# Import data from a CSV file

This example shows you how to create a Smart Action `"Import data"` to import data from a CSV file.

Forest Admin natively supports data creation but it’s sometimes more efficient to simply import it.

![](/files/-M4P93GTR4DfNFnnasu3)

## Requirements

* An admin backend running on forest-express-sequelize/forest-express-mongoose
* [bluebird](https://www.npmjs.com/package/bluebird) npm package
* [parse-data-uri](https://www.npmjs.com/package/parse-data-uri) npm package
* [csv](https://www.npmjs.com/package/csv) npm package

## How it works

### Directory: /models

This directory contains the `products.js` file where the model is declared.

{% tabs %}
{% tab title="Sequelize" %}
{% code title="/models/products.js" %}

```javascript

module.exports = (sequelize, DataTypes) => {
  const { Sequelize } = sequelize;
  const Products = sequelize.define('products', {
    price: {
      type: DataTypes.INTEGER,
    },
    label: {
      type: DataTypes.STRING,
    },
    picture: {
      type: DataTypes.STRING,
    },
    ...
  }, {
    tableName: 'products',
    underscored: true,
    schema: process.env.DATABASE_SCHEMA,
  });
​
  Produtcs.associate = (models) => {
  };
​
  return Products;
};
```

{% endcode %}
{% endtab %}

{% tab title="Mongoose" %}
{% code title="/models/products.js" %}

```javascript
const mongoose = require('mongoose');

const schema = mongoose.Schema({
  'price': Integer,
  'label': String,
  'picture': String,
  ...
}, {
  timestamps: false,
});

module.exports = mongoose.model('products', schema, 'products');
```

{% endcode %}
{% endtab %}
{% endtabs %}

### **Directory: /forest**

This directory contains the `products.js` file where the Smart Action `Import data`is declared.

{% tabs %}
{% tab title="Sequelize" %}
{% code title="/forest/products.js" %}

```javascript
const { collection } = require('forest-express-sequelize');
const models = require('../models');

collection('products', {
  actions: [{
    name: 'Import data',
    endpoint: '/forest/products/actions/import-data',
    type: 'global',
    fields: [{
      field: 'CSV file',
      description: 'A semicolon separated values file stores tabular data (numbers and text) in plain text',
      type: 'File',
      isRequired: true
    }, {
      field: 'Type',
      description: 'Specify the product type to import',
      type: 'Enum',
      enums: ['phone', 'dress', 'toy'],
      isRequired: true
    }]
  }],

  // ...
});
```

{% endcode %}
{% endtab %}

{% tab title="Mongoose" %}
{% code title="/forest/products.js" %}

```javascript
const { collection } = require('forest-express-mongoose');
const models = require('../models');

Liana.collection('products', {
  actions: [{
    name: 'Import data',
    endpoint: '/forest/products/actions/import-data',
    type: 'global',
    fields: [{
      field: 'CSV file',
      description: 'A semicolon separated values file stores tabular data (numbers and text) in plain text',
      type: 'File',
      isRequired: true
    }, {
      field: 'Type',
      description: 'Specify the product type to import',
      type: 'Enum',
      enums: ['phone', 'dress', 'toy'],
      isRequired: true
    }]
  }],

  // ...
});
```

{% endcode %}
{% endtab %}
{% endtabs %}

### **Directory: /routes**

This directory contains the `products.js` file where the implementation of the route is handled. The `POST /forest/actions/import-data` API call is triggered when you click on the Smart Action in the Forest UI. 

The CSV file passed into the body of the API call is serialized using a base64 encoding [Data URI scheme](https://en.wikipedia.org/wiki/Data_URI_scheme).

To deserialize the base64 encoded CSV file, we use the NPM package [parse-data-uri](https://www.npmjs.com/package/parse-data-uri). We also use the [csv parser](https://github.com/adaltas/node-csv) NPM package to iterate over each line of the CSV file.

You can find a sample CSV file we use here to feed our products table on the [Live demo Github repository](https://raw.githubusercontent.com/ForestAdmin/forest-live-demo-lumber/master/seed/walmart-toys.csv).

You may find below the coding examples you need to make this Smart action work:

{% tabs %}
{% tab title="Sequelize" %}
{% code title="/routes/products.js" %}

```javascript
const P = require('bluebird');
const express = require('express');
const router = express.Router();
const faker = require('faker');
const parseDataUri = require('parse-data-uri');
const csv = require('csv');
const models = require('../models');

//...

router.post('/products/actions/import-data',
  (req, res) => {
    let parsed = parseDataUri(req.body.data.attributes.values['CSV file']);
    let productType = req.body.data.attributes.values['Type'];

    csv.parse(parsed.data, { delimiter: ';' }, function (err, rows) {
      if (err) {
        res.status(400).send({
          error: `Cannot import data: ${err.message}` });
      } else {
        return P
          .each(rows, (row) => {
            // Random price for the example purpose. In a real situation, the price 
            // should certainly be available in the CSV file.
            let price = faker.commerce.price(5, 1000) * 100;

            return models.products.create({
              label: row[0],
              price: price,
              picture: row[1]
            });
          })
          .then(() => {
            res.send({ success: 'Data successfuly imported!' });
          });
      }
    });
  });
  
//...

module.exports = router;
```

{% endcode %}
{% endtab %}

{% tab title="Mongoose" %}
{% code title="/routes/products.js" %}

```javascript
const P = require('bluebird');
const express = require('express');
const router = express.Router();
const faker = require('faker');
const parseDataUri = require('parse-data-uri');
const csv = require('csv');
const models = require('../models');

...

router.post('/products/actions/import-data',
  (req, res) => {
    let parsed = parseDataUri(req.body.data.attributes.values['CSV file']);
    let productType = req.body.data.attributes.values['Type'];

    csv.parse(parsed.data, { delimiter: ';' }, function (err, rows) {
      if (err) {
        res.status(400).send({
          error: `Cannot import data: ${err.message}` });
      } else {
        return P
          .each(rows, (row) => {
            // Random price for the example purpose. In a real situation, the price 
            // should certainly be available in the CSV file.
            let price = faker.commerce.price(5, 1000) * 100;

            return models.products.create({
              label: row[0],
              price: price,
              picture: row[1]
            });
          })
          .then(() => {
            res.send({ success: 'Data successfuly imported!' });
          });
      }
    });
  });
  
...

module.exports = router;
```

{% endcode %}
{% endtab %}
{% endtabs %}

### Uploading large files

{% hint style="warning" %}
For large file uploads, you should add an option in your Express Server in your **app.js** file:
{% endhint %}

```javascript
app.use(bodyParser.urlencoded({ extended: true ,limit: '50mb' }));
app.use(bodyParser.json({ limit: '50mb' }));
```


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.forestadmin.com/woodshop/how-tos/import-data-from-a-csv-file.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.