🥇SQL HowTo: scribe aliquod tempus rectum in interrogatione seu "Elementarius tres-via"

Periodice, munus quaerendi notitias relatas sub statuto clavium oritur; donec nos adepto summa numerum records.

Exemplum est maxime "verisimile" ad ostentationem XX difficultates vetustissima, recensentur in album elit (exempli gratia, infra eandem partem). Pro variis moderatorum "dashboards" cum brevibus summariis arearum operis, similis locus saepius requiritur.

In articulo, deductionem in PostgreSQL considerabimus versionem "surgentem" solvendi talem problema, "smertorem" et algorithmum valde complexum. "Loop" in SQL cum conditione exitus ex data inventaquae possunt esse utilia et ad communem progressionem et ad usum in aliis similibus casibus.

Sit scriptor ut a test dataset from prior articulus. Ut tabulae outputa non "salire" interdum cum valores sorted congruunt, extendere subiectum index addendo primaria key. Eodem tempore, hoc statim singularitatem illam dabit nobisque praestabit singularitatem eius modi ordinis;

CREATE INDEX ON task(owner_id, task_date, id);
-- а старый - удалим
DROP INDEX task_owner_id_task_date_idx;

Sicut auditum est, sic scriptum est

Primum, versionem simplicissimam postulationis delineemus, ids agentium transiens ordinata ut initus:

SELECT
  *
FROM
  task
WHERE
  owner_id = ANY('{1,2,4,8,16,32,64,128,256,512}'::integer[])
ORDER BY
  task_date, id
LIMIT 20;

[Aspice explain.tensor.ru]

Aliquantulus tristis - tantum XX tabulas iussimus, et Index Scan nobis reddidit 960 linesquod tum etiam fringilla erat.

unnest + ARRY

Prima consideratio quae nos adiuvat - si opus est summa XX sorted monumentis satis est legere non plus quam XX in eodem ordine digestus est clavis. bonum, idoneus index (dominus_id, task_date, id) habemus.

Eadem mechanismo extrahendi et "convertendi in columnas" utamur. integralis mensa viscus, id est tandem articulum. Et etiam applicate convolutionem ad usum ordinatae functionis ARRAY():

WITH T AS (
  SELECT
    unnest(ARRAY(
      SELECT
        t
      FROM
        task t
      WHERE
        owner_id = unnest
      ORDER BY
        task_date, id
      LIMIT 20 -- ограничиваем тут...
    )) r
  FROM
    unnest('{1,2,4,8,16,32,64,128,256,512}'::integer[])
)
SELECT
  (r).*
FROM
  T
ORDER BY
  (r).task_date, (r).id
LIMIT 20; -- ... и тут - тоже

[Aspice explain.tensor.ru]

Oh, iam multo melius est! XL% citius ac 40 temporibus minus notitia legere debebat.

Materializatio tabularum monumentorum per CTENotabo quod in quibusdam casibus conatum statim operari cum campis recordis postquam eam in subquisitione quaerendo, sine "involutione" in CTE, ducere potest. "multiplicatio" InitPlan pro numero eorumdem agrorum;

SELECT
  ((
    SELECT
      t
    FROM
      task t
    WHERE
      owner_id = 1
    ORDER BY
      task_date, id
    LIMIT 1
  ).*);

Result  (cost=4.77..4.78 rows=1 width=16) (actual time=0.063..0.063 rows=1 loops=1)
  Buffers: shared hit=16
  InitPlan 1 (returns $0)
    ->  Limit  (cost=0.42..1.19 rows=1 width=48) (actual time=0.031..0.032 rows=1 loops=1)
          Buffers: shared hit=4
          ->  Index Scan using task_owner_id_task_date_id_idx on task t  (cost=0.42..387.57 rows=500 width=48) (actual time=0.030..0.030 rows=1 loops=1)
                Index Cond: (owner_id = 1)
                Buffers: shared hit=4
  InitPlan 2 (returns $1)
    ->  Limit  (cost=0.42..1.19 rows=1 width=48) (actual time=0.008..0.009 rows=1 loops=1)
          Buffers: shared hit=4
          ->  Index Scan using task_owner_id_task_date_id_idx on task t_1  (cost=0.42..387.57 rows=500 width=48) (actual time=0.008..0.008 rows=1 loops=1)
                Index Cond: (owner_id = 1)
                Buffers: shared hit=4
  InitPlan 3 (returns $2)
    ->  Limit  (cost=0.42..1.19 rows=1 width=48) (actual time=0.008..0.008 rows=1 loops=1)
          Buffers: shared hit=4
          ->  Index Scan using task_owner_id_task_date_id_idx on task t_2  (cost=0.42..387.57 rows=500 width=48) (actual time=0.008..0.008 rows=1 loops=1)
                Index Cond: (owner_id = 1)
                Buffers: shared hit=4"
  InitPlan 4 (returns $3)
    ->  Limit  (cost=0.42..1.19 rows=1 width=48) (actual time=0.009..0.009 rows=1 loops=1)
          Buffers: shared hit=4
          ->  Index Scan using task_owner_id_task_date_id_idx on task t_3  (cost=0.42..387.57 rows=500 width=48) (actual time=0.009..0.009 rows=1 loops=1)
                Index Cond: (owner_id = 1)
                Buffers: shared hit=4

Idem testimonium "exquisitum est" 4 times… Donec PostgreSQL 11, haec agendi ratio regulariter occurrit, et solutio est "involvere" in CTE, quod est sine condicione finis optimizer in his versionibus.

recursive accumulator

In priore versione, in summa, legitur 200 lines propter necessarias 20. Iam not 960, sed minus - estne?

Experiamur ut opus sit ut scientia summa LX monumentis. Hoc est, nisi data detractione iteramus, donec quantum ad nos pervenitur.

Gradus I: Committitur List

Patet, nostrum "scopum" indicem 20 viscusrum incipere ab "primo" introitu pro una clavium possessoris nostri. Ergo invenimus talem "Primum" pro singulis clavibus et pone in indice, in ordine quo volumus - (negotium, id).

Gradus 2: invenire "proximum" records

Nunc primum ingressum ex indice nostro accipiamus et initium faciamus "gradus" adhuc in indice cum dominus_id-key servato, tunc omnes tabulae inventae sunt proxime sequentes in delectu consequens. Scilicet, nisi donec transire applicatae clavem secundum ingressum in indice.

Si evenit ut "transivimus" secundum ingressum, tunc ultimum legere ingressum debet ad indices loco primi (cum eodem possessore), post quem elenchus iterum digestus est.

Hoc est, semper accipimus ut album non plus quam unum ingressum in singulis clavibus habeat (si ingressus sunt, et "transivimus" non habemus, tunc primus ingressu solum ex indice evanescet et nihil adiicietur. ), et illi semper sorted in ascendendo ordinem clavis applicationis (negotium, id).

Gradus III, Filtering et Expanding Records

In versuum nostrorum parte delectu recursivo, nonnulla monumenta rv duplicantur - primum invenimus ut "terminus 2 introitus elenchus transiens", deinde ut primum ex indice substituimus. Et ideo primo eventum eliquari debet.

Terribilis quaestio finalis

WITH RECURSIVE T AS (
  -- #1 : заносим в список "первые" записи по каждому из ключей набора
  WITH wrap AS ( -- "материализуем" record'ы, чтобы обращение к полям не вызывало умножения InitPlan/SubPlan
    WITH T AS (
      SELECT
        (
          SELECT
            r
          FROM
            task r
          WHERE
            owner_id = unnest
          ORDER BY
            task_date, id
          LIMIT 1
        ) r
      FROM
        unnest('{1,2,4,8,16,32,64,128,256,512}'::integer[])
    )
    SELECT
      array_agg(r ORDER BY (r).task_date, (r).id) list -- сортируем список в нужном порядке
    FROM
      T
  )
  SELECT
    list
  , list[1] rv
  , FALSE not_cross
  , 0 size
  FROM
    wrap
UNION ALL
  -- #2 : вычитываем записи 1-го по порядку ключа, пока не перешагнем через запись 2-го
  SELECT
    CASE
      -- если ничего не найдено для ключа 1-й записи
      WHEN X._r IS NOT DISTINCT FROM NULL THEN
        T.list[2:] -- убираем ее из списка
      -- если мы НЕ пересекли прикладной ключ 2-й записи
      WHEN X.not_cross THEN
        T.list -- просто протягиваем тот же список без модификаций
      -- если в списке уже нет 2-й записи
      WHEN T.list[2] IS NULL THEN
        -- просто возвращаем пустой список
        '{}'
      -- пересортировываем словарь, убирая 1-ю запись и добавляя последнюю из найденных
      ELSE (
        SELECT
          coalesce(T.list[2] || array_agg(r ORDER BY (r).task_date, (r).id), '{}')
        FROM
          unnest(T.list[3:] || X._r) r
      )
    END
  , X._r
  , X.not_cross
  , T.size + X.not_cross::integer
  FROM
    T
  , LATERAL(
      WITH wrap AS ( -- "материализуем" record
        SELECT
          CASE
            -- если все-таки "перешагнули" через 2-ю запись
            WHEN NOT T.not_cross
              -- то нужная запись - первая из спписка
              THEN T.list[1]
            ELSE ( -- если не пересекли, то ключ остался как в предыдущей записи - отталкиваемся от нее
              SELECT
                _r
              FROM
                task _r
              WHERE
                owner_id = (rv).owner_id AND
                (task_date, id) > ((rv).task_date, (rv).id)
              ORDER BY
                task_date, id
              LIMIT 1
            )
          END _r
      )
      SELECT
        _r
      , CASE
          -- если 2-й записи уже нет в списке, но мы хоть что-то нашли
          WHEN list[2] IS NULL AND _r IS DISTINCT FROM NULL THEN
            TRUE
          ELSE -- ничего не нашли или "перешагнули"
            coalesce(((_r).task_date, (_r).id) < ((list[2]).task_date, (list[2]).id), FALSE)
        END not_cross
      FROM
        wrap
    ) X
  WHERE
    T.size < 20 AND -- ограничиваем тут количество
    T.list IS DISTINCT FROM '{}' -- или пока список не кончился
)
-- #3 : "разворачиваем" записи - порядок гарантирован по построению
SELECT
  (rv).*
FROM
  T
WHERE
  not_cross; -- берем только "непересекающие" записи

[Aspice explain.tensor.ru]

Sic, we operatus est L% data legit ad XX% supplicium tempus. Hoc est, si ratio habenda est credere lectionem posse esse longam (exempli gratia, notitia saepe in cella non est, et ad discum pro eo ire necesse est), tunc hoc modo a lectione minus pendere potes.

In casu, exsecutio tempus melius evasit quam in optione prima "insipientiae". Sed quid horum 3 bene utendum est ad vos.

Source: www.habr.com

SQL HowTo: scribe directe in interrogatione vel "elementarii tres modos"

Sicut auditum est, sic scriptum est

unnest + ARRY

recursive accumulator

Gradus I: Committitur List

Gradus 2: invenire "proximum" records

Gradus III, Filtering et Expanding Records

Add a comment Отменить ответ