JSONL to JSON wraps line-delimited records into one pretty JSON array after flattening each record through the tabular model.
From real runs
| Element | Result | What happens |
|---|---|---|
| Line records | Kept | Each non-empty JSONL line became one object in the output array, preserving source order. |
| Output container | Changed | The result is one bracketed JSON array rather than independent newline-delimited JSON objects. |
| Nested objects | Changed | Nested customer and extra fields flattened to dotted object keys in the final JSON. |
| Arrays | Changed | tags became a string containing JSON array text, so consumers must parse that field again if they need an array. |
| Missing fields | Changed | extra.gift was added to the first output object with a null value because the second line introduced that key. |
| Unicode and string IDs | Kept | The pretty JSON output retained 001, 02134, München, Zoë, and 東京 exactly as decoded values. |
Each non-empty JSONL line became one object in the output array, preserving source order.
The result is one bracketed JSON array rather than independent newline-delimited JSON objects.
Nested customer and extra fields flattened to dotted object keys in the final JSON.
tags became a string containing JSON array text, so consumers must parse that field again if they need an array.
extra.gift was added to the first output object with a null value because the second line introduced that key.
The pretty JSON output retained 001, 02134, München, Zoë, and 東京 exactly as decoded values.
We converted our JSON Lines sample (411 B) with Table engine on 2026-09-11. It took 1 ms and produced a 558 B JSON file, 1.4× larger than the input. The engine raised no warnings.
Input · sample.jsonl · 411 B
4 JSON objects, one per line; keys: company, stage, arr, owner, close_date
Output · sample.json · 558 B
Array of 4 objects; keys: company, stage, arr, owner, close_date
First lines of the output
[
{
"company": "Northwind Traders",
"stage": "Proposal",
"arr": 48000,
"owner": "R. Ayers",
"close_date": "2026-10-14"
},
{
"company": "Contoso Ltd",
"stage": "Negotiation",
"arr": 120000,
"owner": "P. Ng",
"close_date": "2026-09-30"
},
{
"company": "Fabrikam",
"stage": "Qualified",
…Python json module
python - <<'PY'
import json
with open("input.jsonl", encoding="utf-8") as f:
rows = [json.loads(line) for line in f if line.strip()]
with open("output.json", "w", encoding="utf-8") as out:
json.dump(rows, out, ensure_ascii=False, indent=2)
PYThis keeps nested JSON unchanged; tabular-ir flattens records before writing the array.
The HTTP API and the MCP server run the exact engine this page uses (Table engine); the response carries the same warnings the web converter shows. Check support first with GET /api/v1/pairs?from=jsonl&to=json.
curl -X POST https://filetypeconverters.com/api/v1/convert \
-H "Authorization: Bearer $FTC_API_KEY" \
-F "file=@input.jsonl" -F "from=jsonl" -F "to=json" \
-F 'options={"encoding":"utf-8"}' \
--output output.json| Key | Applies to | Type | Default |
|---|---|---|---|
| encoding | Reading JSON Lines | utf-8 · latin1 | utf-8 |
| flattenSeparator | Reading JSON Lines | string | . |
| arrayMode | Reading JSON Lines | json · explode | json |
| delimiter | Writing JSON | string | , |
| newline | Writing JSON | · | |
| sheetName | Writing JSON | string | |
| dialect | Writing JSON | postgres · mysql · sqlite | postgres |
| tableName | Writing JSON | string | data |
| createTable | Writing JSON | boolean | true |
No. It creates one JSON array containing those objects. That makes the file easier for editors but less stream-friendly.
Yes. The parser filters empty lines before parsing, so blank separators do not become null records.
The conversion uses the tabular IR, which flattens nested objects into columns. JSON output then writes those columns as object keys.