W. H. Kossebau --- Git commit 8664e6eb1d3b45805abffef91fe212d253750bc0 by Friedrich W. H. Kossebau. Committed on 01/06/2022 at 19:53. Pushed by kossebau into branch 'master'. Avoid unprefixed CamelCase headers generated directly in build dir The removal of the PREFIX argument from ecm_generate_headers() call in 42a061b04c507de11b0fb240808fc65a7528d946 resulted in the headers being directly generated in the build dir, thus no longer only being visible via an explicit prefixed include. FIXED-IN: 5.95.0 M +6-0src/lib/CMakeLists.txt https://invent.kde.org/frameworks/syntax-highlighting/commit/8664e6eb1d3b45805abffef91fe212d253750bc0 -- You are receiving this mail because: You are watching all bug changes.
[frameworks-syntax-highlighting] [Bug 453759] KSyntaxHighlighting does not build with a C++20 compiler on Windows
https://bugs.kde.org/show_bug.cgi?id=453759 --- Comment #12 from Friedrich W. H. Kossebau --- Ah, right, yes, I only thought about building KSyntaxHighlighting itself, as the bug report talks about in title and initial description. Once it itself is built (if that works) and you build against the build artifacts in its build dir, we are back at the problem, I see, -- You are receiving this mail because: You are watching all bug changes.
[frameworks-syntax-highlighting] [Bug 453759] KSyntaxHighlighting does not build with a C++20 compiler on Windows
https://bugs.kde.org/show_bug.cgi?id=453759 --- Comment #10 from Friedrich W. H. Kossebau --- So a work-around for you might be to set the CMake C++ standard explicitly to C++17 for that subdirectory with ksyntaxhighligting, to match what is officially supported. Building the code with C++20 only works by chance, so there might be future issues. )s your code open visible somewhere, to get a better idea of your use-case? Still we want to fix ksyntaxhighlighting from the regression to have the unprefixed headers in the build include dir, that is not that ideal in any case, just wondering how urgent it is. -- You are receiving this mail because: You are watching all bug changes.
[frameworks-syntax-highlighting] [Bug 453759] KSyntaxHighlighting does not build with a C++20 compiler on Windows
https://bugs.kde.org/show_bug.cgi?id=453759 --- Comment #8 from Friedrich W. H. Kossebau --- Nicolas: I am curious: given ksyntaxhighlighting itself is set to use C++17 during its build, does MSVC fail to react to that and still make C++20 headers visible or even use them? Or, when you say "Compile KSyntaxHighlighting with Visual Studio 2022 in C++20 mode" you mean, you modified the normal build and actually enforce C++20 instead? So anyone else should not be really affected for now? And in that case, why would you build with C++20 and not just the default C++17? Does that affect the output in any way? And should it perhaps be an option supported by upstream (in case this is also interesting for others)? -- You are receiving this mail because: You are watching all bug changes.
[frameworks-syntax-highlighting] [Bug 453759] KSyntaxHighlighting does not build with a C++20 compiler on Windows
https://bugs.kde.org/show_bug.cgi?id=453759 --- Comment #7 from Friedrich W. H. Kossebau --- Thanks to a shower thought I guess the problem is rather that the current buildsystem setup for KSyntaxHighlighting generates the forward camelcase headers directly in the build directory of the library, which also is part of the include directories set up for the build, and before any system directories. So the preprocessor on searching to resolve #include will simply first come across the "Format" file in the library build dir, and then things go boom. So the generated headers need to be placed outside of the include directories, or, in this case due to other planned usages like from the examples referencing the generated forward headers, be placed into a subdir matching the prefix dir as used in the installed case. So the part of https://invent.kde.org/frameworks/syntax-highlighting/-/merge_requests/313 which adds the "OUTPUT_DIR" argument to the ecm_generate_headers call should improve things and avoid the name collision. -- You are receiving this mail because: You are watching all bug changes.
[frameworks-syntax-highlighting] [Bug 453759] KSyntaxHighlighting does not build with a C++20 compiler on Windows
https://bugs.kde.org/show_bug.cgi?id=453759 Friedrich W. H. Kossebau changed: What|Removed |Added CC||kosse...@kde.org --- Comment #6 from Friedrich W. H. Kossebau --- The problem is not the header named "Format", but when include directories are set up to also have the very subdirectory of the "Format" file as part of it. Which is not by design, instead only the prefix "KSyntaxHighlighting" directoy should be found in the include directories. So that only #include will work. My question would be: can you tell which include actually triggers the wrong header to be used? I saw no include statement in KSyntaxHighlighting itself that uses a plain "format"/"Format" of any kind? So curious how exactly the wrong pick by the compiler is triggered here? -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 450826] Add additional decoding of 16, 32, and 64-bits on left-side
https://bugs.kde.org/show_bug.cgi?id=450826 Friedrich W. H. Kossebau changed: What|Removed |Added Status|REPORTED|NEEDSINFO Resolution|--- |WAITINGFORINFO --- Comment #1 from Friedrich W. H. Kossebau --- > It would be nice if the data could be displayed as groups of 16, 32, or > 64-bits on the left-hand side rather than clicking on the right to see the > decoding table with options for little or big endian. With "left-hand side", you mean the byte value display (with the options binary, octal, decimal, hexadecimal)? And with "right" the character display? What do you mean exactly by "clicking on the right"? After all the decoding table also adapts to the cursor when placed/moved in the byte values view part? What exactly do you mean by "groups of 16, 32, or 64-bits" or "32 or 64-bit values"? Integer values? If so, curious what use-case there is to combine view of 8-bit character set with mult-byte integer values? Would one not rather want to combine the single byte value (as the most common notation/interpretaion of the 8 bits of a byte) with whatever special data type those bytes are used for to encode? Can you give a visual mockup how you would like to see the display done, and perhaps give a sample of the use-case you have? So far I understand your request is a variant of bug 214131, just with multi-byte integer values instead of float or double? -- You are receiving this mail because: You are watching all bug changes.
[Elisa] [Bug 454372] Proposal: switch to the cherry-pick backwards workflow
https://bugs.kde.org/show_bug.cgi?id=454372 Friedrich W. H. Kossebau changed: What|Removed |Added CC|kosse...@kde.org| --- Comment #2 from Friedrich W. H. Kossebau --- As I got cc::ed I have at least to say: fix bugs in the current product version,. Only then integrate those changes also into the development version, by whatever way. Backports to the product version instead come with increased risks of regressions and thus lower quality. Not involved with Elisa otherwise, so unsubscribing -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 337653] Changed a value in a structure view isn't shown after submitting
https://bugs.kde.org/show_bug.cgi?id=337653 Friedrich W. H. Kossebau changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED Version Fixed In||0.26.9 Latest Commit||https://invent.kde.org/util ||ities/okteta/commit/e86a5b4 ||59f6ea3b43d94223183e4a60695 ||b31a62 --- Comment #5 from Friedrich W. H. Kossebau --- Git commit e86a5b459f6ea3b43d94223183e4a60695b31a62 by Friedrich W. H. Kossebau. Committed on 23/05/2022 at 14:23. Pushed by kossebau into branch '0.26'. Structures tool: properly update tree view after committing a change FIXED-IN: 0.26.9 M +9-4kasten/controllers/view/structures/structurestool.cpp M +1-0kasten/controllers/view/structures/structurestool.hpp https://invent.kde.org/utilities/okteta/commit/e86a5b459f6ea3b43d94223183e4a60695b31a62 -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 421357] Support for variable-length strings w/o terminator
https://bugs.kde.org/show_bug.cgi?id=421357 --- Comment #1 from Friedrich W. H. Kossebau --- Alex, would you remember a reason besides no spare time resources that maxCharCount and maxByteCount could not be also optionally scripted properties? Would look into this otherwise in the next weeks. FTR, as another work-around (though resulting in data reparsing) one can use the updateFunc, starting with maxCharCount or maxByteCount set to 0 and then have the value updated from that type-wide generic function. E.g. with JS: var obj = struct({ length: uint16(), data: string("utf-8").set({ maxByteCount: 0, updateFunc: function() { this.maxByteCount = this.parent.length.value; } }) }); or with XML -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 396057] Pascal-style RLE strings not working properly for length 1, both JS and OSD
https://bugs.kde.org/show_bug.cgi?id=396057 Friedrich W. H. Kossebau changed: What|Removed |Added Version Fixed In||0.26.9 Latest Commit||https://invent.kde.org/util ||ities/okteta/commit/4c95f98 ||a3ce12b8517ea10058e59016336 ||165b17 Status|REPORTED|RESOLVED Resolution|--- |FIXED --- Comment #2 from Friedrich W. H. Kossebau --- Git commit 4c95f98a3ce12b8517ea10058e59016336165b17 by Friedrich W. H. Kossebau. Committed on 23/05/2022 at 09:37. Pushed by kossebau into branch '0.26'. Structures tool: fix condition for string maxBytes meaning empty string FIXED-IN: 0.26.9 M +1-1 kasten/controllers/view/structures/datatypes/strings/utf32stringdata.cpp M +1-1 kasten/controllers/view/structures/datatypes/strings/utf8stringdata.cpp https://invent.kde.org/utilities/okteta/commit/4c95f98a3ce12b8517ea10058e59016336165b17 -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 396057] Pascal-style RLE strings not working properly for length 1, both JS and OSD
https://bugs.kde.org/show_bug.cgi?id=396057 --- Comment #1 from Friedrich W. H. Kossebau --- Git commit aaea006c23cf1bb4471d11e7d3cc8cf45e1184ca by Friedrich W. H. Kossebau. Committed on 23/05/2022 at 09:32. Pushed by kossebau into branch '0.26'. Structures tool: clear any existing string data on reading zero-byres string M +1-0 kasten/controllers/view/structures/datatypes/strings/utf16stringdata.cpp M +1-0 kasten/controllers/view/structures/datatypes/strings/utf32stringdata.cpp M +1-0 kasten/controllers/view/structures/datatypes/strings/utf8stringdata.cpp https://invent.kde.org/utilities/okteta/commit/aaea006c23cf1bb4471d11e7d3cc8cf45e1184ca -- You are receiving this mail because: You are watching all bug changes.
[okteta] [Bug 454083] not working
https://bugs.kde.org/show_bug.cgi?id=454083 Friedrich W. H. Kossebau changed: What|Removed |Added Resolution|--- |DUPLICATE Status|REPORTED|NEEDSINFO --- Comment #1 from Friedrich W. H. Kossebau --- Thanks for your feedback. Do I understand correctly that with "Linux Mint 20.3" 'sudo apt install okteta' you got the version 0.26.3? > when i try to open a file nothing happens. You tried to open a file via the menu entry or the toolbar? Sounds like you hit bug 441802, too bad Qt still does not have that fixed. When it comes to no double byte support in the char table, different issue, interest for it though noted. In some future that might be added, other undone things more priority, so nothing to bet on short term sadly. *** This bug has been marked as a duplicate of bug 441802 *** -- You are receiving this mail because: You are watching all bug changes.