Use the free CSV to CREATE TABLE SQL below: your data never leaves your browser — all work happens on your device, and nothing is uploaded to any server.
Moving data from a spreadsheet into a database starts with one tedious step: writing the CREATE TABLE statement. You must list every column, choose a type for each, and get the syntax exactly right, or the database refuses the whole statement. The CSV to CREATE TABLE SQL tool does that first draft for you. Paste a CSV, and it reads the headers, infers a type for each column, and emits a CREATE TABLE statement as plain text, ready to copy into your database tool.
This tool is built for people setting up a database, not for database experts alone. You do not need to install anything, learn SQL typing rules, or upload your file. Everything happens in your browser, on your own device. Your CSV never leaves the page, and nothing is sent to a server. That matters when the CSV holds customer names, prices, or internal records you would rather not hand to a stranger.
To use it, paste a CSV with a header row first, as the sample shows. Set the table name, using letters, numbers, and underscores, and pick a dialect: generic SQL, MySQL, or PostgreSQL. The main difference is small but real: generic and PostgreSQL output uses TEXT and INTEGER, while MySQL turns TEXT into VARCHAR(255), which is the more common MySQL choice. Press Generate SQL and the statement appears, together with the column count and the number of rows read. Press Copy SQL to copy it to the clipboard.
Type inference follows a simple, stated rule, so you can predict it. A column of whole numbers becomes INTEGER. A column with decimals becomes DECIMAL(10,2), which suits money and measurements with two decimal places. A column of true and false becomes BOOLEAN. A column of dates in YYYY-MM-DD form becomes DATE. Anything else, including a column that mixes shapes, becomes TEXT, which is the safe choice: TEXT accepts anything, while a wrongly narrow type would reject real rows. Empty cells are ignored when inferring, so a half-empty column still gets the type its filled cells suggest.
That safety-first rule has a trade-off you should know. A column of codes such as 00123 looks like an integer but is really an identifier, and storing it as INTEGER would drop the leading zeros. A column of mixed text and numbers becomes TEXT, which is correct but loses numeric ordering. The generated statement is therefore a first draft, not a final schema. Check each inferred type before running it on a real database, especially for identifiers, codes, phone numbers, and money with more than two decimal places, and adjust the types to match what the column really means.
Column names are cleaned automatically: spaces and punctuation become underscores, so a header like First Name becomes First_Name. If two headers clean to the same name, the database will reject the statement, so check for duplicates after cleaning. The tool also reads the CSV with PapaParse, a widely used parsing library loaded from a public CDN, with a built-in fallback parser if that library fails to load, so quoted fields containing commas are handled correctly rather than split in the wrong place.
People use this tool when importing a spreadsheet export into a new project, when setting up a demo database for a class, and when moving a mailing list or product list into an app. The generated statement saves the typing and the syntax errors, and the review step, checking the types, is where the human judgment belongs. A database schema is a commitment: changing a column type later, after rows exist, is harder than choosing well now.
One more honest limit: the tool generates the table definition only, not the INSERT statements for your rows, and it does not connect to any database. That separation is deliberate. Generating SQL as text keeps your data local and lets you review every line before it touches a real database, which is exactly the caution a new schema deserves.
How to generate CREATE TABLE SQL from CSV
- Paste your CSV with a header row first.
- Set the table name and pick a dialect.
- Press Generate SQL.
- Check each inferred type, especially identifiers and codes.
- Press Copy SQL and paste it into your database tool.
Frequently asked questions
Why did my code column become INTEGER? It looks like whole numbers. Change it to TEXT yourself to keep leading zeros.
Why is a mixed column TEXT? TEXT is the safe type that accepts every shape in the column. A narrower type would reject real rows.
Does it insert my rows? No. It generates the table definition only. Review it, then load rows separately.
Is my data uploaded? No. All parsing and generation happens in your browser.
Paste a CSV you need in a database and generate the first draft, then spend your attention where it matters: checking that each type matches what the column really means.
Pingback: Convertir CSV a CREATE TABLE SQL Gratis - Genera DDL desde tu CSV