From 8ec888732fc7045c1a8167db7209ba3ddd2ea702 Mon Sep 17 00:00:00 2001
From: Mike Rosseel <mike.rosseel@gmail.com>
Date: Sun, 22 Mar 2026 20:09:08 +0100
Subject: [PATCH] fix: handle MPC comet data with unparseable orbital elements

Skyfield's load_comets_dataframe() can return string dtypes for numeric
columns when the MPC data file contains rows it can't parse (e.g. the
MPEC 2026-F34 batch of 27 historical comets with malformed fields).
Even a few bad rows cause pandas to fall back to dtype=object for the
entire perihelion_year column, which then crashes comet_orbit() for
*every* comet with a numpy UFuncNoLoopError.

Fix: force numeric coercion with pd.to_numeric() after loading, drop
rows with unparseable essential fields, and add a try/except safety
net around individual comet processing.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---
 python/PiFinder/comets.py | 40 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 39 insertions(+), 1 deletion(-)

diff --git a/python/PiFinder/comets.py b/python/PiFinder/comets.py
index 9430094ff..a3185c054 100644
--- a/python/PiFinder/comets.py
+++ b/python/PiFinder/comets.py
@@ -4,6 +4,7 @@
 from skyfield.constants import GM_SUN_Pitjeva_2005_km3_s2 as GM_SUN
 from PiFinder.utils import Timer, comet_file
 from PiFinder.calc_utils import sf_utils
+import pandas as pd
 import requests
 import os
 import logging
@@ -187,6 +188,39 @@ def calc_comets(
         with open(comet_file, "rb") as f:
             comets_df = mpc.load_comets_dataframe(f)
 
+        # Ensure orbital element columns are numeric — MPC data sometimes
+        # contains rows that Skyfield's parser can't handle (e.g. MPEC
+        # 2026-F34 historical comets), which causes the entire column to
+        # become dtype=object (strings).  Skyfield's comet_orbit() then
+        # crashes on every comet with a numpy UFuncNoLoopError.
+        numeric_cols = [
+            "perihelion_year",
+            "perihelion_month",
+            "perihelion_day",
+            "argument_of_perihelion_degrees",
+            "longitude_of_ascending_node_degrees",
+            "inclination_degrees",
+            "eccentricity",
+            "perihelion_distance_au",
+            "magnitude_g",
+            "magnitude_k",
+        ]
+        for col in numeric_cols:
+            if col in comets_df.columns:
+                comets_df[col] = pd.to_numeric(comets_df[col], errors="coerce")
+
+        # Drop rows where essential orbital elements couldn't be parsed
+        essential = [
+            "perihelion_year",
+            "perihelion_month",
+            "perihelion_day",
+            "eccentricity",
+            "perihelion_distance_au",
+        ]
+        comets_df = comets_df.dropna(
+            subset=[c for c in essential if c in comets_df.columns]
+        )
+
         # Report progress after file loading (roughly 33% of setup time)
         if progress_callback:
             progress_callback(1)
@@ -208,7 +242,11 @@ def calc_comets(
 
         for comet in comet_data:
             if comet_names is None or comet[0] in comet_names:
-                result = process_comet(comet, dt)
+                try:
+                    result = process_comet(comet, dt)
+                except Exception as e:
+                    logger.warning(f"Skipping comet {comet[0]}: {e}")
+                    result = {}
                 if result:
                     comet_dict[result["name"]] = result