Jim Bosch added a comment - 19/Jul/18 11:38 AM Sophie Reed and Jim Bosch are doing this as a pair-programming exercise.

Jim Bosch added a comment - 19/Jul/18 1:59 PM We've made changes in geom and afw that remove the default for invert in order to catch code that is using the old default.  So far we have not found any cases where the invert=True behavior was desired.  We'll return to this in the future.

Yusra AlSayyad added a comment - 02/Aug/18 2:29 PM - edited Dan Taranu and Yusra AlSayyad joined in on the Party Programming today to inspect Box2I and Box2D constructors and add explicit invert=False . So far haven't found any cases where invert=True made sense. Update: We looked at and updated the following 30 packages. I got afw  compiling in the evening. Now all the packages that depend on afw and geom need to be tested.   geom Jim+Sophie pushed afw Jim+Sophie+Yusra  pushed display_matplotlib Sophie no branch shapelet Sophie pushed ctrl_orca Sophie no branch display_ds9 Sophie no branch obs_base Sophie pushed skymap Sophie pushed display_firefly Sophie no branch ctrl_execute Sophie no branch obs_test Sophie pushed ctrl_platform_lsstvc Sophie no branch pipe_base Sophie no branch ctrl_pool Sophie no branch cp_pipe Sophie no branch coadd_utils Sophie pushed meas_base Sophie pushed coadd_chisquared Yusra pushed meas_algorithms Yusra pushed meas_extensions_photometryKron Yusra no branch meas_deblender Yusra pushed meas_extensions_psfex Yusra no branch ip_diffim Yusra pushed ip_isr Yusra pushed meas_astrom Yusra pushed meas_modelfit Yusra pushed meas_extensions_simpleShape Yusra pushed meas_extensions_astrometryNet Yusra pushed meas_extensions_convolved Yusra pushed obs_monocam Yusra pushed meas_extensions_shapeHSM Yusra pushed obs_ctio0m9 Yusra no branch pipe_tasks Yusra pushed obs_sdss Dan pushed obs_lsstSim Dan pushed obs_cfht Dan pushed synpipe Dan no branch pipe_drivers Dan pushed obs_subaru Dan pushed ci_ctio0m9 Dan no branch obs_decam Dan pushed meas_mosaic Yusra pushed obs_comCam Dan pushed validation_data_decam Dan no branch lsst_obs Dan no branch validate_drp Dan no branch validation_data_cfht Dan no branch lsst_apps Dan no branch testdata_jointcal Dan no branch jointcal Yusra pushed

Sophie Reed added a comment - 17/Mar/21 6:34 PM Having checked all the branches above we are happy that no one was using this with invert=True, therefore we propose just changing the default behaviour.

Jim Bosch added a comment - 17/Mar/21 7:46 PM Note that the final result on the RFC is to remove the invert option as well as changing the default behavior; that's what we should now do.