Every national independence day on Earth: 207 holidays across 168 countries, one row per country-celebration, because some countries celebrate twice (Afghanistan honors leaving the United Kingdom in 1919 and the Soviet Union in 1989, both rows carried through the table's rowspans intact). The decolonization waves are among history's clearest time-series patterns and they land exactly where textbooks put them: 1821 freed nine countries of Latin America, 1918 ten as empires collapsed, the Year of Africa 1960 counts eighteen, 1975 eight as Portugal let go, and 1991 fourteen when the Soviet flag came down. The Independent_From column makes empire visible as a single groupby: the United Kingdom released 50 countries, France 26, Spain 24, the Soviet Union 14, the Ottoman Empire 9 and Portugal 8, a leaderboard verified by test with the UK required on top. Dates split into analytics-ready Holiday_Date (MM-DD), month and day columns, so the calendar question (which month hosts the most independence days?) is one histogram; the year column spans from the ancient (San Marino's founding era) to the modern, each with its decade; every row carries the commemorated event's name and a note on what actually happened, citation-stripped. Anchored by test on the Fourth of July (1776, Kingdom of Great Britain, as the table correctly says rather than the anachronistic United Kingdom), India's 15 August 1947 and Brazil's 7 September 1822. Built with an 11-test acceptance suite.
207 rows in one CSV file, with 10 columns in total: Country, Event, Holiday_Date, Holiday_Month, Holiday_Day, Independence_Year, Decade, Independent_From, Note, Country_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.