RELEASE_NOTES_6.2.2 - TheDaniel166/moira GitHub Wiki

Moira 6.2.2 - Lunar Node And Lilith Honesty

Release date: 2026-08-15

Public upgrade path: 6.2.1 to 6.2.2. Callers still on 6.2.0 inherit the wheel catalog plus this node/Lilith honesty patch.

Moira 6.2.2 is a lunar-point honesty release. It binds Mean Lilith to the same IERS conventional arguments as the mean node, and it stops reporting a mean-rate proxy as the speed of True Node and True Lilith. Public function signatures and REST paths do not change.

In this release

  • IERS Mean Lilith ā€” Body.LILITH / mean_lilith() is the IERS 2003 secular mean apogee F + Ī© āˆ’ l + 180Ā° from _fundamental_args, with the 5.2.2 nutation= frame unchanged. ERFA faf03 + faom03 āˆ’ fal03 + 180Ā° is the authority. Modern longitudes move by less than 0.2ā€³ versus the previous Meeus polynomial.
  • True Node and True Lilith speed ā€” NodeData.speed is the circular finite-difference of the DE osculating true-of-date longitude (0.002-day TT step, same span as planetary longitude rates). The previous constant mean-node rate and borrowed mean-Lilith rate are removed. Speeds may be large or retrograde; that is the geometry.
  • Receipts ā€” Lilith reduction stages name iers_2003_mean_apogee_solution.
  • Swiss comparison doctrine ā€” a shared astrological name is not a shared number. Swiss digits are not a target. The 4ā€“7ā€² residual versus SE_MEAN_APOG is expected (IERS secular mean vs ELP hybrid). True Lilith versus SE_OSCU_APOG remains the close numerical match.

See MIGRATING_FROM_SWISS_EPHEMERIS.md.

Not in 6.2.2

  • smoothed_lilith() and Body.SMOOTHED_LILITH are not implemented.
  • Kernel presence does not change which product Body.LILITH is.
  • Swiss SE_MEAN_APOG digits are not a parity target.
  • No Track B, visibility, Track C, or website entitlement change.

Install

pip install moira-astro==6.2.2

from moira.nodes import mean_lilith, true_lilith, true_node

# Mean Lilith: IERS secular mean. Do not assert arcminute parity with swe.MEAN_APOG.
mean_lilith(jd_ut)

# True Lilith / True Node: longitude unchanged; speed is dĪ»/dt of that point.
true_lilith(jd_ut, reader=reader)
true_node(jd_ut, reader=reader)

Addresses

TheDaniel166/moira#18 is answered by this release: the Mean Lilith residual versus Swiss SE_MEAN_APOG is a series difference, not a defect. The issue is not closed as a Swiss-parity bug.