Experimental patches to use PAL instead of C SLALIB

[timj]

Untested (lsst.SConsUtils does not exist) but hopefully should give an idea of what is required.

[rhiannonlynne]

Ooh, this option makes me happy. It also makes me wish we had unit tests
and integration tests. TBD!

On Tue, Sep 15, 2015, 7:23 PM Tim Jenness [email protected] wrote:

Untested (lsst.SConsUtils does not exist) but hopefully should give an

idea of what is required.

You can view, comment on, or merge this pull request online at:

#1
Commit Summary

• Experimental patches to use PAL instead of C SLALIB

File Changes

• M SConstruct
  https://github.com/lsst/mops_daymops/pull/1/files#diff-0 (4)
• M src/Makefile
  https://github.com/lsst/mops_daymops/pull/1/files#diff-1 (16)
• M src/MopsDetection.cc
  https://github.com/lsst/mops_daymops/pull/1/files#diff-2 (12)
• M ups/mops_daymops.table
  https://github.com/lsst/mops_daymops/pull/1/files#diff-3 (2)

Patch Links:

• https://github.com/lsst/mops_daymops/pull/1.patch
• https://github.com/lsst/mops_daymops/pull/1.diff

—
Reply to this email directly or view it on GitHub
#1.

[timj]

In terms of testing, SLA and PAL share the same test suite with slight modifications, and PAL was a direct port of the Fortran SLA in many cases. Where there are differences, these come from the use of ERFA behind the scenes which uses newer models for nutation and precession and the like than SLA implements.