# Import files from Amazon S3

Robot: `/s3/import`

🤖/s3/import imports whole directories of files from your S3 bucket.

If you are new to Amazon S3, see our tutorial on [using your own S3 bucket](/docs/faq/how-to-set-up-an-amazon-s3-bucket.md).

The URL to the result file in your S3 bucket will be returned in the <dfn>Assembly Status JSON</dfn>.

> [!Warning]
> **Use DNS-compliant bucket names**. Your bucket name [must be DNS-compliant](https://docs.aws.amazon.com/AmazonS3/latest/userguide/bucketnamingrules.html) and must not contain uppercase letters. Any non-alphanumeric characters in the file names will be replaced with an underscore, and spaces will be replaced with dashes. If your existing S3 bucket contains uppercase letters or is otherwise not DNS-compliant, rewrite the result URLs using the <dfn>Robot</dfn>’s `url_prefix` parameter.

<span id="minimum-s3-iam-permissions" aria-hidden="true"></span>

## Limit access

You will also need to add permissions to your bucket so that Transloadit can access it properly. Here is an example IAM policy that you can use. Following the [principle of least privilege](https://en.wikipedia.org/wiki/Principle_of_least_privilege), it contains the **minimum required permissions** to import a file from your S3 bucket using Transloadit. You may require more permissions (especially viewing permissions) depending on your application.

Please change `{BUCKET_NAME}` in the values for `Sid` and `Resource` accordingly. Also, this policy will grant the minimum required permissions to all your users. We advise you to create a separate Amazon IAM user, and use its User ARN (can be found in the "Summary" tab of a user [here](https://console.aws.amazon.com/iam/home#users)) for the `Principal` value. More information about this can be found [here](https://docs.aws.amazon.com/AmazonS3/latest/dev/AccessPolicyLanguage_UseCases_s3_a.html).

```json
{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Sid": "AllowTransloaditToImportFilesIn{BUCKET_NAME}Bucket",
      "Effect": "Allow",
      "Action": ["s3:GetBucketLocation", "s3:ListBucket"],
      "Resource": ["arn:aws:s3:::{BUCKET_NAME}", "arn:aws:s3:::{BUCKET_NAME}/*"]
    }
  ]
}
```

The `Sid` value is just an identifier for you to recognize the rule later. You can name it anything you like.

The policy needs to be separated into two parts, because the `ListBucket` action requires permissions on the bucket while the other actions require permissions on the objects in the bucket. When targeting the objects there's a trailing slash and an asterisk in the `Resource` parameter, whereas when the policy targets the bucket, the slash and the asterisk are omitted.

In order to build proper result URLs we need to know the region in which your S3 bucket resides. For this we require the `GetBucketLocation` permission. Figuring out your bucket's region this way will also slow down your Assemblies. To make this much faster and to also not require the `GetBucketLocation` permission, we have added the `bucket_region` parameter to the /s3/store and /s3/import Robots. We recommend using them at all times.

Please keep in mind that if you use bucket encryption you may also need to add `"sts:*"` and `"kms:*"` to the bucket policy. Please read [here](https://docs.aws.amazon.com/kms/latest/developerguide/kms-api-permissions-reference.html) and [here](https://aws.amazon.com/blogs/security/how-to-restrict-amazon-s3-bucket-access-to-a-specific-iam-role/) in case you run into trouble with our example bucket policy.

Stage: ga

## Usage example

Import files from the \`path/to/files\` directory and its subdirectories:

```json
{
  "steps": {
    "imported": {
      "robot": "/s3/import",
      "credentials": "YOUR_AWS_CREDENTIALS",
      "path": "path/to/files/",
      "recursive": true
    }
  }
}
```

## Parameters

* `interpolate`: Controls whether Assembly Variables are interpolated for individual instruction fields.

  By default, most Robot instruction fields interpolate Assembly Variables. Set this to `false` to treat every instruction field as literal text, or set an individual field path to `false` to treat only that field as literal text. For Robot-specific fields that are literal by default, set this to `true` or set that field path to `true` to opt back into interpolation.

  Use field names such as `path`, or dotted paths such as `ffmpeg.vf` for nested objects.

* `output_meta`: Allows you to specify a set of metadata that is more expensive on CPU power to calculate, and thus is disabled by default to keep your Assemblies processing fast.

  For images, you can add `"has_transparency": true` in this object to extract if the image contains transparent parts and `"dominant_colors": true` to extract an array of hexadecimal color codes from the image.

  For images, you can also add `"blurhash": true` to extract a [BlurHash](https://blurha.sh) string — a compact representation of a placeholder for the image, useful for showing a blurred preview while the full image loads.

  For videos, you can add the `"colorspace": true` parameter to extract the colorspace of the output video.

  For videos, you can also add `"interlaced": true` to detect whether the video is interlaced. This combines the cheap ffprobe `field_order` flag with a bounded `idet` sampling pass over the first frames of the source, exposing `interlaced`, `field_order`, and a diagnostic `interlace_detection` object under `file.meta`. This is computationally expensive and billed accordingly.

  For audio, you can add `"mean_volume": true` to get a single value representing the mean average volume of the audio file.

  You can also set this to `false` to skip metadata extraction and speed up transcoding.

* `result`: Whether the results of this Step should be present in the Assembly Status JSON

* `queue`: Setting the queue to 'batch', manually downgrades the priority of jobs for this step to avoid consuming Priority job slots for jobs that don't need zero queue waiting times

* `force_accept`: Force a Robot to accept a file type it would have ignored.

  By default, Robots ignore files they are not familiar with.
  [🤖/video/encode](/docs/robots/video-encode.md), for
  example, will happily ignore input images.

  With the `force_accept` parameter set to `true`, you can force Robots to accept all files thrown at them.
  This will typically lead to errors and should only be used for debugging or combatting edge cases.

* `ignore_errors`: Ignore errors during specific phases of processing.

  Setting this to `["meta"]` will cause the Robot to ignore errors during metadata extraction.

  Setting this to `["execute"]` will cause the Robot to ignore errors during the main execution phase.

  Setting this to `["import"]` will cause the Robot to ignore errors while importing the source file. The `"import"` phase is only available to import Robots.

  Setting this to `true` is equivalent to `["meta", "import", "execute"]` and will ignore errors in all three phases.

* `force_name`: Custom name for the imported file(s). By default file names are derived from the source.

* `import_on_errors`: Setting this to `["meta"]` will still import the file on metadata extraction errors. `ignore_errors` is similar, it also ignores the error and makes sure the Robot doesn't stop, but it doesn't import the file.

* `credentials`: Please create your associated <dfn>Template Credentials</dfn> in your Transloadit account and use the name of your <dfn>Template Credentials</dfn> as this parameter's value. They will contain the values for your S3 `bucket`, `key`, `secret` and `bucket_region`.

  While we recommend to use <dfn>Template Credentials</dfn> at all times, some use cases demand dynamic credentials for which using <dfn>Template Credentials</dfn> is too unwieldy because of their static nature. If you have this requirement, feel free to use the following parameters instead: `"bucket"`, `"bucket_region"` (for example: `"us-east-1"` or `"eu-west-2"`), `"key"`, `"secret"`.

* `bucket`

* `bucket_region`

* `key`

* `secret`

* `path`: The path in your bucket to the specific file or directory. If the path points to a file, only this file will be imported. For example: `images/avatar.jpg`.

  If it points to a directory, indicated by a trailing slash (`/`), then all files that are direct descendants to this directory will be imported. For example: `images/`.

  Directories are **not** imported recursively. If you want to import files from subdirectories and sub-subdirectories, enable the `recursive` parameter.

  If you want to import all files from the root directory, please use `/` as the value here. In this case, make sure all your objects belong to a path. If you have objects in the root of your bucket that aren't prefixed with `/`, you'll receive an error: `A client error (NoSuchKey) occurred when calling the GetObject operation: The specified key does not exist.`

  You can also use an array of path strings here to import multiple paths in the same <dfn>Robot</dfn>'s <dfn>Step</dfn>.

* `recursive`: Setting this to `true` will enable importing files from subdirectories and sub-subdirectories (etc.) of the given path.

  Please use the pagination parameters `page_number` and `files_per_page` wisely here.

* `page_number`: The pagination page number. For now, in order to not break backwards compatibility in non-recursive imports, this only works when recursive is set to `true`.

  When doing big imports, make sure no files are added or removed from other scripts within your path, otherwise you might get weird results with the pagination.

* `files_per_page`: The pagination page size. This only works when recursive is `true` for now, in order to not break backwards compatibility in non-recursive imports.

* `return_file_stubs`: If set to `true`, the Robot will not yet import the actual files but instead return an empty file stub that includes a URL from where the file can be imported by subsequent Robots. This is useful for cases where subsequent Steps need more control over the import process, such as with 🤖/video/ondemand. This parameter should only be set if all subsequent Steps use Robots that support file stubs.

* `range`: Allows you to specify one or more byte ranges to import from the file. S3 must support range requests for this to work.

  **Single range**: Use a string like `"0-99"` to import bytes 0-99 (the first 100 bytes).

  **Multiple ranges**: Use an array like `["0-99", "200-299"]` to import multiple separate ranges. The resulting file will contain all requested ranges concatenated together, with zero bytes (\0) filling any gaps between non-contiguous ranges.

  **Range formats**:

  * `"0-99"`: Bytes 0 through 99 (inclusive)
  * `"100-199"`: Bytes 100 through 199 (inclusive)
  * `"-100"`: The last 100 bytes of the file

  **Important notes**:

  * S3 supports range requests by default
  * Overlapping ranges are allowed and will be included as requested
  * The resulting file size will be the highest byte position requested, with gaps filled with zero bytes
  * Each range is fetched in a separate request to ensure compatibility with S3