Details
-
Type:
Bug
-
Status: Won't Fix
-
Resolution: Done
-
Fix Version/s: None
-
Component/s: ap_association
-
Labels:None
-
Story Points:1
-
Epic Link:
-
Team:Alert Production
-
Urgent?:No
Description
When running the Gen 3 pipeline multiple times in the same run (see DM-27325), I get many copies of the following error:
diaPipe.associator.diaCalculation WARN: Error in ap_HTMIndex.calculate: __init__(): incompatible constructor arguments. The following argument types are supported:
|
1. lsst.geom.SpherePoint()
|
2. lsst.geom.SpherePoint(longitude: lsst.geom.Angle, latitude: lsst.geom.Angle)
|
3. lsst.geom.SpherePoint(longitude: float, latitude: float, units: lsst.geom.AngleUnit)
|
4. lsst.geom.SpherePoint(vector: lsst.sphgeom._sphgeom.Vector3d)
|
5. lsst.geom.SpherePoint(unitVector: lsst.sphgeom._sphgeom.UnitVector3d)
|
6. lsst.geom.SpherePoint(lonLat: lsst.sphgeom._sphgeom.LonLat)
|
7. lsst.geom.SpherePoint(other: lsst.geom.SpherePoint)
|
|
|
Invoked with: diaObjectId
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
Name: ra, dtype: object, diaObjectId
|
50809677860045963 0.0387833 rad
|
50809677860045963 0.0387833 rad
|
50809677860045963 0.0387833 rad
|
50809677860045963 0.0387833 rad
|
50809677860045963 0.0387833 rad
|
50809677860045963 0.0387833 rad
|
50809677860045963 0.0387833 rad
|
Name: decl, dtype: object
|
This error is hard to diagnose because the argument's __repr__ method doesn't even name the type of the object. Please implement a __repr__ that is more consistent with Python conventions.
Attachments
Activity
Respectfully, the error message generated by pybind11 always reports the repr form of each individual argument to the function/method, and does not group them in any way. In this case, that means that the first argument is
diaObjectId
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
50809677860045963 2.61725 rad
|
Name: ra, dtype: object
|
I think you said at one point that this is the representation of a Pandas Dataframe, and therefore outside our control. Is that no longer the case?
Krzysztof Findeisen
added a comment - - edited Respectfully, the error message generated by pybind11 always reports the repr form of each individual argument to the function/method, and does not group them in any way. In this case, that means that the first argument is
diaObjectId
50809677860045963 2.61725 rad
50809677860045963 2.61725 rad
50809677860045963 2.61725 rad
50809677860045963 2.61725 rad
50809677860045963 2.61725 rad
50809677860045963 2.61725 rad
50809677860045963 2.61725 rad
Name: ra, dtype: object
I think you said at one point that this is the representation of a Pandas Dataframe, and therefore outside our control. Is that no longer the case? That is correct, it is a Pandas Series object.
Perhaps the issue could simply be closed as Invalid or Won't Fix?
Krzysztof Findeisen
added a comment - Perhaps the issue could simply be closed as Invalid or Won't Fix? Sounds good to me. I'll mark it as "won't fix"
Output shown here is due to the _repr_ definition in Pandas, not within the DM Stack, hence it will not be fixed.
Further, the error here is triggered by an upstream break in the assumption that DiaObjects are unique in the catalogs processed within ap_association. This issue and safeguards for it are currently being address in DM-28555.
The error reported shows that a multiple values were input to a function that expects single floats or other single objects.