Convert JSONL to CSV Online | JSON Lines to CSV Converter

Convert JSON Lines (JSONL/NDJSON) files to CSV format for spreadsheet analysis, database imports, and data sharing. Transform streaming data exports into the universal tabular format supported by Excel, Google Sheets, and data tools.

Perfect for analyzing log files, ML training data exports, or any JSONL dataset using familiar spreadsheet tools and SQL databases.

Why Convert JSONL to CSV?

Spreadsheet Analysis: Open your data in Excel or Google Sheets for sorting, filtering, pivot tables, and charting.
Database Import: CSV is universally supported for importing into MySQL, PostgreSQL, SQLite, and other databases.
Data Sharing: Share data with colleagues who are comfortable with spreadsheets but not JSON formats.
Report Generation: Use CSV data with reporting tools, BI platforms, and data visualization software.

JSONL to CSV: choosing the output layout

Each JSONL record is a JSON object, and nested fields rarely map cleanly onto a single flat table. The Output layout option lets you choose exactly how nested fields become rows and columns in the CSV. Pick the shape that matches how you will use the data:

Tabletable

One column per field, one row per record. The default, and the best fit for flat or lightly nested JSON.

For JSONL this default uses a fast engine that keeps nested arrays compact. To expand nested fields into rows or columns, pick one of the layouts below.

Input (JSON Lines)

json

{"id":"1001","item":[{"sku":"A1"},{"sku":"B7"}]}
{"id":"1002","item":[{"sku":"C3"}]}

Output (CSV)

csv

"id","item"
"1001","[{'sku': A1}, {'sku': B7}]"
"1002","[{'sku': C3}]"

Key / Valuekey_value

Two columns, the field path and its value, with one row per value. Ideal for deeply nested or irregular JSON where a fixed set of columns does not fit.

Input (JSON Lines)

json

{"id":"1001","item":[{"sku":"A1"},{"sku":"B7"}]}
{"id":"1002","item":[{"sku":"C3"}]}

Output (CSV)

csv

"key","value"
"record/id","1001"
"record/item/sku","A1"
"record/item/sku","B7"
"record/id","1002"
"record/item/sku","C3"

One row per elementelement_rows

Every value on its own row alongside its full path. Useful for flattening or auditing all values in one long list.

Input (JSON Lines)

json

{"id":"1001","item":[{"sku":"A1"},{"sku":"B7"}]}
{"id":"1002","item":[{"sku":"C3"}]}

Output (CSV)

csv

"path","text","attributes"
"record/id","1001",
"record/item/sku","A1",
"record/item",,
"record/item/sku","B7",
"record/item",,
"record",,
"record/id","1002",
"record/item/sku","C3",
"record/item",,
"record",,

Nested levelspath_levels

The path is split across columns (level 1, level 2, and so on) followed by the value. Good for pivot-style analysis by depth.

Input (JSON Lines)

json

{"id":"1001","item":[{"sku":"A1"},{"sku":"B7"}]}
{"id":"1002","item":[{"sku":"C3"}]}

Output (CSV)

csv

"level_1","level_2","level_3","value","attributes"
"record","id",,"1001",
"record","item","sku","A1",
"record","item",,,
"record","item","sku","B7",
"record","item",,,
"record",,,,
"record","id",,"1002",
"record","item","sku","C3",
"record","item",,,
"record",,,,

One row per recordNewtranspose

Keeps each record on a single row and spreads repeating nested lists into indexed columns (for example line_items/0/amount, line_items/1/amount). The best fit when every record must stay one row even with nested arrays.

Input (JSON Lines)

json

{"id":"1001","item":[{"sku":"A1"},{"sku":"B7"}]}
{"id":"1002","item":[{"sku":"C3"}]}

Output (CSV)

csv

"record","id","item/0","item/0/sku","item/1","item/1/sku"
,"1001",,"A1",,"B7"
,"1002",,"C3",,

Conversion Details

By default each JSONL record becomes one CSV row, with top-level keys as the column headers. Use the Output layout option (above) to control how nested fields expand into rows and columns - from a compact table to fully flattened key/value pairs.

Format Resources

JSON Lines - Format specification
RFC 4180 - CSV format specification

Related Converters

CSV to JSONL - Convert CSV to JSON Lines format
JSONL to Excel - Convert directly to Excel spreadsheet
JSONL to Parquet - Convert to columnar format

Frequently Asked Questions

How are nested objects handled?

It depends on the Output layout you choose. The default table keeps nested values compact; the key/value, one-row-per-element, nested-levels and one-row-per-record layouts expand nested fields into rows and columns using slash-separated paths (for example record/item/sku).

What if records have different fields?

The CSV will include columns for all unique fields found across all records. Missing values are represented as empty cells.