The measurements Wikidata does not have, extracted from where they actually live: 5,322 insects with body length, wingspan, mass, adult lifespan or measured strength pulled from Wikipedia prose, because the probe that opened this build found 288 body lengths across ALL animal taxa in the graph. Every row earned its place by stating at least one number: 3,673 body lengths from fairyfly scale to the 700 mm stick-insect ceiling, 1,878 wingspans topping out at the white witch's 300, 680 adult lifespans from the mayfly's day to ant queens' three decades (the 10-year cap lifts only for sentences about queens, so a butterfly can never live fifty years again), and the strength column crowned correctly: Onthophagus taurus, the horned dung beetle, measured pulling 1,141 times its own body mass, a number the article writes without a comma, which defeated the first number regex and now anchors the test suite. The extraction gauntlet is documented in the scraper and worth reading: a 2 mm bark beetle briefly weighed the 120-gram ethanol lure from its own trap description, butterflies lived for 50 years because their woodland was replanted in the last 50 years, and milligram-scale beetles rounded to zero until mass kept four decimals. Mass records the maximum stated at any life stage (the Goliath beetle's famous 100 g is its larva); ranges keep their upper bound, one disclosed rule. Taxonomy, common names and IUCN status join from Wikidata on 99 percent of rows; Lepidoptera lead with 2,019. Ships with flagship gates (Goliath, Atlas moth, mayfly) that block the fetch when extraction drifts. Built with a 12-test acceptance suite.
5,322 rows in one CSV file, with 13 columns in total: Insect, Insect_ID, Common_Name, Rank, Order, Family, Body_Length_mm, Wingspan_mm, Mass_g, Adult_Lifespan_Days, Strength_Multiple, IUCN_Status, Insect_URL.
It is built by programmatically scraping en.wikipedia.org, last pulled on 2026-10-12. 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.