Comparing periods with lag

The Idaho field office wants to see how many tests are performed in each test period (year and month) that it is selected, and compare it with the prior one. They'd like to start with test site 13903 in Minidoka county.

Este exercício faz parte do curso

Introduction to Redshift

Ver curso

Instruções do exercício

Build a CTE, minidoka_13903, selecting all the columns from idaho_sample and converting the sample_date to a date as an additional column (date_sampled) for station 13903.
In the main query:
- Count the records for the current month
- Count the records for the prior month via a window function.

Exercício interativo prático

Experimente este exercício completando este código de exemplo.

WITH minidoka_13903 AS (
    SELECT *,
           -- Converts the sample_date to a date
           ___(___ AS ___) as date_sampled
    FROM public_intro_redshift.idaho_samples
    WHERE fk_monitoringlocation = 13903
)
SELECT DATE_PART('year', date_sampled) as sample_year,
       DATE_PART('month', date_sampled) as sample_month,
       -- Count the records
       ___,
       -- Count the records for the prior month
       ___(___, ___) OVER (
           ORDER BY DATE_PART('year', date_sampled),
                    DATE_PART('month', date_sampled)
       ) AS prior_readings
  -- Use the prefiltered CTE you created above
  FROM ___
 GROUP BY sample_year,
          sample_month
 ORDER BY sample_year DESC,
          sample_month DESC;

Editar e executar o código