JSON

JSON_EXTRACT_PATH_TEXT

DuckDB

Seperti JSON_EXTRACT_PATH tapi mengembalikan hasil sebagai VARCHAR, bukan JSON.

Tipe hasil: VARCHARDiperbarui: 6 Jan 2026

Syntax

SQL

JSON_EXTRACT_PATH_TEXT(json, path1, path2, ...)

Parameter

jsonJSON/VARCHARwajib

JSON object sumber

path1, path2, ...VARCHARwajib

Sequence of keys untuk navigate ke nilai

Contoh Penggunaan

Extract sebagai Text

SQL

1 SELECT
2   JSON_EXTRACT_PATH('{"name": "Budi"}', 'name') as json_result,
3   JSON_EXTRACT_PATH_TEXT('{"name": "Budi"}', 'name') as text_result;

Perbedaan hasil JSON vs TEXT.

Hasil

json_result	text_result
"Budi"	Budi

Langsung Gunakan untuk String Ops

SQL

1 SELECT
2   id,
3   UPPER(JSON_EXTRACT_PATH_TEXT(data, 'user', 'name')) as name_upper,
4   JSON_EXTRACT_PATH_TEXT(data, 'user', 'email') as email
5 FROM users_json;

Hasil bisa langsung dipakai untuk string functions.

Hasil

id	name_upper	email
1	BUDI	budi@email.com
2	ANI	ani@email.com

Filter Berdasarkan JSON Value

SQL

1 SELECT *
2 FROM products_json
3 WHERE JSON_EXTRACT_PATH_TEXT(metadata, 'category') = 'Electronics'
4   AND CAST(JSON_EXTRACT_PATH_TEXT(metadata, 'price') AS INTEGER) < 1000000;

Gunakan TEXT version untuk comparison dan filtering.

Hasil

id	name	metadata
1	Phone	{"category":"Electronics","price":"800000"}
3	Tablet	{"category":"Electronics","price":"500000"}

Fungsi Terkait

JSON_EXTRACT_PATHJSON_EXTRACT

Kembali ke DuckDB Functions

1	SELECT
2	JSON_EXTRACT_PATH('{"name": "Budi"}', 'name') as json_result,
3	JSON_EXTRACT_PATH_TEXT('{"name": "Budi"}', 'name') as text_result;

1	SELECT
2	id,
3	UPPER(JSON_EXTRACT_PATH_TEXT(data, 'user', 'name')) as name_upper,
4	JSON_EXTRACT_PATH_TEXT(data, 'user', 'email') as email
5	FROM users_json;

1	SELECT *
2	FROM products_json
3	WHERE JSON_EXTRACT_PATH_TEXT(metadata, 'category') = 'Electronics'
4	AND CAST(JSON_EXTRACT_PATH_TEXT(metadata, 'price') AS INTEGER) < 1000000;