You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
According to the docs, there is currently no -s flag on the //stack command to also stack the selection.
There is actually an undocumented effect of -s to //stack, though not what I would expect. Currently specifying -s leads to a selection around the last pasted instance of the stack.
What I would expect is for the selection to include the original and all pasted instances.
Use case: Currently, stacking in two directions requires two commands and adjusting the selection inbetween. This feature would make the middle step obsolete.
//stack 3 east
[Adjust selection, could be replaced by -s]
//stack 3 north
A Solution
If current -s behaviour is unintentional: Change the flag to the behaviour described above.
If current -s behaviour is intentional: Add a new flag that has the behaviour described above (e.g. -ss, if duplicate letters are no problem, although I suspect they are).
In both cases, the flags should be added to the documentation.
For clarification, the -s flag is documented as "Shift the selection to the last stacked copy" – its purpose is to allow easily continuing the stack from that final point (eg, to continue it along, or stack in another direction)
4. I think we're open to this under -S, to separate it from -s.
Sounds great!
2. For clarification, the -s flag is documented as "Shift the selection to the last stacked copy" – its purpose is to allow easily continuing the stack from that final point (eg, to continue it along, or stack in another direction)
I think it's worth considering a separate "expand" flag that keeps the original selection too. Thought being that if you want to make a grid starting with one cell, you'd have to stack that cell in one direction, re-select the original cell, then stack in the perpendicular direction.
Also worth noting that non-axis-aligned offsets or non-cuboid selections may have unexpected behaviors here - WorldEdit will happily stack a sphere multiple times and select the last one, but it can't select multiple spheres without degrading to their AABB. This, at the least, needs to be told to the user if they're stacking in one of these ways.
Activity
me4502 commentedon Oct 17, 2026
For clarification, the
-sflag is documented as "Shift the selection to the last stacked copy" – its purpose is to allow easily continuing the stack from that final point (eg, to continue it along, or stack in another direction)octylFractal commentedon Oct 17, 2026
The documentation in question: https://worldedit.enginehub.org/en/latest/commands/#command-//stack
octylFractal commentedon Oct 17, 2026
I think we're open to this under
-S, to separate it from-s.p-98 commentedon Oct 21, 2026
Sounds great!
I see, thanks for the link. Maybe those flags should also be mentioned in the usage section (https://worldedit.enginehub.org/en/latest/usage/regions/regionops/#stacking), which is where I looked.
wizjany commentedon Oct 24, 2026
I think it's worth considering a separate "expand" flag that keeps the original selection too. Thought being that if you want to make a grid starting with one cell, you'd have to stack that cell in one direction, re-select the original cell, then stack in the perpendicular direction.
Also worth noting that non-axis-aligned offsets or non-cuboid selections may have unexpected behaviors here - WorldEdit will happily stack a sphere multiple times and select the last one, but it can't select multiple spheres without degrading to their AABB. This, at the least, needs to be told to the user if they're stacking in one of these ways.