Paul Price added a comment - 09/Mar/16 3:30 PM Perry Gee , would you please review these two commits? price@price-laptop:~/LSST/shapelet (tickets/DM-5281=) $ git sub commit 9b8f33080e7e1778e4bc3a3aab13bee5d6af65f6 Author: Jim Bosch <jbosch@astro.princeton.edu> Date: Wed Jul 17 16:57:27 2013 -0400   Let the value to normalize to be specified in [Multi]ShapeletFunction   include/lsst/shapelet/MultiShapeletFunction.h | 4 ++-- include/lsst/shapelet/ShapeletFunction.h | 4 ++-- src/MultiShapeletFunction.cc | 6 +++--- src/ShapeletFunction.cc | 4 ++-- 4 files changed, 9 insertions(+), 9 deletions(-)     price@price-laptop:~/LSST/skymap (tickets/DM-5281=) $ git sub commit 0dee8ed02e2e75ddb8083cca0e579b50a7f7f760 Author: Paul Price <price@astro.princeton.edu> Date: Tue Feb 26 16:56:08 2013 +0900   Set vertex list in ExplicitTractInfo.   python/lsst/skymap/tractInfo.py | 3 +++ 1 file changed, 3 insertions(+)

Perry Gee added a comment - 13/Mar/16 1:14 PM My only question about ExplicitTractInfo is: how did it every work before? _vertexCoordList was just left empty? Should setting this list from bbox when vertexCoordList is empty be a default behavior of TractInfo __init _? Not saying that. Just wondering. ShapeletFunction.h and ShapeletFunction.cc – changes are fine. If you feel the need for a pull request, let me know. I am not requesting any changes.

Paul Price added a comment - 14/Mar/16 10:01 AM I don't think the _vertexCoordList is really used much: I haven't been able to find any code that uses it, and I can't remember what prompted this change. But setting it to something should be preferred to leaving it empty. I don't think it's the base class' job to set it, because it's the subclass that's providing the empty vertexCoordList , and setting the bbox using a different method. It's just that we forgot when writing the subclass to actually set it when done.

Paul Price added a comment - 14/Mar/16 10:02 AM Merged to master.