v1.0.1

Bug fix release

Known Issues

Fixes for the issues affecting the original v1.0.1 are readily available in the attached 1.0.1-2 tarball; on the main branch; and will be included in the v1.0.2 bug fix release.

• #28 -- d_light() divide by zero if called with a zero vector pos argument (e.g. Sun or SSB position, depending on the solar-system calculator type).
• #29 -- [CRITICAL] All calculations involving the IAU2000 nutation or ee_ct() yield invalid results on non-Intel x86 platforms, such as ARMv7 or PowerPC, due to a different signedness of the char integer type on these platforms.

Fixes

• #15 -- cirs_to_itrs(), itrs_to_cirs(), tod_to_itrs(), and itrs_to_tod() all had a unit conversion bug in using the ut1_to_tt argument [s] when converting TT-based Julian date to UT1-based JD [day] internally.
• #17 -- Fixes to included examples.c

Additions

• Added cirs_to_app_ra() and app_to_cirs_ra() for convenience to convert between right ascensions measured from the CIO (for CIRS) vs measured from the true equinox of date, on the same true equator of date.

Changes

• #18 -- Changed definition of NOVAS_AU to the IAU definition of exactly 149597870.7 km (a <10 m change from the previously defined value, which has been renamed to DE405_AU).
• Various corrections and changes to documentation.

v1.0

[1.0.0] - 2024-03-01

This is the initial release of the SuperNOVAS library. Changes are indicated w.r.t. the upstream NOVAS C 3.1 library
from which SuperNOVAS is forked from.

Known Issues

• #15 -- cirs_to_itrs(), itrs_to_cirs(), tod_to_itrs(), and itrs_to_tod() all have a unit conversion bug in using the ut1_to_tt argument [s] when converting TT-based Julian date to UT1-based JD [day] internally. A fix for this bug is included in the main branch and the v1.0.1 bug fix release.

• #28 -- d_light() divide by zero if called with pos argument is a zero vector (e.g. Sun or SSB, depending on the solar-system calculator type). Fix available on the main branch and will be included in the v1.0.2 bug fix release.

• #29 -- [CRITICAL] All calculations in this release involving the IAU2000 nutation or ee_ct() yield invalid results on non-Intel x86 platforms, such as ARMv7 or PowerPC, due to a different signedness of the char integer type on these platforms. Fix available on the main branch and will be included in the v1.0.2 bug fix release.

Fixed

• Fixes the sidereal_time bug, whereby the sidereal_time() function had
  an incorrect unit cast. This is a known issue of NOVAS C 3.1.

• Some remainder calculations in NOVAS C 3.1 used the result from fmod() unchecked, which led to the wrong results
  when the numerator was negative. This affected the calculation of the mean anomaly in solsys3.c (line 261) and
  the fundamental arguments calculated in fund_args() and ee_ct() for dates prior to J2000. Less critically, it
  also was the reason cal_date() did not work for negative JD values.

• Fixes antedating velocities and distances for light travel time in ephemeris(). When getting positions and
  velocities for Solar-system sources, it is important to use the values from the time light originated from the
  observed body rather than at the time that light arrives to the observer. This correction was done properly for
  positions, but not for velocities or distances, resulting in incorrect observed radial velocities or apparent
  distances being reported for spectroscopic observations or for angular-physical size conversions.

• Fixes bug in ira_equinox() which may return the result for the wrong type of equinox (mean vs. true) if the the
  equinox argument was changing from 1 to 0, and back to 1 again with the date being held the same. This affected
  routines downstream also, such as sidereal_time().

• Fixes accuracy switching bug in cio_basis(), cio_location(), ecl2equ(), equ2ecl_vec(), ecl2equ_vec(),
  geo_posvel(), place(), and sidereal_time(). All these functions returned a cached value for the other
  accuracy if the other input parameters are the same as a prior call, except the accuracy.

• Fixes multiple bugs related to using cached values in cio_basis() with alternating CIO location reference
  systems.

• Fixes bug in equ2ecl_vec() and ecl2equ_vec() whereby a query with coord_sys = 2 (GCRS) has overwritten the
  cached mean obliquity value for coord_sys = 0 (mean equinox of date). As a result, a subsequent call with
  coord_sys = 0 and the same date as before would return the results GCRS coordinates instead of the requested mean
  equinox of date coordinates.

• Fixes aberration() returning NaN vectors if the ve argument is 0. It now returns the unmodified input vector
  appropriately instead.

• Fixes unpopulated az output value in equ2hor() at zenith. While any azimuth is acceptable really, it results in
  unpredictable behavior. Hence, we set az to 0.0 for zenith to be consistent.

• Fixes potential string overflows and eliminates associated compiler warnings.

• Fixes the ephem_close bug, whereby ephem_close() in
  eph_manager.c did not reset the EPHFILE pointer to NULL. This is a known issue of NOVAS C 3.1.

• Supports calculations in parallel threads by making cached results thread-local.

Added

• New debug mode and error traces. Simply call novas_debug(NOVAS_DEBUG_ON) or novas_debug(NOVAS_DEBUG_EXTRA)
  to enable. When enabled, any error conditions (such as NULL pointer arguments, or invalid input values etc.) will
  be reported to the standard error, complete with call tracing within the SuperNOVAS library, s.t. users can have
  a better idea of what exactly did not go to plan (and where). The debug messages can be disabled by passing
  NOVAS_DEBUF_OFF (0) as the argument to the same call.

• Added Doxygen markup of source code and header.

• Added Makefile for GNU make.

• Added continuous integration on GitHub, including regression testing, static analysis, and doxygen validation.

• Added an number of precompiler constants and enums in novas.h to promote consistent usage and easier to read
  code.

• New runtime configurability:

  • The planet position calculator function used by ephemeris() can be set at runtime via set_planet_provider(),
    and set_planet_provider_hp() (for high precision calculations). Similarly, if planet_ephem_provider() or
    planet_ephem_provider_hp() (defined in solsys-ephem.c) are set as the planet calculator functions, then
    set_ephem_provider() can set the user-specified function to use with these to actually read ephemeris data
    (e.g. from a JPL ephemeris file).

  • If CIO locations vs GSRS are important to the user, the user may call set_cio_locator_file() at runtime to
    specify the location of the binary CIO interpolation table (e.g. cio_ra.bin) to use, even if the library was
    compiled with the different default CIO locator path.

  • The default low-precision nutation calculator nu2000k() can be replaced by another suitable IAU 2006 nutation
    approximation via set_nutation_lp_provider(). For example, the user may want to use the iau2000b() model
    instead or some custom algorithm instead.

• New intutitive XYZ coordinate conversion functions:

  • for GCRS - CIRS - ITRS (IAU 2000 standard): gcrs_to_cirs(), cirs_to_itrs(), and itrs_to_cirs(),
    cirs_to_gcrs().
  • for GCRS - J2000 - TOD - ITRS (old methodology): gcrs_to_j2000(), j2000_to_tod(), tod_to_itrs(), and
    itrs_to_tod(), tod_to_j2000(), j2000_to_gcrs().

• New itrs_to_hor() and hor_to_itrs() functions to convert Earth-fixed ITRS coordinates to astrometric azimuth
  and elevation or back. Whereas tod_to_itrs() followed by itrs_to_hor() is effectively a just a more explicit
  2-step version of the existing equ2hor() for converting from TOD to to local horizontal (old methodology), the
  cirs_to_itrs() followed by itrs_to_hor() does the same from CIRS (new IAU standard methodology), and had no
  prior equivalent in NOVAS C 3.1.

• New ecl2equ() for converting ecliptic coordinates to equatorial, complementing existing equ2ecl().

• New gal2equ() for converting galactic coordinates to ICRS equatorial, complementing existing equ2gal().

• New refract_astro() complements the existing refract() but takes an unrefracted (astrometric) zenith angle as
  its argument.

• New convenience functions to wrap place() for simpler specific use: place_star(), place_icrs(),
  place_gcrs(), place_cirs(), and place_tod().

• New radec_star() and radec_planet() as the common point for existing functions such as astro_star()
local_star(), virtual_planet(), topo_planet() etc.

• New time conversion utilities tt2tdb(), get_utc_to_tt(), and get_ut1_to_tt() make it simpler to convert
  between UTC, UT1, TT, and TDB time scales, and to supply ut1_to_tt arguments to place() or topocentric
  calculations.

• Co-existing solarsystem() variants. It is possible to use the different solarsystem() implementations
  provided by solsys1.c, solsys2.c, solsys3.c and/or solsys-ephem.c side-by-side, as they define their
  functionalities with distinct, non-conflicting names, e.g. earth_sun_calc() vs planet_jplint() vs
  planet_eph_manager vs planet_ephem_provider(). See the section on
  Building and installation further above on including a selection of these in your library
  build.)

• New novas_case_sensitive(int) to enable (or disable) case-sensitive processing of object names. (By default NOVAS
  object names were converted to upper-case, making them effectively case-insensitive.)

• New make_planet() and make_ephem_object() to make it simpler to configure Solar-system objects.

Changed

• Changed to support for calculations in parallel threads by making cached results thread-local.
  This works using the C11 standard _Thread_local or else the earlier GNU C >= 3.3 standard __thread modifier.
  You can also set the preferred thread-local keyword for your compiler by passing it via -DTHREAD_LOCAL=... in
  config.mk to ensure that your build is thread-safe. And, if your compiler has no support whatsoever for
  thread_local variables, then SuperNOVAS will not be thread-safe, just as NOVAS C isn't.

• SuperNOVAS functions take enums as their option arguments instead of raw integers. These enums are defined in
  novas.h. The same header also defines a number of useful constants. The enums allow for some compiler checking,
  and make for more readable code that is easier to debug. They also make it easy to see what choices are available
  for each function argument, without having to consult the documentation each and every time.

• All SuperNOVAS functions check for the basic validity of the suppli...