Recursive exact matching
Walk nested objects and arrays while changing only values whose complete field name matches.
Batch edit values across nested JSON: multiply prices or other numbers, control decimal precision, and safely rename keys.
Match numeric values by exact field name. Multiplication runs first, then results are rounded to optional decimal places; JSON numbers do not retain trailing zeroes.
When enabled, safely replace matching keys at every nested level.
Processed locally in your browser
Walk nested objects and arrays while changing only values whose complete field name matches.
Handle numbers and strict numeric strings with rounding from 0 to 100 decimal places; practical precision is limited by JavaScript number representation.
Calculate before renaming, and stop without partial output when rules duplicate or keys collide.
Multiply integer prices in nested product data by 0.01 and round to at most two decimal places; JSON numbers do not retain trailing zeroes.
Adjust prices, tax rates, percentages, or other numeric fields while controlling rounding.
Replace an old API key throughout a JSON payload when a schema changes.
Normalize deep arrays and objects before sending exported data to another system.
Enter a field, multiplier, and optional rounding decimals. Enable renaming only when you need to change keys.
Paste the source, upload one local .json file, or load the example to check the default rule.
Run the editor, review the summary and output, then copy or download the formatted JSON.
No. The transformation runs in this browser and does not require a Toop processing API call.
Yes. It recursively walks arrays and objects and matches complete field names at every level.
Yes. Strict numeric strings such as "1299" or "12.5" become JSON numbers. Values such as "12px", empty strings, booleans, and null are skipped. JSON numbers do not retain trailing zeroes, and practical precision is limited by JavaScript number representation.
Calculation runs against the original key first, then that key is renamed, so both rules can work together in one pass.
If the new key would overwrite another field in the same object, processing stops to prevent silent data loss.