# Importing and exporting data with CSV
You can import and export data to/from your cloud service in the CSV (Comma Separated Values) format. This is done in the cloud services dashboard.
To export, browse the data for a specific record class and then you find the export commands in the toolbar at the top.

You can either export all records in a class (max 10K) or only the selected records. Select an option and then save the exported CSV.
To import a CSV you must also browse to the record class that you want to import into. Then you will find the import command in the `Edit` menu at the very bottom.

Click and then pick the CSV file on your computer that you want to use for import. The CSV file must follow a certain format for the import to work well.
### String
The input will be assigned directly as a string.
### Boolean
When the column data type is a boolean, there are a few different formats that are possible:
```js
"True" // == true
"true" // == true
"1" // == true
1 // == true
"FalsE" // == false
"false" // == false
"0" // == false
0 // == false
```
### Number
When the column data type is a number, it will convert the input to a number and assign if the value is not invalid or NaN.
### Date
Everything that is available via `new Date(...)` will be possible to import.
For example:
```json
"Thu Apr 13 2023 10:07:29 GMT+0200 (Central European Summer Time)"
"2023-04-13T08:02:46.447Z"
"Thu, 13 Apr 2023 08:06:11 GMT"
```
### Object
You can import an object type property by specifying it as JSON:
```json
{
"some_property":10
}
```
### Array
Properties of array types can also be provided as JSON.
```json
[1,2,3]
```
### GeoPoint
A propety of type GeoPoint can be imported either as a string with latitude and longitude:
```json
"30,30"
```
Or as an object with lat/long as properties:
```json
{"latitude": 30, "longitude": 30}
```
### Pointer
A point is simply a string with the `Id` of the target record.
### Relation
This is currently not supported.
### File
```json
{
"url": "path-to-file",
"name: "filename.png"
}
```