SQL Challenger
0 of 250 complete

Lessons

0 of 250 complete

Saved in this browser

Lesson 141 of 250

Moderate

Read a JSON field

Your task

Return each receipt ID and the currency in its metadata.

JSON_EXTRACT can read a value from structured text using a path beginning with $.

Sample data

Table:receipts
idsold_onamountchannelmetadata
12025-01-0136 ONLINE {"currency":"EUR","tags":["gift","jazz"]}
22025-01-1622Store{"currency":"EUR","tags":["indie"]}
32025-01-3154 online{"currency":"EUR","tags":["returning","bundle"]}
42025-03-0438STORE {"currency":"EUR","tags":[]}
52025-04-0944Online{"currency":"EUR","tags":["indie"]}

Expected output

idJSON_EXTRACT(metadata,'$.currency')
1EUR
2EUR
3EUR
4EUR
5EUR
Show a hint

Use JSON_EXTRACT(metadata, '$.currency').

Dates, text, and JSON guide

Work with dates as ISO text

The receipt dates use YYYY-MM-DD, so their text order matches calendar order. Use an inclusive start and exclusive end for periods. This query returns receipts in the first quarter, without including April 1.

SELECT id, sold_on, amount
FROM receipts
WHERE sold_on >= '2025-01-01'
  AND sold_on < '2025-04-01';

Normalize values before comparing them

The receipt channel values vary in capitalization and spacing. TRIM removes surrounding spaces and LOWER handles these ASCII names. Use the same normalization when grouping values that represent the same category.

SELECT LOWER(TRIM(channel)) AS channel, COUNT(*) AS receipts
FROM receipts
GROUP BY LOWER(TRIM(channel));

Extract a field or expand an array

JSON_EXTRACT reads a value at a path, such as the currency code. JSON_EACH turns an array into rows with a value column. After expansion, receipt IDs can repeat and a receipt with no tags contributes no rows.

SELECT r.id, JSON_EXTRACT(r.metadata, '$.currency') AS currency,
       tag.value AS tag
FROM receipts AS r
CROSS JOIN JSON_EACH(r.metadata, '$.tags') AS tag;
SQLiteCtrl / ⌘ + Enter

Press Control+Space for SQL suggestions. Use the arrow keys to select an option and Enter to insert it.

Results

Loading SQLite…