Article source size histograms on skwiki

SQL

USE skwiki_p;

-- Log2 bins
SELECT
    FLOOR(LOG2(p.page_len)) as size_bucket,
    POW(2, FLOOR(LOG2(p.page_len))) as size_from,
    POW(2, FLOOR(LOG2(p.page_len)) + 1) - 1 as size_to,
    COUNT(*) as frequency
  FROM
    page p
  WHERE
    p.page_namespace = 0 and
    p.page_is_redirect = 0
  GROUP BY
    size_bucket
  ORDER BY
    size_bucket
  ;
  
-- Log10 bins
SELECT
    FLOOR(LOG10(p.page_len)) as size_bucket,
    POW(2, FLOOR(LOG10(p.page_len))) as size_from,
    POW(2, FLOOR(LOG10(p.page_len)) + 1) - 1 as size_to,
    COUNT(*) as frequency
  FROM
    page p
  WHERE
    p.page_namespace = 0 and
    p.page_is_redirect = 0
  GROUP BY
    size_bucket
  ORDER BY
    size_bucket
  ;
  
-- Linear 1000 B bins
SQL

Checking query status...
