Supplementary data #8
Conversation
|
Is it possible to decouple this from the other PR? It makes it a bit difficult to isolate which issues are from this PR solely or within both PR's |
raphacan
force-pushed
the
supplementary_data
branch
from
6ff87e9
to
f5ee5c0
Compare
raphacan
force-pushed
the
supplementary_data
branch
from
4e599e0
to
a2f7b27
Compare
|
@lcweaver & @jessegv : I rebased the supplementary branch to the latest of the copy,swap move, so that when this is merged, we can continue with this PR. If you want to see the supplementary additions, you should compare it to the swap/move branch: @florian: You could update the 9.2 branch with it. |
| { | ||
| if (exemplar->img[i]->base.supplementary_data) | ||
| { | ||
| status = vxSetSupplementaryUserDataObject(&pmd->img[i]->base, exemplar->img[i]->base.supplementary_data); |
There was a problem hiding this comment.
Will this also need to be done similarly for other exemplars such as object array?
There was a problem hiding this comment.
I need to check the supplementary data extension itself if we need to set the supplementary on the object array level. I don't think so because the objarray is only a container, only the objects contained inside have supplementary data.
There was a problem hiding this comment.
Thanks. Please let us know the results of your findings
There was a problem hiding this comment.
@lcweaver
from the extension:
"Object Arrays
An object array (the container) may also have supplementary data, distinct from the data of the elements of the array. ...."
There was a problem hiding this comment.
Thanks. In that case, is it the case that we will need similar logic added to the object array exemplar function?
I would think so given that if the exemplar was created with different supplementary data on the object array vs the element of the object array, then the resulting objects would also need to be set in a similar manner
There was a problem hiding this comment.
can you elaborate ? I do not understand your point.
to me, the obj array does neither inherit nor give its supplementary to the elements, they can have their own supplementary. So this is not similar to the pyramid.
|
It looks like Lucas and I have marked the conversations which we verified as fixed as "resolved". There are still several open not yet resolved. |
| { | ||
| if (exemplar->img[i]->base.supplementary_data) | ||
| { | ||
| status = vxSetSupplementaryUserDataObject(&pmd->img[i]->base, exemplar->img[i]->base.supplementary_data); |
There was a problem hiding this comment.
Thanks. In that case, is it the case that we will need similar logic added to the object array exemplar function?
I would think so given that if the exemplar was created with different supplementary data on the object array vs the element of the object array, then the resulting objects would also need to be set in a similar manner
jessegv
force-pushed
the
supplementary_data
branch
from
4a40b4a
to
c5cc792
Compare
jessegv
force-pushed
the
supplementary_data
branch
from
c5cc792
to
98de587
Compare
jessegv
force-pushed
the
supplementary_data
branch
from
a5822e8
to
8421421
Compare
Signed-off-by: raphacan <[email protected]>
jessegv
force-pushed
the
supplementary_data
branch
from
fbdd3fa
to
0e468ca
Compare
test now running, one remaining issue with the test_util and dangling ref count in the test read-only param and child objects.
maybe it is worth to check it in combination with the copy/swap/move PR.