Reputation: 622
I have the following JSON format stored as JSONB column in postgresql.
{DisplayName":"Bob marley","FName":"Bob","Title":null,"LName":"Marley","State":null,"EmailAddresses":["[email protected]","[email protected]"]},
{DisplayName":"Bob martin","FName":"Bob ","Title":null,"LName":"Martin","State":null,"EmailAddresses":["[email protected]","[email protected]"]}
I want to query for EmailAddresses Array where it starts with Bob. I am able to query for normal string type keys using the LIKE and ILIKE operators, however for searching inside ARRAY types, is posing a problem. Please advise.
Upvotes: 0
Views: 97
Reputation: 3266
If you want to check if any element of array in jsonb field match to your mask:
SELECT *
FROM jtable
WHERE EXISTS (
SELECT 1
FROM jsonb_array_elements_text(jfield->'EmailAddresses') AS j
WHERE j ~ 'bob'
)
jfield
----------------------------------------------------------------------------------------------------------------------------------------------------------------------
{"FName": "Bob", "LName": "Marley", "State": null, "Title": null, "DisplayName": "Bob marley", "EmailAddresses": ["[email protected]", "[email protected]"]}
{"FName": "Bob ", "LName": "Martin", "State": null, "Title": null, "DisplayName": "Bob martin", "EmailAddresses": ["[email protected]", "[email protected]"]}
(2 rows)
Or if you want list of email that match to mask:
WITH w AS (
SELECT jsonb_array_elements_text(jfield->'EmailAddresses') AS emails,
*
FROM jtable
)
SELECT * FROM w WHERE emails ~ 'bobm'
emails | jfield
---------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------
[email protected] | {"FName": "Bob", "LName": "Marley", "State": null, "Title": null, "DisplayName": "Bob marley", "EmailAddresses": ["[email protected]", "[email protected]"]}
[email protected] | {"FName": "Bob ", "LName": "Martin", "State": null, "Title": null, "DisplayName": "Bob martin", "EmailAddresses": ["[email protected]", "[email protected]"]}
(2 rows)
Upvotes: 1