gistfile1.sql

SELECT DISTINCT ON(random.first, random.last) coalesce(mf.name, ff.name), l.name FROM (
  SELECT
    RANDOM() * 90.020 AS first -- dataset is 90% of most popular
    , RANDOM() * 90.483 AS last
    , CASE WHEN RANDOM() > 0.5 THEN 'male' ELSE 'female' END AS sex
    , CASE WHEN RANDOM() > 0.5 THEN 'model' ELSE 'vehicle' END AS detail
  FROM generate_series(1,10,1)
) AS random
LEFT OUTER JOIN census.names AS mf
  ON random.sex = 'male'
  AND mf.name_type = 'MALE_FIRST'
  AND mf.cumfreq > random.first
LEFT OUTER JOIN census.names AS ff
  ON random.sex = 'female'
  AND ff.name_type = 'FEMALE_FIRST'
  AND ff.cumfreq > random.first
LEFT OUTER JOIN census.names AS l
  ON l.name_type = 'LAST'
  AND l.cumfreq > random.first
ORDER BY random.first, random.last, mf.cumfreq, ff.cumfreq, l.cumfreq
;