I have a jsonb field with an array like this one below:
[
{
"type":"discount",
"title":"Discount 10%"
},
{
"file":"zx5rP8EoacyfhqGndcSOnP8VYtkr9Ya8Nvf7oYL98YDsM1CLMYIurYvfVUU4AGkzBsovwssT0bq.pdf",
"type":"menu",
"title":"Some menu title etc"
}
]
I want to get the file attribute in case there is a type=menu in the array.
What I managed to do is to know if there is one, but how can I eventually extract the file value?
case when offers #> '[{"type":"menu"}]' then true else false end
I don't want to do something like this below because the array may not contain a discount type.
offers->1->'file'
Use jsob_array_elements() and ->> operator (see JSON Functions and Operators.)
with a_table(json_col) as (
values (
'[
{
"type":"discount",
"title":"Discount 10%"
},
{
"file":"zx5rP8EoacyfhqGndcSOnP8VYtkr9Ya8Nvf7oYL98YDsM1CLMYIurYvfVUU4AGkzBsovwssT0bq.pdf",
"type":"menu",
"title":"Some menu title etc"
}
]'::jsonb)
)
select value->>'file' as filename
from a_table,
lateral jsonb_array_elements(json_col)
where value->>'type' = 'menu'
filename
---------------------------------------------------------------------------------
zx5rP8EoacyfhqGndcSOnP8VYtkr9Ya8Nvf7oYL98YDsM1CLMYIurYvfVUU4AGkzBsovwssT0bq.pdf
(1 row)
Eg:
t=# with a as (with v as (select '[
{
"type":"discount",
"title":"Discount 10%"
},
{
"file":"zx5rP8EoacyfhqGndcSOnP8VYtkr9Ya8Nvf7oYL98YDsM1CLMYIurYvfVUU4AGkzBsovwssT0bq.pdf",
"type":"menu",
"title":"Some menu title etc"
}
]'::jsonb j)
select jsonb_array_elements(j) r from v) select r->>'file' from a where r->>'type' = 'menu';
?column?
---------------------------------------------------------------------------------
zx5rP8EoacyfhqGndcSOnP8VYtkr9Ya8Nvf7oYL98YDsM1CLMYIurYvfVUU4AGkzBsovwssT0bq.pdf
(1 row)
Related
I have to query a table with 2 columns, id and content. Id is just a uuid and the content column looks like
{
"fields": [
{
"001": "mig00004139229"
},
{
"856": {
"ind1": " ",
"ind2": " ",
"subfields": [
{
"u": "https://some.domain.com"
},
{
"z": "some text"
}
]
}
},
{
"999": {
"subfields": [
{
"i": "81be1acf-11df-4d13-a5c6-4838e3a808ee"
},
{
"s": "3a6aa357-8fd6-4451-aedc-13453c1f2296"
}
]
}
}
]
}
I need to select the id, 001, and 856 elements where the subfield "u" domain matches a string "domain.com" so the output would be
id
001
856
81be1acf-11df-4d13-a5c6-4838e3a808ee
mig00004139229
https://some.domain.com
If this were a flat table, the query would correspond with "select id, 001, 856 from table where 856 like '%domain.com%'"
I can select individual columns based on the criteria I need, but they appear in separate rows except the id which appears with any other individual field in a regular select statement. How would I get the other fields to appear in the same row since it's part of the same record?
Unfortunately, my postgres version doesn't support jsonb_path_query, so I've been trying something along the lines of:
SELECT id, jsonb_array_elements(content -> 'fields') -> '001',
jsonb_array_elements(content -> 'fields') -> '856' -> 'subfields'
FROM
mytable
WHERE....
This method returns the data I need, but the individual elements arrive on separate rows with the with the id in the first column and nulls for every element that is neither the 001 nor 856 e.g.
id
001
856
id_for_first_record
001_first_record
null
id_for_first_record
null
null
id_for_first_record
null
null
id_for_first_record
null
856_first_record
id_for_second_record
001_second_record
null
id_for_second_record
null
null
id_for_second_record
null
null
id_for_second_record
null
856_second_record
Usable, but clunky so I'm looking for something better
I think my query can help you. There are different ways to resolve this, I am not sure if this is the best approach.
I use jsonb_path_query() function with the path for the specified JSON value.
SELECT
id,
jsonb_path_query(content, '$.fields[*]."001"') AS "001",
jsonb_path_query(content, '$.fields[*]."856".subfields[*].u') AS "856"
FROM t
WHERE jsonb_path_query_first(content, '$.fields[*]."856".subfields[*].u' )::text ilike '%domain%';
Output:
id
001
856
81be1acf-11df-4d13-a5c6-4838e3a808ee
"mig00004139229"
"https://some.domain.com"
UPDATED: because of Postgresql version is prior to 12.
You could try something like this, but I think there must be a better approach:
SELECT
t.id,
max(sq1."001") AS "001",
max(sq2."856") AS "856"
FROM t
INNER JOIN (SELECT id, (jsonb_array_elements(content -> 'fields') -> '001')::text AS "001" FROM t) AS sq1 ON t.id = sq1.id
INNER JOIN (SELECT id, (jsonb_array_elements(jsonb_array_elements(content -> 'fields') -> '856' -> 'subfields') -> 'u')::text AS "856" FROM t) AS sq2 ON t.id = sq2.id
WHERE sq2."856" ilike '%domain%'
GROUP BY t.id;
In a Postgres (9+) table there is a column of type jsonb with the following json:
{
"dynamicFields":[
{
"name":"040",
"subfields":[
{
"name":"a",
"value":"abc"
},
{
"name":"a",
"value":"xyz"
}
]
}
]
}
I would like to write a query that return only the rows where the field name equals 040 and subfield a equals xyz.
This is as far as I got, so far:
select e.obj from my_table
cross join lateral jsonb_array_elements(my_column-> 'dynamicFields') as e(obj)
where e.obj ->> 'name' = '040' and e.obj ->> 'subfields' #> '{"name": "a", "value": "xyz"}'::jsonb
How should this query be to achieve this?
e.obj ->> 'subfields' has a text result. You'll want to use e.obj -> 'subfields' that returns the jsonb value where the #> operator works. Also the containment checks needs to have another array as the right hand side, so that it will test whether all values in the right array are contained in the left array - it doesn't work to pass the element object directly.
select e.obj from my_table
cross join lateral jsonb_array_elements(my_column-> 'dynamicFields') as e(obj)
where e.obj ->> 'name' = '040' and e.obj -> 'subfields' #> '[{"name": "a", "value": "xyz"}]'::jsonb
-- ^ ^ ^
(online demo)
As you have an equality condition you can use the "contains" operator directly by providing a JSON value of what you want. There is no need to unnest the arrays.
select *
from my_table
where my_column -> 'dynamicFields' #> '[{"name": "040", "subfields": [{"name":"a", "value": "xyz"}]}]'
Starting with Postgres 12 an alternative is to a SQL/JSON path operator:
select *
from my_table
where my_column #? '$.dynamicFields[*] ? (#.name == "040").subfields[*] ? (#.name == "a" && #.value == "xyz")'
I have a column in my Metabase table where the column entry is like the following:
{ “text_fields”: { “Weight”: “{:optional=>true, :priority=>4, :index=>false}” }, “checkbox_fields”: {}, “dropdown_fields”: { “Brand Name”: “{:optional=>false, :priority=>1, :index=>false, :options=>[“Non Branded”]}” }}
I want to get a net count of
text_fields
checkbox_fields
dropdown_fields
The desired answer, in this case, will be: 2 (1 text field + 0 checkbox field + 1 dropdown field)
Use jsonb_each to iterate over object keys/values, and jsonb_object_keys to extract the keys only.
Example (with sample data included):
SELECT * FROM mytable ;
mycolumn
---------------------------------------------------------------------------------------------
{"text_fields": {"Weight": 1}, "checkbox_fields": {}, "dropdown_fields": {"Brand Name": 1}}
(1 row)
SELECT *,
(SELECT sum((SELECT count(*) FROM jsonb_object_keys(v1)))
FROM jsonb_each(mycolumn) AS j1(k1,v1)
WHERE k1 IN ('text_fields', 'checkbox_fields', 'dropdown_fields')
) AS mytotal
FROM mytable;
mycolumn | mytotal
---------------------------------------------------------------------------------------------+---------
{"text_fields": {"Weight": 1}, "checkbox_fields": {}, "dropdown_fields": {"Brand Name": 1}} | 2
(1 row)
Assuming the following jsonb data :
{
"3f210b": {
"hash": "e60ab24c549ac7060fzedfefe563e8493d759bb"
},
"Siaa1b3": {
"hash": "d24b37efgregr1a2c6db3b9334b3bf4fef3f22bfc9a43f"
}
}
Is it possible, in PostgreSQL, to do a deep jsonb search without know the key (3f210b & Siaa1b3) but we know the hash.
With Postgres 11 you will need to unnest that value:
select *
from the_table t
where exists (select *
from jsonb_each(t.the_column) as x(ky, item)
where x.item ->> 'hash' = 'd24b37efgregr1a2c6db3b9334b3bf4fef3f22bfc9a43f');
With Postgres 12, you could use the new JSON/Path expression:
select *
from the_table t
where jsonb_path_exists(the_column, '$.*.hash ? (# == "d24b37efgregr1a2c6db3b9334b3bf4fef3f22bfc9a43f")')
I am trying to build a recursive CTE in Postgres that supports both arrays and objects, to return a list of key-value pairs and don't seem to be able to find a good example. This is my current code.
with recursive jsonRecurse as
(
select
j.key as Path
,j.key
,j.value
from jsonb_each(to_jsonb('{
"key1": {
"key2": [
{
"key3": "test3",
"key4": "test4"
}
]
},
"key5": [
{
"key6":
[
{
"key7": "test7"
}
]
}
]
}'::jsonb)) j
union all
select
jr.path || '.' || jr2.Key
,jr2.key
,jr2.value
from jsonRecurse jr
left join lateral jsonb_each(jr.value) jr2 on true
where jsonb_typeof(jr.value) = 'object'
)
select
*
from jsonRecurse;
As you can see the code stops recursing as soon as I hit an array instead of an object. I've tried playing around with using a case statement and putting the function call to jsonb_each or jsonb_array_element in the case statement instead but I get an error telling me to use lateral joins instead.
I have used this example table to make the query more readable:
create table my_table(id serial primary key, jdata jsonb);
insert into my_table (jdata) values
('{
"key1": {
"key2": [
{
"key3": "test3",
"key4": "test4"
}
]
},
"key5": [
{
"key6":
[
{
"key7": "test7"
}
]
}
]
}');
You have to join both jsonb_each(value) and jsonb_array_elements(value) conditionally, depending on the type of value:
with recursive extract_all as
(
select
key as path,
value
from my_table
cross join lateral jsonb_each(jdata)
union all
select
path || '.' || coalesce(obj_key, (arr_key- 1)::text),
coalesce(obj_value, arr_value)
from extract_all
left join lateral
jsonb_each(case jsonb_typeof(value) when 'object' then value end)
as o(obj_key, obj_value)
on jsonb_typeof(value) = 'object'
left join lateral
jsonb_array_elements(case jsonb_typeof(value) when 'array' then value end)
with ordinality as a(arr_value, arr_key)
on jsonb_typeof(value) = 'array'
where obj_key is not null or arr_key is not null
)
select *
from extract_all;
Output:
path | value
--------------------+------------------------------------------------
key1 | {"key2": [{"key3": "test3", "key4": "test4"}]}
key5 | [{"key6": [{"key7": "test7"}]}]
key1.key2 | [{"key3": "test3", "key4": "test4"}]
key5.0 | {"key6": [{"key7": "test7"}]}
key1.key2.0 | {"key3": "test3", "key4": "test4"}
key5.0.key6 | [{"key7": "test7"}]
key1.key2.0.key3 | "test3"
key1.key2.0.key4 | "test4"
key5.0.key6.0 | {"key7": "test7"}
key5.0.key6.0.key7 | "test7"
(10 rows)
Elements of json arrays have no keys, we should use their indexes to build a path. Therefore the function jsonb_array_elements() should be called with ordinality. Per the documentation (see 7.2.1.4. Table Functions):
If the WITH ORDINALITY clause is specified, an additional column of type bigint will be added to the function result columns. This column numbers the rows of the function result set, starting from 1.
The function call
jsonb_array_elements(case jsonb_typeof(value) when 'array' then value end)
with ordinality as a(arr_value, arr_key)
returns pairs (value, ordinality) aliased as (arr_value, arr_key).