JSON Types

JavaScript Object Notation (JSON) data types are used for storing JSON data.

It can be an independent scalar, an array, or a key-value object. An array and an object can be called a container.

1. Scalar: a number, Boolean, string, or null
2. Array: defined in a pair of square brackets ([]), in which elements can be of any JSON data type, and are not necessarily of the same type.
3. Object: defined in a pair of braces ({}), in which objects are stored in the format of key:value. Each key must be a string enclosed in double quotation marks (""), and its value can be of any JSON data type. In case of duplicate keys, the last key-value pair will be used.

The GaussDB(DWS) memory has two data types: JSON and JSONB, which can be used to store JSON data. JSON data is a complete copy of the input strings, retaining the entered spaces, duplicate keys, and sequence, while JSONB stores data in a decomposed binary form, removing semantic-irrelevant details and duplicate keys, and sorting key-values. Therefore, JSONB data does not need to be parsed.

Both JSON and JSONB are of JSON data type, and the same strings can be entered as input. The main difference between JSON and JSONB is the efficiency. Because JSON data is an exact copy of the input text, the data must be parsed on every execution. In contrast, JSONB data is stored in a decomposed binary form and can be processed faster, though this makes it slightly slower to input due to the conversion mechanism. In addition, because the JSONB data form is normalized, it supports more operations, for example, comparing sizes according to a specific rule. JSONB also supports indexing, which is a significant advantage.

Null (null-json): Only null is supported, and all letters are in lowercase.

1
2

select 'null'::json;   -- suc
select 'NULL'::jsonb;  -- err

Number (num-json): The value can be a positive or negative integer, decimal fraction, or 0. The scientific notation is supported.

1
2
3
4

select '1'::json;
select '-1.5'::json;
select '-1.5e-5'::jsonb, '-1.5e+2'::jsonb;
select '001'::json, '+15'::json, 'NaN'::json;  -- Redundant leading zeros, plus signs (+), NaN, and infinity are not supported.

Boolean (bool-json): The value can only be true or false in lowercase.

1
2

select 'true'::json;
select 'false'::jsonb;  

String (str-json): The value must be a string enclosed in double quotation marks ("").

1
2

select '"a"'::json;
select '"abc"'::jsonb; 

Array (array-json): Arrays are enclosed in square brackets ([]). Elements in the array can be any valid JSON data, and are unnecessarily of the same type.

1
2
3

select '[1, 2, "foo", null]'::json;
select '[]'::json;
select '[1, 2, "foo", null, [[]], {}]'::jsonb; 

Object (object-json): The value is enclosed in braces ({}). The key must be a JSON-compliant string, and the value can be any valid JSON string.

1
2
3

select '{}'::json;
select '{"a": 1, "b": {"a": 2,  "b": null}}'::json;
select '{"foo": [true, "bar"], "tags": {"a": 1, "b": null}}'::jsonb;  

• Note that 'null'::json and null::json are two different concepts. The difference is similar to that between the strings str="" and str=null.
• For numbers, when scientific notation is used, JSONB expands them, while JSON stores an exact copy of the input text.

Compares sizes

Format normalization ensures that only one form of JSONB data exists in the same semantics. Therefore, sizes can be compared according to a specific rule.

1. First, type comparison: object-jsonb > array-jsonb > bool-jsonb > num-jsonb > str-jsonb > null-jsonb
2. Content is compared if the data type is the same:
   • str-json: The default text sorting rule of the database is used for comparison. A positive value indicates greater than, a negative value indicates less than, and 0 indicates equal.
   • num-json: numeric comparison
   • bool-json: true > false
   • array-jsonb: long elements > short elements. If the lengths are the same, compare each element in sequence.
   • object-jsonb: long key-value pairs > short key-value pairs. If the lengths are the same, compare each key-value pair in sequence, first the key and then the value.

For comparison within the object-jsonb type, the final result after format sorting is used for comparison. Therefore, the comparison result may not be so intuitive as direct input.

Creates an index

B-Tree and GIN indexes can be created for the JSONB type.

Includes elements in a JSON

An important capability of JSONB is to query whether a JSON contains some elements or whether some elements exist in a JSON.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17

-- Simple scalar/original values contain only the same value:
SELECT '"foo"'::jsonb @> '"foo"'::jsonb;

-- The array on the left contains the string on the right.
SELECT '[1, "aa", 3]'::jsonb ? 'aa';

-- The array on the left contains all elements of the array on the right. The sequence and repetition are not important.
SELECT '[1, 2, 3]'::jsonb @> '[1, 3, 1]'::jsonb;

-- The object-json on the left contains all key-values of the object-json on the right.
SELECT '{"product": "PostgreSQL", "version": 9.4, "jsonb":true}'::jsonb @> '{"version":9.4}'::jsonb;

-- The array on the left does not contain all elements in the array on the right, because the three elements in the array on the left are 1, 2, and [1,3], and the elements on the right are 1 and 3.
SELECT '[1, 2, [1, 3]]'::jsonb @> '[1, 3]'::jsonb; -- false

-- The array on the right does not contain all elements in the array on the left in the following example:
SELECT '{"foo": {"bar": "baz"}}'::jsonb @> '{"bar": "baz"}'::jsonb; -- false