Open JSON Online as a Table

An API export or a JSON Lines log reads like an ordinary table: with sorting, filters, statistics and export to Excel.

Free for personal use. .json, .jsonl and .ndjson are supported. Windows version — 3.4 MB, no installation: details. Organizations: license.

The table finds itself

If the array is not at the root but under a key such as data.items, it is found and suggested. Another array can be chosen in “File parsing”.

Nesting into columns

{"owner": {"name": "Smith"}} becomes the column owner.name. Arrays of primitives are joined with commas, objects stay JSON.

JSON Lines

A file with a separate object on every line is recognized automatically. Broken lines are skipped and counted instead of aborting the parse.

Types as in the document

Numbers stay numbers, booleans stay booleans, strings holding dates are recognized as dates. Nothing is converted to text without need.

Your file never leaves your computer. Parsing happens in the browser, in a background thread. The page's security policy forbids sending data to third-party addresses — you can check this in the developer tools. After the first visit the app keeps working offline.

What to do with large files

A nested API response in JSON opened as a table

JSON is a format that has to be parsed as a whole: it has no lines from which an index could be built. So for files over a hundred megabytes we warn about memory use, and above half a gigabyte we honestly refuse — rather than hang the tab.

For such volumes JSON Lines is the answer: it is read line by line and behaves like CSV. See opening large JSON files and JSON Lines and NDJSON.

From JSON to a table and back

The filtered result can be exported to Excel, CSV, SQL, Markdown, HTML and DBF. Export back to JSON and JSON Lines is available too — handy when you need to pick a subset out of a big export and pass it on.

Guides: JSON to CSV, nested JSON to a table and API response to Excel.

FAQ

Why are there more columns than keys in one object?

Columns are collected as the union of the keys of all elements: if some objects have extra fields, they become columns too and are empty for the rest.

Can I leave nesting unfolded?

Yes, switch unfolding off in “File parsing” — nested objects stay in the column as JSON.

Is NDJSON supported?

Yes, the format is detected from the content; the extension is not required.

Is the data really not sent anywhere?

Really. Parsing is done in the browser; there are no network requests carrying the file's content.

In depth

How to open JSONHow to open and read JSON: browser, editors, online viewers, a table. What to pick for a quick look, debugging and working with large exports.
JSON to Excel and CSVHow to export JSON to CSV and XLSX: what to do with nesting and arrays, how to keep long IDs intact and how to handle paginated API exports.
Nested JSON to a flat tableWhat to do with nested objects and arrays when turning JSON into a table: dotted keys, arrays of primitives, arrays of objects and when you need a second table.
JSON Lines: a format you do not have to read in fullWhat JSON Lines is, how it differs from regular JSON, why logs and exports use it, and how to view a .jsonl or .ndjson file as a table.
API response → table → ExcelHow to get a table out of an API response: where to find the data array, what to do with nesting and pagination, and how to export to XLSX with correct types.
Large JSON and why it hangs everythingWhy JSON of hundreds of megabytes hangs editors and browsers, how much memory parsing takes, how JSON Lines helps and how to still look at such a file.