#1947 closed defect (fixed)
unbalanced model - Wrong root in connection graph
| Reported by: | Christian Schubert | Owned by: | Adrian Pop |
|---|---|---|---|
| Priority: | critical | Milestone: | 1.9.0 |
| Component: | Frontend | Version: | trunk |
| Keywords: | unbalanced connection graph | Cc: |
Description
I found a bug in the connection graph handling.
If the world object, which defines Connection.root(), is used in a submodel with an outer prefix, it seems that the root is not recognized as such.
See the attached test model (CGraphBug.TopModel). From the equations it can be seen, that mass.frame_a is chosen as root, as the following equations are generated:
27 (22): if mass.useQuaternions then mass.phi_dd[3] = 0.0 mass.phi_dd[2] = 0.0 [...]
This branch is only selected if Connections.isRooted(mass.frame_a) becomes true.
Attachments (4)
Change History (7)
by , 12 years ago
| Attachment: | CGraphBug.mo added |
|---|
by , 12 years ago
| Attachment: | CGraphBug.mos added |
|---|
by , 12 years ago
| Attachment: | CGraphBug-GrapVizFile.png added |
|---|
comment:1 by , 12 years ago
Hm, it seems we're missing the connection:
connect(world.frame_b, frame_a)
from SubModel1 as far as I can see. I'll look more into it.
by , 12 years ago
| Attachment: | CGraphBug-GrapVizFile-correct.png added |
|---|
generated with +d=cgraphGraphVizFile
comment:3 by , 12 years ago
Christian, can you add this model to the testsuite so we don't break this stuff again?
Thanks!
Note:
See TracTickets
for help on using tickets.
generated with +d=cgraphGraphVizFile