CSV to SQL INSERT

How the CSV to SQL INSERT works

The CSV is read with a standard (RFC 4180) parser: quoted fields may contain the delimiter, line breaks and doubled quotes, and blank lines are skipped. The first row supplies the column names. Empty or repeated names are made unique, and rows with fewer values than the header get NULL for the missing columns.

Before writing anything, every column is checked: if all of its values (apart from NULLs) are plain numbers, the column is written without quotes; otherwise every value in it is written as a string, with single quotes doubled, N'...' for non-ASCII text on SQL Server and escaped backslashes on MySQL. That keeps values such as postal code 01067 exactly as they are. Rows are then grouped into INSERT ... VALUES statements of the size you choose. The SQL is only generated as text and is never run.

How to use the CSV to SQL INSERT

1. Paste CSV with a header row into the input box, open a .csv file, or select Load example.
2. Enter the table name, choose the dialect, the delimiter (or leave it on Detect automatically) and how many rows go into each INSERT.
3. Choose how empty cells and the text NULL are treated, and tick Include CREATE TABLE if you want a table definition too.
4. Select Convert to SQL, or press Ctrl + Enter (Cmd + Enter on a Mac), then copy the SQL or download it as a .sql file.

Example

This CSV, with table name customers and PostgreSQL:

id,name,postal_code,credit_limit
1,"Smith, Jane",01067,2500.00
2,Seán O'Connor,,NULL

produces:

INSERT INTO "customers" ("id", "name", "postal_code", "credit_limit")
VALUES
    (1, 'Smith, Jane', '01067', 2500.00),
    (2, 'Seán O''Connor', NULL, NULL);

Common use cases

• Importing a spreadsheet saved from Excel or Google Sheets as CSV into a database table.
• Turning a CSV export from another system into a script that can be reviewed and run later.
• Seeding lookup tables such as countries, currencies or product categories.
• Sharing a small dataset as a SQL script that anyone can run without an import wizard.

Common errors and how to fix them

Unterminated quoted field starting on line N

A field opened with a double quote but never closed it, so the parser reads the rest of the file as one value. Find that line and add the closing quote, remembering that a quote inside a quoted field is written twice.

The CSV has a header row but no data rows

Only the column names were pasted. Add at least one row of values below the header.

The first data row was treated as column names

The first row is always read as the header. If your file has no header, type the column names into the Columns box instead.

Numbers or dates were quoted as text

Values are taken from the file as written, and CSV carries no type information. Set the column types for the target table yourself, and check the generated statements before running them.

Frequently asked questions