GitLab

Original: 30d976610f86eb4edeb13784325d897f8d69d2d0

Original: 136222252030abe406e3ec00f907eeb1f6ed043f

This is more consistent with usage elsewhere in the code.
Original: 07019313082d3ff7d5b2688622d3330a708aa6bd

This is useful for setting up separate HL and HLV analyses.
Original: 51c27d236eb9b7f3ce0fd27c597e2e4636a49fa0

- use sin-based expression for angleV instead of cos-based, which yields exact 0 result
  in case of identical input vectors and therefore angleV=0
Original: 60388400adc577ffa8f4af64b46a2e942b03a980

Original: d3ed67fae74605cc968b85d0ec670f85d69d5336

Original: 2b043e1748eb2297c3b87fc3db92ddb27d3d9c20

- this happened when requested interval is below SFT catalog interval
- throw warnings if the requested interval is empy
- throw error if the requested intervall is invalid
- fixes #5608
Original: cfc0b643b4004d9481108a10a37c31b478b2f622

- check for 0 before taking log() instead of trapping ValueError so that the exception handling doesn't hide other issues.
Original: c5a11873399a0092a62053eff4bb4b2915ae7485

- should have been -inf, call self.normalize() to do whatever's correct.
Original: 8d0d216ce8e38d3935329eba5d879c6132f8e09f

Original: 1f093fb1e859a6f81265dc981a15406176aa6fac

Original: 04bc5b7fab240883502ac3e785a53ced21b534e8

Original: a45c6fc5c37f7721d535535f38b28f5f67b2b1c6

 -new function write_transientFstatMap_to_fp()
  with more atomic inputs, useful for SWIG usage
 -write_transientCandidateAll_to_fp() now just a wrapper
 -closes #5616
Original: c8e513ffab5b95db4ce01e27fc77ee42fd8d94be

- 'last' marker on bitflags needs to be exactly one higher than
  maximal possible combined bitflag value
- that bug was introduced in a086d950f6 as part of issue #5206
- fixes #5620
Original: 5947e32b90fe1ee05a942a412d59956742f0a69a

Original: 7a6ecfd62b8ed2cd8f552652c016e4b941b8e922

Original: 1b70c0a63ee6807f964b13ceb9440f82240504f1

- add commented-out implementation of the N-way coincidence rate calculation based on qhull's half-space intersection code.  requires scipy >= 0.19 (not available on reference platforms, nor, unfortunately the *next* reference platform.  sigh)
Original: b722708c3c00d5ffa60fb20ee51bf99330d5946c

Original: e7265fb3572ee5d314b6d40c5d2ba4453f5581fc

- move an attribute look-up out of a loop
Original: 6063f047a4ea9c3350459145202ad9af66e5c37b

Original: dde2e9909b5237e4ccb6959f0c1d3616a60128e4

- Divide reduced supersky parameter space (two unit
  disks) into roughly square patches of equal area
Original: 4e2c021995167de708d725c6eabfcdad231f78b6

- Does all supported physical parameters returned in 2 x PulsarDopplerParams
Original: d93004925226b756bbd1281c0dc79c58148c1481

Original: 53d382bcca1662c3e492e5a6ad0288ef8d091644

Original: ea7135170d96f67e600bb8cf413affe6eb8a3dea

- XLALScaleSuperskyMetricsFiducialFreq() should be used to rescale all metrics
Original: 7cd64a6d65123960208c833810ba77b560f70fa8

Original: 8ae6cc370f1dbaf6e408f989921c996771baea68

Original: dd5c8b2126ee9bc0b60df90a2bf2f98a61ba8b19

Original: da330ad623cba54eedba25a19d2b7ec2e2e720b3

- In anticipation of Arunava adding the binary metric into this function
Original: c9cb803079b8e7c89f734f9899e3f79d534540e8

- Replace packed GSL matrix with more sensible/extensible struct for
  storing coordinate transform data
- Use FITS table instead of array to write coordinate transform data,
  which will BREAK existing FITS files!
- Test FITS file writing in SuperskyMetricsTest
- FIXME: replace 'SMAX' with either 'nspins', 'nsky_offsets - 1',
  or something else...
Original: 77769cf401b07360e4091b8bbdd2513c241bf211

- Return >0 if a limited-size heap is full, 0 otherwise
  (or if heap has unlimited size), or <0 on error
Original: c566bb3373e7caa5742a9e23dbe0732de62dc453

- Copy table column data to a (resizeable) internal buffer, large enough to
  permit some overflow, then copy the correct portion into the table record
Original: 2eedba11a6fdf0bd6eb086b75e2beff93db3bc41

- Definitely should use 'TINT' which maps to int (ffgpv() calls
  ffgpvk() if datatype == TINT) and int is probably 4 bytes long...
- Need to adjust the output of fits_get_eqcoltype() as CFITSIO is
  getting confused between TLONG and TINT, as are we all...
Original: 25f6899f3af184dd47b565710b371429beb3fc4e

Original: 05960a4038077f3e2f7267032535454e6a5cdd85

- Change from a gsl_matrix into an opaque struct SuperskyTransformData,
  though internally for now this just contains a gsl_matrix pointer 'data'
- SuperskyMetricsTest must have its own definition of SuperskyTransformData
Original: 7f94322b84ba4e5f045c2afbec7cbb418f9d0a27

Original: 6ef2ae567a40a2dc982734944cfee84430a35fb7

Original: b0ebccba49aa5da56ed8d6cde06a8024c5597732

Original: 8c740eb0cabe2669d637c10df58410b894a00461

Original: 43074b15249edf3e8effb1951cfb7f46b7361a83