[krita] [Bug 408749] New: Krita crashes when drawing on the canvas while animation is playing and no frames are added
https://bugs.kde.org/show_bug.cgi?id=408749 Bug ID: 408749 Summary: Krita crashes when drawing on the canvas while animation is playing and no frames are added Product: krita Version: 4.2.1 Platform: MS Windows OS: MS Windows Status: REPORTED Severity: crash Priority: NOR Component: Animation Assignee: krita-bugs-n...@kde.org Reporter: kapyia@gmail.com Target Milestone: --- SUMMARY Krita crashes when drawing on the canvas while animation is playing but there are no frames are added. STEPS TO REPRODUCE 1. Open up a new clean document and switch over to the Animation workspace 2. Immediately go over to the Animation docker and press Play 3. With the Freehand Brush Tool try to draw on the canvas while the Timeline is playing OBSERVED RESULT Krita instantly locks up, leaving no kritacrash.log file behind. EXPECTED RESULT I would expect Krita to stop the animation and continue as if you'd never hit Play in the first place, possibly triggering a warning saying "No active animation frames could be found". SOFTWARE/OS VERSIONS Windows 10 ADDITIONAL INFORMATION When there ARE active frames available in the Timeline, Krita prevents the lock up by switching over to an available frame, pausing the timeline, and drawing onto it. Perhaps the software gets stuck looking for active frames and not finding any to draw onto? -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 409617] Krita goes unresponsive while editing text
https://bugs.kde.org/show_bug.cgi?id=409617 Kapyia changed: What|Removed |Added Ever confirmed|0 |1 CC||kapyia@gmail.com Status|REPORTED|CONFIRMED --- Comment #4 from Kapyia --- I'm able to reproduce this bug on my Windows system, independent of font, size, alignment, etc.. Krita will crash if the first new-line is the "Shift+Return" variant. However, if the first new-line is a regular "Return" one the software doesn't seem to crash for me, no matter what you use afterwards. I noticed that a regular "Return" new-line will start a new paragraph in the SVG-source, whereas the "Shift+Return" variant just adds a new-line to the SVG-source itself. -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 409748] Brush preset dropdown keeps changing to a tag
https://bugs.kde.org/show_bug.cgi?id=409748 Kapyia changed: What|Removed |Added CC||kapyia@gmail.com --- Comment #2 from Kapyia --- I'm having a similar experience, though for me the Brush Presets Docker always start at the Digital Atilier tag. And I know it's kept opening at the Pixel brush tag before. It seems like Krita just sporadically picks a tag and sticks to opening with that one for a while. Note that this is happening on my Windows 10 system. I've held off reporting it since I know the entire resource system is being reworked. -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 410807] New: Python Debugger throws exception when ran
https://bugs.kde.org/show_bug.cgi?id=410807 Bug ID: 410807 Summary: Python Debugger throws exception when ran Product: krita Version: 4.2.5 Platform: MS Windows OS: MS Windows Status: REPORTED Severity: normal Priority: NOR Component: Scripting Assignee: krita-bugs-n...@kde.org Reporter: kapyia@gmail.com Target Milestone: --- Created attachment 122059 --> https://bugs.kde.org/attachment.cgi?id=122059&action=edit Python Debugger Traceback SUMMARY When trying to run the Scripter's built-in debugger it immediately throws an exception and won't run. STEPS TO REPRODUCE 1. Launch Krita and open the Scripter 2. Load or create a Python script file 3. Try to debug the script OBSERVED RESULT The Scripter throws an exception. The traceback as been attached to the report. SOFTWARE/OS VERSIONS Windows 10 Krita x64 4.2.5 Portable (bug is present at least as far back as 4.2.3) -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 410807] Python Debugger throws exception when ran
https://bugs.kde.org/show_bug.cgi?id=410807 --- Comment #2 from Kapyia --- The exception is thrown even when debugging a simple Hello World script or even empty files :( -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 416126] New: Mislabeled method: setRotation()
https://bugs.kde.org/show_bug.cgi?id=416126 Bug ID: 416126 Summary: Mislabeled method: setRotation() Product: krita Version: 4.2.8 Platform: MS Windows OS: MS Windows Status: REPORTED Severity: minor Priority: NOR Component: Scripting Assignee: krita-bugs-n...@kde.org Reporter: kapyia@gmail.com Target Milestone: --- The method 'setRotation' in libkis/Canvas.cpp is either mislabeled or incomplete. It rotates the canvas by the amount of the given angle instead of setting the canvas' rotation to that angle. This seems to be because 'setRotation' passes its unaltered argument on to a different method called 'rotateCanvas' without making adjustments. The fix is to subtract the canvas' current rotation from the value of the argument before passing it on to 'rotateCanvas'. Or you could rename the function and adjust the documentation :P (I would submit a MR with a fix, but I don't have an environment set up to check if what I submit is valid C++ code) -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 416793] New: API creates broken layer masks
https://bugs.kde.org/show_bug.cgi?id=416793 Bug ID: 416793 Summary: API creates broken layer masks Product: krita Version: 4.2.8 Platform: MS Windows OS: MS Windows Status: REPORTED Severity: normal Priority: NOR Component: Scripting Assignee: krita-bugs-n...@kde.org Reporter: kapyia@gmail.com Target Milestone: --- Created attachment 125442 --> https://bugs.kde.org/attachment.cgi?id=125442&action=edit Script to create a mask node This problem was brought to light thanks to Frohman and tiar over at Krita-Artists.org. There's currently a problem with the libkis API where trying to create a mask node of any kind results in an empty/broken layer mask object. The created node always has the name "effect" (kis_effect_mask.cc, line 27) and has no bounds/size. >From what I can tell, the problem stems from the method 'createNode' in Document.cpp creating mask nodes somewhat incorrectly, resulting in them being "incomplete". On my system it also leads to a crash when I try to make Krita display the mask (eg. trying to isolate the layer). I've attached a small script that helps reproduce the problem in Krita. -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 421289] New: Broken Channels management
https://bugs.kde.org/show_bug.cgi?id=421289 Bug ID: 421289 Summary: Broken Channels management Product: krita Version: 4.2.9 Platform: MS Windows OS: MS Windows Status: REPORTED Severity: normal Priority: NOR Component: Scripting Assignee: krita-bugs-n...@kde.org Reporter: kapyia@gmail.com Target Milestone: --- Created attachment 128336 --> https://bugs.kde.org/attachment.cgi?id=128336&action=edit Broken Channels script SUMMARY I've been running into a couple of issues while experimenting with manipulating individual channels using the Python API. First thing I noticed is that the 'channels' function of the Node class returns a QList with the channels ordered as "BGRA". This could be intentional for all I know and not a bug, but I thought/assumed them to be ordered as "RGBA". - Second problem I ran into was that the 'pixelData' function of a Channel returns a byte array with a full 4 bytes per pixel, but only the alpha channel of each pixel holds data. Not too confusing, but it would've been nice if the documentation was a bit clearer about this :) This feeds into the third problem, which is that the 'setPixelData' function doesn't account for only the alpha channel/byte actually has valuable data. Writing the pixel data from one channel to another results in only every 4 pixels actually having data, whilst the remaining 3/4 of the pixels are empty. The second and third problem could probably be solved with one swoop if either the QBA returned by the 'pixelData' function only had one byte per pixel, or if the 'setPixelData' only wrote from the alpha channel/byte of the input QBA :) - The fourth problem I've run into is that I seem to only be able to get the data from the Blue channel. No matter which index in the list retrieved using the 'channels' function in the Node class, it's always the Blue channel. This is evident when writing the pixel data to the alpha channel of a node and comparing the thumbnails in the Channels docker. - Fifth and final problem I've encountered is that when I tried writing pixel data from one channel to another it appears the data gets corrupted, resulting in unwanted artefacts. I've attached a script to help test these issues :) SOFTWARE/OS VERSIONS Windows 10 x64 -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 421289] Broken Channels management
https://bugs.kde.org/show_bug.cgi?id=421289 --- Comment #1 from Kapyia --- Created attachment 128337 --> https://bugs.kde.org/attachment.cgi?id=128337&action=edit Color diagram to help test the issues Color diagram to help test the issues -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 421289] Broken Channels management
https://bugs.kde.org/show_bug.cgi?id=421289 --- Comment #2 from Kapyia --- Created attachment 128338 --> https://bugs.kde.org/attachment.cgi?id=128338&action=edit The results from me using the script The results from me using the script -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 396479] Smudging is weakened inside selections
https://bugs.kde.org/show_bug.cgi?id=396479 --- Comment #2 from Kapyia --- Created attachment 113921 --> https://bugs.kde.org/attachment.cgi?id=113921&action=edit Video demonstration I haven't made any changes to the brush preset, it is the default one pre-packaged with Krita. I could attach a .kra file if you'd like, but to me this problem isn't bound to any specific project and happens with all smudge brush presets. It might be worth mentioning that I'm not using Instant Preview mode, and I tried turning off Canvas Graphics Acceleration as well, but the problem persists. -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 396483] New: Dulling mode of the Smudge Brush engine ignores Overlay Mode
https://bugs.kde.org/show_bug.cgi?id=396483 Bug ID: 396483 Summary: Dulling mode of the Smudge Brush engine ignores Overlay Mode Product: krita Version: 4.1.0 Platform: MS Windows OS: MS Windows Status: UNCONFIRMED Severity: normal Priority: NOR Component: Brush engines Assignee: krita-bugs-n...@kde.org Reporter: kapyia@gmail.com Target Milestone: --- Don't think I can explain it further than the title + video demonstration, sorry :) -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 396483] Dulling mode of the Smudge Brush engine ignores Overlay Mode
https://bugs.kde.org/show_bug.cgi?id=396483 --- Comment #1 from Kapyia --- Video attachment too large, here's a link: https://drive.google.com/file/d/1S9-qbE6CLmWd2XscJGyJMy8o7d1wABfr/view?usp=sharing -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 396615] New: The areas around undone selections remain active
https://bugs.kde.org/show_bug.cgi?id=396615 Bug ID: 396615 Summary: The areas around undone selections remain active Product: krita Version: nightly build (please specify the git hash!) Platform: MS Windows OS: MS Windows Status: UNCONFIRMED Severity: grave Priority: NOR Component: Tools Assignee: krita-bugs-n...@kde.org Reporter: kapyia@gmail.com Target Milestone: --- Created attachment 113991 --> https://bugs.kde.org/attachment.cgi?id=113991&action=edit Video demonstration In Krita 4.2.0 Pre-Alpha (krita-nightly-x64-v4.0.0.51-1735-g024d1a36eb), undone selected areas aren't properly unloaded. Steps to reproduce: - Open a new document - Select a large area with any Selection Tool (I recommend Outline Selection Tool) - Make new smaller selections (note: Selection Tools set to Replace mode), making sure the new selections overlap the first one - Undo actions until you're back at the first selection - Paint inside of the selection - Painted areas will be restricted to boxes of the undone selections Occurs with and without Canvas Graphics Acceleration. Note: This bug has occasionally enabled me to be able to draw *outside* of a selected area, but I have yet to figure out how to reproduce this consistently. -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 396606] Undoing a crop action causes data corruption
https://bugs.kde.org/show_bug.cgi?id=396606 Kapyia changed: What|Removed |Added CC||kapyia@gmail.com --- Comment #1 from Kapyia --- Present for me in 4.2.0-pre-alpha (git 024d1a3) on Win 10 as well -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 396605] Cancelling a transformation action causes data corruption
https://bugs.kde.org/show_bug.cgi?id=396605 Kapyia changed: What|Removed |Added CC||kapyia@gmail.com --- Comment #1 from Kapyia --- Present for me in 4.2.0-pre-alpha (git 024d1a3) on Win 10 as well -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 396615] The areas around undone selections remain active
https://bugs.kde.org/show_bug.cgi?id=396615 --- Comment #2 from Kapyia --- (In reply to Boudewijn Rempt from comment #1) > Could you check whether this happens with the 4.1.0 stable release? If not, > I suspect it's to do with Dmitry's selection work... Not present in 4.1.0 stable! -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 396704] Using Tarnsform Tool (ctrl+t) on a selection, when applied, the contents of the selection disappears.
https://bugs.kde.org/show_bug.cgi?id=396704 Kapyia changed: What|Removed |Added CC||kapyia@gmail.com --- Comment #2 from Kapyia --- Created attachment 114034 --> https://bugs.kde.org/attachment.cgi?id=114034&action=edit Video demonstration I can confirm with my similar setup (nVidia GPU, Intel CPU, Win 10) that the Transform tool is broken in the 4.2.0 pre-alpha (git 9a2372d). Undoing + redoing Transform actions results in duplicates on the canvas, and Transforming a layer multiple times results in data corruption. -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 396704] Using Tarnsform Tool (ctrl+t) on a selection, when applied, the contents of the selection disappears.
https://bugs.kde.org/show_bug.cgi?id=396704 Kapyia changed: What|Removed |Added Ever confirmed|0 |1 Status|UNCONFIRMED |CONFIRMED -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 403436] New: channel[].visible() always returns False
https://bugs.kde.org/show_bug.cgi?id=403436 Bug ID: 403436 Summary: channel[].visible() always returns False Product: krita Version: 4.1.7 Platform: MS Windows OS: MS Windows Status: REPORTED Severity: minor Priority: NOR Component: Scripting Assignee: krita-bugs-n...@kde.org Reporter: kapyia@gmail.com Target Milestone: --- SUMMARY Checking for a channel's visibility using the visible() function in Python always returns False. STEPS TO REPRODUCE 1. Create a new document in Krita 2. Run the following script in the Scripter tool: ### from krita import * doc = Krita.instance().activeDocument() channel = doc.activeNode().channels()[0] channel.setVisible(True) print(channel.visible()) doc.refreshProjection() ### OBSERVED RESULT Output is always 'False'. Setting a channel's visibility works, but the flag doesn't seem to be updated. Additionally, changing a channel's visibility using setVisible() isn't reflected in the Channels docker. EXPECTED RESULT Output from visible() should reflect the visibility of the channel. OS Windows 10 -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 403436] channel[].visible() always returns False
https://bugs.kde.org/show_bug.cgi?id=403436 --- Comment #2 from Kapyia --- Disregard the line about the Channels docker. This script manipulates the individual layer's channels, so there's no reason for the visibility status in the Channels docker to be affected. -- You are receiving this mail because: You are watching all bug changes.
