从PostgreSQL 14开始，你可以使用jsonb下标直接访问JSON字段的元素，并最终更新它们。

UPDATE test SET data['a'] = '5' WHERE data['b'] = '2';

遗憾的是，我没有在文档中找到任何东西，但您可以使用一些变通方法，例如您可以编写一些扩展函数。

例如，在Python中:

CREATE or REPLACE FUNCTION json_update(data json, key text, value json)
returns json
as $$
from json import loads, dumps
if key is None: return data
js = loads(data)
js[key] = value
return dumps(js)
$$ language plpython3u

然后

update test set data=json_update(data, 'a', to_json(5)) where data->>'b' = '2';

UPDATE test
SET data = data::jsonb - 'a' || '{"a":5}'::jsonb
WHERE data->>'b' = '2'

这似乎在PostgreSQL 9.5上有效

select * from pg_settings where name = 'deadlock_timeout';
begin;
create temp table a2(data jsonb);
insert into a2 values('{
    "key0": {
        "key01": "2018-05-06T12:36:11.916761+00:00",
        "key02": "DEFAULT_WEB_CONFIGURATION",
    "key1": {
        "key11": "Data System",
        "key12": "<p>Health,<p>my address<p>USA",
        "key13": "*Please refer to main screen labeling"
    }
}}'::jsonb);
commit;

嵌套的jsonb结构更新。可应用于删除。

update a2 set data =
    data::jsonb #-  '{key0, key1, key13}'
        || '{"key13":"screen labeling"}'::jsonb
returning *;

更新:在PostgreSQL 9.5中，在PostgreSQL本身中有一些jsonb操作功能(但没有json;需要类型转换来操作json值)。

合并2个(或更多)JSON对象(或连接数组):

SELECT jsonb '{"a":1}' || jsonb '{"b":2}', -- will yield jsonb '{"a":1,"b":2}'
       jsonb '["a",1]' || jsonb '["b",2]'  -- will yield jsonb '["a",1,"b",2]'

所以，设置一个简单的键可以使用:

SELECT jsonb '{"a":1}' || jsonb_build_object('<key>', '<value>')

其中<key>应该是字符串，<value>可以是to_jsonb()接受的任何类型。

要在JSON层次结构中设置深层值，可以使用jsonb_set()函数:

SELECT jsonb_set('{"a":[null,{"b":[]}]}', '{a,1,b,0}', jsonb '{"c":3}')
-- will yield jsonb '{"a":[null,{"b":[{"c":3}]}]}'

jsonb_set()的完整参数列表:

jsonb_set(target         jsonb,
          path           text[],
          new_value      jsonb,
          create_missing boolean default true)

path也可以包含JSON数组的索引&出现在那里的负整数从JSON数组的末尾开始计数。然而，一个不存在但为正的JSON数组索引将把元素追加到数组的末尾:

SELECT jsonb_set('{"a":[null,{"b":[1,2]}]}', '{a,1,b,1000}', jsonb '3', true)
-- will yield jsonb '{"a":[null,{"b":[1,2,3]}]}'

对于插入JSON数组(同时保留所有原始值)，可以使用jsonb_insert()函数(在9.6+;本节只提供这个函数):

SELECT jsonb_insert('{"a":[null,{"b":[1]}]}', '{a,1,b,0}', jsonb '2')
-- will yield jsonb '{"a":[null,{"b":[2,1]}]}', and
SELECT jsonb_insert('{"a":[null,{"b":[1]}]}', '{a,1,b,0}', jsonb '2', true)
-- will yield jsonb '{"a":[null,{"b":[1,2]}]}'

jsonb_insert()的完整参数列表:

jsonb_insert(target       jsonb,
             path         text[],
             new_value    jsonb,
             insert_after boolean default false)

同样，出现在JSON数组末尾的路径计数中的负整数。

所以,f.ex。附加到JSON数组的末尾可以使用:

SELECT jsonb_insert('{"a":[null,{"b":[1,2]}]}', '{a,1,b,-1}', jsonb '3', true)
-- will yield jsonb '{"a":[null,{"b":[1,2,3]}]}', and

但是，当target中的路径是JSON对象的键时，这个函数的工作方式与jsonb_set()略有不同。在这种情况下，它只会在不使用键时为JSON对象添加一个新的键-值对。如果使用了它，它将引发一个错误:

SELECT jsonb_insert('{"a":[null,{"b":[1]}]}', '{a,1,c}', jsonb '[2]')
-- will yield jsonb '{"a":[null,{"b":[1],"c":[2]}]}', but
SELECT jsonb_insert('{"a":[null,{"b":[1]}]}', '{a,1,b}', jsonb '[2]')
-- will raise SQLSTATE 22023 (invalid_parameter_value): cannot replace existing key

从JSON对象(或数组)中删除键(或索引)可以使用-操作符:

SELECT jsonb '{"a":1,"b":2}' - 'a', -- will yield jsonb '{"b":2}'
       jsonb '["a",1,"b",2]' - 1    -- will yield jsonb '["a","b",2]'

从JSON层次结构深处删除可以用#-操作符完成:

SELECT '{"a":[null,{"b":[3.14]}]}' #- '{a,1,b,0}'
-- will yield jsonb '{"a":[null,{"b":[]}]}'

对于9.4，您可以使用原始答案的修改版本(如下所示)，但是您可以使用json_object_agg()直接将其聚合为JSON对象，而不是聚合JSON字符串。

原来的答案:这是可能的(没有plpython或plv8)在纯SQL(但需要9.3+，将不能与9.2)

CREATE OR REPLACE FUNCTION "json_object_set_key"(
  "json"          json,
  "key_to_set"    TEXT,
  "value_to_set"  anyelement
)
  RETURNS json
  LANGUAGE sql
  IMMUTABLE
  STRICT
AS $function$
SELECT concat('{', string_agg(to_json("key") || ':' || "value", ','), '}')::json
  FROM (SELECT *
          FROM json_each("json")
         WHERE "key" <> "key_to_set"
         UNION ALL
        SELECT "key_to_set", to_json("value_to_set")) AS "fields"
$function$;

SQLFiddle

编辑:

一个版本，它设置多个键和值:

CREATE OR REPLACE FUNCTION "json_object_set_keys"(
  "json"          json,
  "keys_to_set"   TEXT[],
  "values_to_set" anyarray
)
  RETURNS json
  LANGUAGE sql
  IMMUTABLE
  STRICT
AS $function$
SELECT concat('{', string_agg(to_json("key") || ':' || "value", ','), '}')::json
  FROM (SELECT *
          FROM json_each("json")
         WHERE "key" <> ALL ("keys_to_set")
         UNION ALL
        SELECT DISTINCT ON ("keys_to_set"["index"])
               "keys_to_set"["index"],
               CASE
                 WHEN "values_to_set"["index"] IS NULL THEN 'null'::json
                 ELSE to_json("values_to_set"["index"])
               END
          FROM generate_subscripts("keys_to_set", 1) AS "keys"("index")
          JOIN generate_subscripts("values_to_set", 1) AS "values"("index")
         USING ("index")) AS "fields"
$function$;

编辑2:正如@ErwinBrandstetter指出的那样，上述函数的工作方式类似于所谓的UPSERT(如果字段存在，则更新字段，如果字段不存在则插入字段)。这是一个变体，它只更新:

CREATE OR REPLACE FUNCTION "json_object_update_key"(
  "json"          json,
  "key_to_set"    TEXT,
  "value_to_set"  anyelement
)
  RETURNS json
  LANGUAGE sql
  IMMUTABLE
  STRICT
AS $function$
SELECT CASE
  WHEN ("json" -> "key_to_set") IS NULL THEN "json"
  ELSE (SELECT concat('{', string_agg(to_json("key") || ':' || "value", ','), '}')
          FROM (SELECT *
                  FROM json_each("json")
                 WHERE "key" <> "key_to_set"
                 UNION ALL
                SELECT "key_to_set", to_json("value_to_set")) AS "fields")::json
END
$function$;

编辑3:这里是递归变量，它可以设置(UPSERT)一个叶值(并使用这个答案中的第一个函数)，位于键路径(其中键只能指向内部对象，不支持内部数组):

CREATE OR REPLACE FUNCTION "json_object_set_path"(
  "json"          json,
  "key_path"      TEXT[],
  "value_to_set"  anyelement
)
  RETURNS json
  LANGUAGE sql
  IMMUTABLE
  STRICT
AS $function$
SELECT CASE COALESCE(array_length("key_path", 1), 0)
         WHEN 0 THEN to_json("value_to_set")
         WHEN 1 THEN "json_object_set_key"("json", "key_path"[l], "value_to_set")
         ELSE "json_object_set_key"(
           "json",
           "key_path"[l],
           "json_object_set_path"(
             COALESCE(NULLIF(("json" -> "key_path"[l])::text, 'null'), '{}')::json,
             "key_path"[l+1:u],
             "value_to_set"
           )
         )
       END
  FROM array_lower("key_path", 1) l,
       array_upper("key_path", 1) u
$function$;

更新:增加了用另一个给定键替换现有json字段的键的函数。对于在迁移或其他场景(如数据结构修改)中更新数据类型非常方便。

CREATE OR REPLACE FUNCTION json_object_replace_key(
    json_value json,
    existing_key text,
    desired_key text)
  RETURNS json AS
$BODY$
SELECT COALESCE(
(
    SELECT ('{' || string_agg(to_json(key) || ':' || value, ',') || '}')
    FROM (
        SELECT *
        FROM json_each(json_value)
        WHERE key <> existing_key
        UNION ALL
        SELECT desired_key, json_value -> existing_key
    ) AS "fields"
    -- WHERE value IS NOT NULL (Actually not required as the string_agg with value's being null will "discard" that entry)

),
    '{}'
)::json
$BODY$
  LANGUAGE sql IMMUTABLE STRICT
  COST 100;

更新:函数现在被压缩。

在PostgreSQL 9.4中，我们实现了以下python函数。它也可以用于PostgreSQL 9.3。

create language plpython2u;

create or replace function json_set(jdata jsonb, jpaths jsonb, jvalue jsonb) returns jsonb as $$
import json

a = json.loads(jdata)
b = json.loads(jpaths)

if a.__class__.__name__ != 'dict' and a.__class__.__name__ != 'list':
  raise plpy.Error("The json data must be an object or a string.")

if b.__class__.__name__ != 'list':
   raise plpy.Error("The json path must be an array of paths to traverse.")

c = a
for i in range(0, len(b)):
  p = b[i]
  plpy.notice('p == ' + str(p))

  if i == len(b) - 1:
    c[p] = json.loads(jvalue)

  else:
    if p.__class__.__name__ == 'unicode':
      plpy.notice("Traversing '" + p + "'")
      if c.__class__.__name__ != 'dict':
        raise plpy.Error("  The value here is not a dictionary.")
      else:
        c = c[p]

    if p.__class__.__name__ == 'int':
      plpy.notice("Traversing " + str(p))
      if c.__class__.__name__ != 'list':
        raise plpy.Error("  The value here is not a list.")
      else:
        c = c[p]

    if c is None:
      break    

return json.dumps(a)
$$ language plpython2u ;

使用示例:

create table jsonb_table (jsonb_column jsonb);
insert into jsonb_table values
('{"cars":["Jaguar", {"type":"Unknown","partsList":[12, 34, 56]}, "Atom"]}');

select jsonb_column->'cars'->1->'partsList'->2, jsonb_column from jsonb_table;

update jsonb_table
set jsonb_column = json_set(jsonb_column, '["cars",1,"partsList",2]', '99');

select jsonb_column->'cars'->1->'partsList'->2, jsonb_column from jsonb_table;

Note that for a previous employer, I have written a set of C functions for manipulating JSON data as text (not as a json or jsonb type) for PostgreSQL 7, 8 and 9. For example, extracting data with json_path('{"obj":[12, 34, {"num":-45.67}]}', '$.obj[2]['num']'), setting data with json_path_set('{"obj":[12, 34, {"num":-45.67}]}', '$.obj[2]['num']', '99.87') and so on. It took about 3 days work, so if you need it to run on legacy systems and have the time to spare, it may be worth the effort. I imagine the C version is much faster than the python version.