如何将两个 json 对象合并到 postgres 中的数组中

Question

我有两个 jsonb 值。我想合并它但是我坚持将它组合在一起。这是我的代码

CREATE OR REPLACE FUNCTION emr.azintin(p_msg text)
 RETURNS character varying
 LANGUAGE plpgsql
AS $function$
DECLARE
    v_msgar     text[];
    v_msgln     text;
    v_currusr   afm_usr.user_id%type := 'SYSEXCH';
    v_currdt    afm_usr.insert_date%type := CURRENT_TIMESTAMP;
    macres      azt_macres%rowtype;
    orddet      lbt_orddet%rowtype;
    v_resultid  character varying ;
    v_reading   character varying ;

BEGIN   
         v_msgar := array['OBR||||^^^3972^2102001420||||20210227133222|25||||3
                            OBX|||0^RBC||4.99|^10â¶/Î¼L|3.85-5.78||||F
                            OBX|||1^HGB||14.4|^g/dL|12.0-17.2||||F
                            OBX|||2^MCV||84.5|^fL|78.0-96.0||||F
                            OBX|||3^HCT||42.2|^%|34.8-50.9||||F
                            OBX|||4^MCH||28.90|^pg|26.40-33.20||||F
                            OBX|||5^MCHC||34.2|^g/dL|31.8-36.7|||<EOT>|F
                            OBX|||6^RDWsd||39.2|^fL|0.0-0.0||||F'];
              
            SELECT json_agg(jsonb_build_object('resultValue',split_part(parts[6],'^',1)))
               INTO v_reading
               FROM unnest(v_msgar) as elem,
                     regexp_split_to_array(elem, '\|') as parts
               WHERE parts[1] = 'OBX';
              
            SELECT json_agg(jsonb_build_object('resultId',split_part(parts[4],'^',2)))
               INTO v_resultid
               FROM unnest(v_msgar) as elem,
                     regexp_split_to_array(elem, '\|') as parts
               WHERE parts[1] = 'OBX';
              
            select concat(v_resultid || v_reading) 
               INTO macres.test_result
               FROM unnest(v_msgar) as elem,
                     regexp_split_to_array(elem, '\|') as parts
               WHERE parts[1] = 'OBX';
        
            SELECT split_part(items[5], '^', 5)
                  INTO macres.specimen_id
                FROM (
                  SELECT string_to_array(element, '|') as items
                 FROM unnest(v_msgar) as t(element)) t   
            WHERE items[1] = 'OBR';
    

    END IF;
END
$function$
;

我设法得到 v_reading 和 v_resultid 看起来像这样

 [{"resultValue": "5.20"}, {"resultValue": "14.2"}, {"resultValue": "81.7"}, {"resultValue": "42.5"}, {"resultValue": "27.20"}, {"resultValue": "33.3"}, {"resultValue": "40.6"}, {"resultValue": "15.9"}, {"resultValue": "188"}, {"resultValue": "8.1"}, {"resultValue": "0.15"}, {"resultValue": "9.4"}, {"resultValue": "46.2"}, {"resultValue": "4.6"}, {"resultValue": "1.1"}, {"resultValue": "23.6"}, {"resultValue": "0.4"}, {"resultValue": "8.7"}, {"resultValue": "3.1"}, {"resultValue": "67.7"}, {"resultValue": "56"}, {"resultValue": "105"}, {"resultValue": ""}]

[{"resultId": "RBC"}, {"resultId": "HGB"}, {"resultId": "MCV"}, {"resultId": "HCT"}, {"resultId": "MCH"}, {"resultId": "MCHC"}, {"resultId": "RDWsd"}, {"resultId": "RDWcv"}, {"resultId": "PLT"}, {"resultId": "MPV"}, {"resultId": "PCT"}, {"resultId": "PDWsd"}, {"resultId": "PDWcv"}, {"resultId": "WBC"}, {"resultId": "LYM"}, {"resultId": "LYMP"}, {"resultId": "MID"}, {"resultId": "MIDP"}, {"resultId": "GRA"}, {"resultId": "GRAP"}, {"resultId": "PLCR"}, {"resultId": "PLCC"}, {"resultId": "RBCHistogram"}]

我怎样才能将这两个组合起来 json 看起来像这样

[{"resultid": "RBC", "resultValue": "5.20"}, {"resultid": "HGB", "resultValue": "5.20"}, {"resultid": "MCV", "resultValue": "5.20"}, {"resultid": "HCT", "resultValue": "5.20"}, {"resultid": "MCH", "resultValue": "5.20"}, {"resultid": "MCHC", "resultValue": "5.20"}, {"resultid": "RDWsd", "resultValue": "5.20"}, {"resultid": "RDWcv", "resultValue": "5.20"}, {"resultid": "PLT", "resultValue": "5.20"}, {"resultid": "MPV", "resultValue": "5.20"}, {"resultid": "PCT", "resultValue": "5.20"}, {"resultid": "PDWsd", "resultValue": "5.20"}, {"resultid": "PDWcv", "resultValue": "5.20"}, {"resultid": "WBC", "resultValue": "5.20"}, {"resultid": "LYM", "resultValue": "5.20"}, {"resultid": "LYMP", "resultValue": "5.20"}, {"resultid": "MID", "resultValue": "5.20"}, {"resultid": "MIDP", "resultValue": "5.20"}, {"resultid": "GRA", "resultValue": "5.20"}, {"resultid": "GRAP", "resultValue": "5.20"}, {"resultid": "PLCR", "resultValue": "5.20"}, {"resultid": "PLCC", "resultValue": "5.20"}, {"resultid": "RBCHistogram", "resultValue": "5.20"}]

我可以简单地这样做

            SELECT json_agg(jsonb_build_object('resultid',split_part(parts[4],'^',2), 'resultValue',split_part(parts[6],'^',1)))
               INTO macres.test_result
               FROM unnest(v_msgar) as elem,
                     regexp_split_to_array(elem, '\|') as parts
               WHERE parts[1] = 'OBX';

但问题是我需要在 split_part(parts[4],'^',2) 中添加一个条件，这就是为什么我将它分成 v_reading 和 v_resultid .

我要放条件

IF v_resultid = 'MCHC' THEN
   v_reading = null
END ID

但如果数组找到 MCHC，它仍然以某种方式读取读数。我想从 json

中删除 MCHC

Answer 1

加入两个 JSON 个对象：

demo:db<>fiddle

SELECT
    jsonb_agg(a.element || b.element)                        -- 4
FROM (
    SELECT * FROM jsonb_array_elements(                      -- 1
        '[{"resultValue": "5.20"}, {"resultValue": "14.2"}, {"resultValue": "81.7"}, {"resultValue": "42.5"}, {"resultValue": "27.20"}, {"resultValue": "33.3"}, {"resultValue": "40.6"}, {"resultValue": "15.9"}, {"resultValue": "188"}, {"resultValue": "8.1"}, {"resultValue": "0.15"}, {"resultValue": "9.4"}, {"resultValue": "46.2"}, {"resultValue": "4.6"}, {"resultValue": "1.1"}, {"resultValue": "23.6"}, {"resultValue": "0.4"}, {"resultValue": "8.7"}, {"resultValue": "3.1"}, {"resultValue": "67.7"}, {"resultValue": "56"}, {"resultValue": "105"}, {"resultValue": ""}]'::jsonb
    ) WITH ORDINALITY as a(element, index)                   -- 2
) a
JOIN (
    SELECT * FROM jsonb_array_elements(
        '[{"resultId": "RBC"}, {"resultId": "HGB"}, {"resultId": "MCV"}, {"resultId": "HCT"}, {"resultId": "MCH"}, {"resultId": "MCHC"}, {"resultId": "RDWsd"}, {"resultId": "RDWcv"}, {"resultId": "PLT"}, {"resultId": "MPV"}, {"resultId": "PCT"}, {"resultId": "PDWsd"}, {"resultId": "PDWcv"}, {"resultId": "WBC"}, {"resultId": "LYM"}, {"resultId": "LYMP"}, {"resultId": "MID"}, {"resultId": "MIDP"}, {"resultId": "GRA"}, {"resultId": "GRAP"}, {"resultId": "PLCR"}, {"resultId": "PLCC"}, {"resultId": "RBCHistogram"}]'::jsonb
    ) WITH ORDINALITY as b(element, index)
) b
ON a.index = b.index                                         -- 3

将 JSON 个数组提取为每个元素一行。
WITH ORDINALITY为每条展开的记录添加行数，以标识数组中的原始位置
将两个提取的 JSON 元素集加入索引
使用 || 运算符合并两个 JSON 列（仅对类型 jsonb 有效，对类型 json 无效，但仍应使用）。最后，您可以将这些合并的对象聚合到一个新的 JSON 数组中。

如何将两个 json 对象合并到 postgres 中的数组中

How to merge two json object into an array in postgres

postgresql

merge

json

stored-procedures

jsonb