Tablizer

Convert JSON to SQL Online

Paste a JSON table below, edit it if you need to, and get SQL back. The conversion runs in your browser, so nothing is uploaded, there is no size limit and there is no signup.

From

The only part of this site that uses a server. The address is sent to it so the page can be fetched, because a browser is not allowed to fetch someone else's page itself. Your own files are never involved.

Nothing here is uploaded. The parsing happens on this page.
Edit
Paste or upload something above and it will appear here.
ToDataMarkupDocumentsCodeSchemasImages

The output will appear here once there is a table to convert.

What JSON is

JSON is a text format for nested data: objects with named keys, arrays, strings, numbers, booleans and null. It has no date type and no comments.

Every web API, most configuration files and anything that talks to JavaScript.

What SQL is

SQL here means the statements that move a table into a database: INSERT INTO with a VALUES list, and optionally the CREATE TABLE that defines the columns.

Anyone seeding a database, moving a spreadsheet into an application, or writing a migration.

What changes when you convert JSON to SQL

Four shapes all read correctly: an array of objects, an array of arrays with a header row, an object of column arrays, and an object keyed by row id. A single wrapper key holding the array, which is how most APIs reply, is unwrapped automatically. Nested objects are flattened one level into dotted keys; anything deeper is kept as JSON text in the cell, because a table cannot hold a tree.

String values are escaped by doubling any single quote. Statements are written in batches so a large table does not become one statement megabytes long, which many clients refuse. Ask for CREATE TABLE as well and column types are inferred: a column stays INT until a value does not fit, then widens to DECIMAL, then to VARCHAR.

What carries over from JSON to SQL

One property of the JSON has no home in SQL, and it is worth knowing which before you convert.

JSON can hold a value that is itself a list or an object. SQL has only flat cells, so nested values are flattened into one cell rather than being spread across columns.

SQL marks header cells differently from data cells, so the keys are lifted out and written once as a marked header row rather than repeated on every row.

The result can be read a record at a time, and appended to by adding to the end of it. That is worth having for a table too large to hold in memory, and it means a SQL file that is cut off part way through still gives you every record before the cut.

Both formats record types, so numbers, booleans and nulls survive as themselves rather than as text.

The identifier 007 comes out of the SQL as 007. Reading it as a number would have made it 7, and an id that changes value is worse than one that stays text.

The role "Analyst, data" survives with its comma, in one cell rather than split across two. That is the first thing to check in any converted table, and the usual place a SQL file goes wrong.

The quotes around Jonah "Jo" Pryce are left as a plain quote character in the SQL.

A worked example

Three rows of staff data, with an identifier that has a leading zero, a value containing a comma, a value containing quotes and one blank cell. Those are the four places formats disagree, so they are the four places to look.

JSON in
[
  { "id": "007", "name": "Halima Yusuf", "role": "Analyst, data", "started": "2024-03-15", "hours": "38.5" },
  { "id": "012", "name": "Jonah \"Jo\" Pryce", "role": "Engineer", "started": "2025-11-02", "hours": "" },
  { "id": "104", "name": "Wei Chen", "role": "Manager", "started": "2023-06-30", "hours": "40" }
]
SQL out
INSERT INTO `staff` (`id`, `name`, `role`, `started`, `hours`) VALUES
  ('007', 'Halima Yusuf', 'Analyst, data', '2024-03-15', '38.5'),
  ('012', 'Jonah "Jo" Pryce', 'Engineer', '2025-11-02', NULL),
  ('104', 'Wei Chen', 'Manager', '2023-06-30', '40');

Questions

How do I convert JSON to SQL?

Paste your JSON into the box above or drop the file onto it. Check the table in the grid, then copy or download the SQL from the output panel. It takes one step and the data never leaves your browser.

Why did my zip code lose its leading zero?

Type inference turned the text 07030 into the number 7030. Tablizer leaves inference off by default for exactly this reason, and refuses to convert any value starting with a zero followed by a digit even when it is on.

Are the generated INSERT statements safe to run?

Quotes in your data are escaped, so the statements are syntactically correct. They are not parameterised, which is fine for a one-off import you are running yourself and not fine as a pattern for application code that takes user input.

What happens to nested values?

SQL has no nested cell, so a list or object inside a JSON value is flattened into a single cell rather than expanded into extra columns. Split it before converting if the parts need to be separate.

Is there a limit on file size?

No. The work happens on your own machine, so the limit is your machine's memory rather than an upload cap. A file with tens of thousands of rows converts in a second or two.

Is my data uploaded anywhere?

No. The parsing and generating are done by JavaScript running on this page. You can watch the network tab while you convert and see that nothing is sent.

Related conversions