A CodeSights flagship and the densest registry on the platform: 299 planetary moons where nearly every row carries real orbital mechanics, and the acceptance suite's showpiece test recovers Kepler's third law from the shipped data itself: regressing log period on log semi-major axis within the Jupiter, Saturn and Uranus systems must return slope 1.5 with correlation above 0.995, and it does. The build earned its correctness the hard way, catalogued in the scraper: a guessed natural-satellite class QID resolved to an entity named Varbak, so the Moon itself was asked what it is; Pluto's moons live in their own demoted class and had to be rejoined; moon masses arrive in teragrams through yottagrams and convert to kilograms; Earth briefly had a second moon named Taiyin, a lunar deity, until the facts-or-article rule showed it the door; and a physics guard nulled the two orbit pairs (Cyllene, Geirrod) that violated Kepler by 10x and 29x, graph errors wearing orbital elements. The counts are the solar system's: Saturn 162, Jupiter 85, Uranus 29, Neptune 15, Pluto 5 including Charon (discovered 1978), Mars 2, Earth exactly 1 by test. Columns: diameter (preferring stated values, falling back to twice the radius), mass, orbital period in days with retrograde negatives normalized, semi-major axis, eccentricity, discovery year (floored at Galileo's 1610, which opens the era by test) and discoverer, with most moons discovered after 1990, the CCD survey boom the suite asserts. Twelve acceptance tests.
299 rows in one CSV file, with 11 columns in total: Moon, Moon_ID, Planet, Diameter_km, Mass_kg, Orbital_Period_Days, Semi_Major_Axis_km, Eccentricity, Discovery_Year, Discoverer, Moon_URL.
It is built by programmatically scraping wikidata.org, last pulled on 2026-10-13. Datasets are versioned; older versions stay downloadable.
Yes. Every CodeSights dataset is completely free as a CSV download; a free account is all it takes. Anyone can preview the data without signing in.
Yes. The exact Python scraper that built it is viewable on the dataset page by any signed-in member, so every number is reproducible.
Automated scraping leaves room for error and the underlying sources change over time, so no version is guaranteed accurate or complete. If a number matters, verify it against the original source.