[PIPE2D-85] Make changes to handle ndarray now using size_t for its array sizes Created: 24/Sep/16 Updated: 07/Oct/16 Resolved: 07/Oct/16 |
|
| Status: | Done |
| Project: | DRP 2-D Pipeline |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Story | Priority: | Major |
| Reporter: | rhl | Assignee: | rhl |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Sprint: | 2014-16 |
| Reviewers: | aritter |
| Description |
|
In the v12_1 release, we're switching to using ndarry as an external dependency, and one change is that it now uses size_t not int for array sizes (in conformity with the C++ STL). Unfortunately, things like
fail to compile (as shape is size_t and 1 is int). The fix is
|
| Comments |
| Comment by rhl [ 24/Sep/16 ] |
|
There's also a problem in the swig interface (it's using Size and wanting ndarray::Size). I have a hack, but would rather come up with a real fix, so am waiting on that to send this for review. |
| Comment by rhl [ 05/Oct/16 ] |
|
Removing some of the extra %template statements resolved the ndarray::Size problem. |
| Comment by aritter [ 05/Oct/16 ] |
|
drp_stella does not compile anymore with lsst_apps 12.0. |
| Comment by rhl [ 07/Oct/16 ] |
|
We are going to standardise on the v12_1 release. So done and merging to master |