Pick Shortest Path tool, inconsistent selection #55745
Labels
No Label
Interest
Alembic
Interest
Animation & Rigging
Interest
Asset Browser
Interest
Asset Browser Project Overview
Interest
Audio
Interest
Automated Testing
Interest
Blender Asset Bundle
Interest
BlendFile
Interest
Collada
Interest
Compatibility
Interest
Compositing
Interest
Core
Interest
Cycles
Interest
Dependency Graph
Interest
Development Management
Interest
EEVEE
Interest
EEVEE & Viewport
Interest
Freestyle
Interest
Geometry Nodes
Interest
Grease Pencil
Interest
ID Management
Interest
Images & Movies
Interest
Import Export
Interest
Line Art
Interest
Masking
Interest
Metal
Interest
Modeling
Interest
Modifiers
Interest
Motion Tracking
Interest
Nodes & Physics
Interest
OpenGL
Interest
Overlay
Interest
Overrides
Interest
Performance
Interest
Physics
Interest
Pipeline, Assets & IO
Interest
Platforms, Builds & Tests
Interest
Python API
Interest
Render & Cycles
Interest
Render Pipeline
Interest
Sculpt, Paint & Texture
Interest
Text Editor
Interest
Translations
Interest
Triaging
Interest
Undo
Interest
USD
Interest
User Interface
Interest
UV Editing
Interest
VFX & Video
Interest
Video Sequencer
Interest
Virtual Reality
Interest
Vulkan
Interest
Wayland
Interest
Workbench
Legacy
Blender 2.8 Project
Legacy
Milestone 1: Basic, Local Asset Browser
Legacy
OpenGL Error
Meta
Good First Issue
Meta
Papercut
Meta
Retrospective
Meta
Security
Module
Animation & Rigging
Module
Core
Module
Development Management
Module
EEVEE & Viewport
Module
Grease Pencil
Module
Modeling
Module
Nodes & Physics
Module
Pipeline, Assets & IO
Module
Platforms, Builds & Tests
Module
Python API
Module
Render & Cycles
Module
Sculpt, Paint & Texture
Module
Triaging
Module
User Interface
Module
VFX & Video
Platform
FreeBSD
Platform
Linux
Platform
macOS
Platform
Windows
Priority
High
Priority
Low
Priority
Normal
Priority
Unbreak Now!
Status
Archived
Status
Confirmed
Status
Duplicate
Status
Needs Info from Developers
Status
Needs Information from User
Status
Needs Triage
Status
Resolved
Type
Bug
Type
Design
Type
Known Issue
Type
Patch
Type
Report
Type
To Do
No Milestone
No project
No Assignees
5 Participants
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: blender/blender#55745
Loading…
Reference in New Issue
No description provided.
Delete Branch "%!s(<nil>)"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
System Information
Suse Linux 42.2
Blender Version
Broken: 2.79b
f4dc9f9d68
Worked: (optional)
Short description of error
The Pick Shortest Path tool has a few oddities:
Exact steps for others to reproduce the error
blender_pick-shortest-path_slideshow_v0.2_upload.mp4
blender_pick-shortest-path_slideshow_v0.2_upload.mkv
Edit: reuploaded video (separate MP4 and MKV provided in case they don't play in all browsers).
pick_shortest_path_selection_seq.blend
See the attached video. But in case it's unavailable in future:
1, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, etc
0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, etc
Alternatively, these steps create a different sequence and might make it easier to see what's happening:
1, 0, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, etc
0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, etc
Added subscriber: @intracube
Added subscriber: @mont29
Please follow our submission template and guidelines, also read these tips about bug reports, and make a complete, valid bug report, with required info, precise description of the issue, precise steps to reproduce it, small and simple .blend and/or other files to do so if needed, etc.
Videos and/or links to external sites etc. are not acceptable as bug report (they can be provided as additional information only).
I'm unsure what the issue is.
I've included the requested info; system info, Blender version, short description of error, exact steps both in video form and written as a step-by-step list.
Since the issue is with an interactive tool, a .blend can't be used to show the issue.
Also, whether it's intended or not, using bold text like that comes across as unnecessary and abrasive.
Added subscribers: @ideasman42, @jenkm
@ideasman42 I also believe there is a problem with "Checker Select", in how it calculates the Offset.
Indeed, something looks wrong in the checker logic, will investigate.
PS: What I find abrasive is having to lose time and copy-paste same canned answer countless times every week because people do not read our basic report rules. tracker is already an huge time and energy black hole, those rules are made to help us devs, so that we can quickly check and validate the bug, instead of spending time trying to carefully decipher the report and reproduce it ourselves.
This issue was referenced by
bc675d90d0
Changed status from 'Open' to: 'Resolved'
Hi,
The offset now looks OK, but the first face is still permanently selected so we get sequences like this:
Also the Nth value seems to define the number of consecutive selected faces +1
So a value of 16 selects 15 faces. The tooltip also says it 'skips' though it's not clear what.
You select all elements but skip (not select) every Nth, (like 3, 6, 9...).
The Skip - number of consecutive elements to skip (not select) at once. (That breaks the "every Nth" logic.)
That's crazy.
Actually, there should be simply the "Number of Selected" and "Number of Unselected" elements.
Hi, any update on the first issue?
First face is still always selected, so this should be reopened.
Yep, Skip set to any value other than 1 breaks the Nth element spacing.
Here's how the tool could logically work:
It might be best to split this into a new task, since it wasn't part of the original report.
Changed status from 'Resolved' to: 'Open'
The "first" element is not the first, it is the previous selection area,
before you perform Pick Shortest Path (or Shortest Path from menu),
so it should remain unchanged.
But then there's still the wrong offset.
And yes, the "Skip Nth" UX/UI very confusing and must be redesigned,
but it's not a bug, so it's not for this bug tracker.
But the first face is included in the sequence pattern, otherwise we'd see an extra selected face at the start: 1110011001100 and that isn't what's happening.
So again, it's kind of inconsistent.
The offset looks fixed in builds after the 14th Aug. What settings make it work incorrectly for you?
It's not just confusing, it doesn't do what the manual or interface text suggest.
Right now the tool is actually very close to the alternative logic you described further up: "there should be simply the "Number of Selected" and "Number of Unselected" elements." - if we pretend that the Nth element is 'number of selected' and Skip is 'number of unselected' then the only issue is that the Nth element value is out by one.
You have some selected area, in your case it's one element but can be any number (it is the previous selection area).
Then you perform Pick Shortest Path and it adds a new selection.
The previous selection should remain unchanged, i.e. first face in your case.
Ah ok, so the tool is intentionally additive selection. That makes sense.
The logic all now seems OK if the controls are interpreted as 'number of consecutve selected elements + 1' and 'number of consecutive skipped elements'.
So aside from the Nth value being one unit too high, It's just the general naming/labeling that's out of step.
@intracube can you suggest exact changes, I'll check over them and commit if they make sense regarding the existing logic.
@ideasman42 I can't think of a clearer description with the current logic so the existing labels/tooltips are probably best left as-is.
Anything else would need changes to the Nth and Skip, which would then break user scripts. The earlier commit might already do that...
@intracube, I wouldn't worry about breaking user scripts for this particular operator, also with 2.8 we allow some breakage.
Changed status from 'Open' to: 'Archived'
This report talks about subtle changes in behavior, we could address but don't think its urgent.
No response in over a week, closing.
There's still the wrong default Offset.
You have one (in this case) selected element:
Then you Ctrl-click (Pick Shortest Path) last element:
Note that here the first element is not part of the Pick Shortest Path operation.
By default, we have the wrong offset:
(in 2.79b) Offset = -2 gives correct results:
(in 2.80) after this changes
bc675d90d0
it works differently but also wrong.Changed status from 'Archived' to: 'Open'
Tested and the results seem fine to me.
Saying "Note that here the first element is not part of the Pick Shortest Path operation." is more of an opinion.
The way it works now gives even selection grouping which happens to include the first face.
From you're reply it's not clear what you consider correct behavior either.
This first face (on the screenshots) is not added during the "Pick Shortest Path", it's not affected by the "Pick Shortest Path" options, and always remains selected.
So, the first (0 index) face for the "Pick Shortest Path" is the face marked in green, here begins the first group.
Let's say we have some selection:
And we add a new one with "Pick Shortest Path":
By default, with Offset = 0, we have a pattern offset by one element.
An even clearer example:
you can easily see that these two vertices are not affected (always remain selected) by the Shortest Path settings.
Other improvements:
The "Skip every Nth element" logic breaks if the Skip value is not equal to 1.
If the "Nth Element" = 1, it means skip each, that is, there is no effect at all, all elements skipped, but now it works the opposite way.
I suggest changing it to "skip X elements after each Y elements".
Skip After = Nth Element - 1
It's a small change, and it's suitable for both "Checker Select" and "Checker Deselect".
This issue was referenced by
0381727663
Changed status from 'Open' to: 'Resolved'