Selextract

Wikipedia Table Scraper

Turn the tables on any Wikipedia page into clean rows, with footnote markers removed and merged cells filled in. Works in every language.

SelextractOfficialVersion 1 · updated Nov 4, 2026

Inputs

Wikipedia pages
Required
pages · List · 1 to 10 Wikipedia article links in any language, like https://en.wikipedia.org/wiki/List_of_largest_companies_by_revenue.
Maximum rows
Optional
max_rows · Whole number · Up to 5,000 rows in total. · default 1000
Table number (optional)
Optional
table_number · Whole number · Which table to get, counting data tables from the top of the page. Use 0 for every table. · default 0

Sample output

The publisher hasn't shared a sample.

About

What you get

One row per table row, with the table's column headings as fields, plus the page, the table's number and its caption. Merged cells are filled in, and footnote markers like [1] are removed.

How to use it

Enter up to 10 Wikipedia article links in any language. You get every data table on each page, or choose one table by its number.

Good to know

  • Only data tables are included, not infoboxes or navigation boxes.
  • Values stay as Wikipedia shows them, like $611.3 billion.
  • Wikipedia's text is available under the CC BY-SA license. Credit Wikipedia when you publish the data.
  • Runs use your own Selextract credit.