Project import and export API

Use the project import and export API to import and export projects using file transfers.

Before using the project import and export API, you might want to use the group import and export API.

Prerequisites

For prerequisites for project import and export API, see:

Schedule an export

Start a new export.

The endpoint also accepts an upload hash parameter. It contains all the necessary information to upload the exported project to a web server or to any S3-compatible platform. For exports, GitLab:

  • Only supports binary data file uploads to the final server.
  • Sends the Content-Type: application/gzip header with upload requests. Ensure that your pre-signed URL includes this as part of the signature.
  • Can take some time to complete the project export process. Make sure the upload URL doesn’t have a short expiration time and is available throughout the export process.
  • Administrators can modify the maximum export file size. By default, the maximum is unlimited (0). To change this, edit max_export_size using either:
  • Has a fixed limit for the maximum import file size on GitLab.com. For more information, see Account and limit settings.

The upload[url] parameter is required if the upload parameter is present.

For uploads to Amazon S3, refer to Generating a pre-signed URL for uploading objects documentation scripts to generate the upload[url].

POST /projects/:id/export
AttributeTypeRequiredDescription
idinteger or stringyesThe ID or URL-encoded path of the project owned by the authenticated user.
upload[url]stringyesThe URL to upload the project.
descriptionstringnoOverrides the project description.
uploadhashnoHash that contains the information to upload the exported project to a web server.
upload[http_method]stringnoThe HTTP method to upload the exported project. Only PUT and POST methods allowed. Default is PUT.
curl --request POST --header "PRIVATE-TOKEN: <your_access_token>" \
    "https://gitlab.example.com/api/v4/projects/1/export" \
    --data "upload[http_method]=PUT" \
    --data-urlencode "upload[url]=https://example-bucket.s3.eu-west-3.amazonaws.com/backup?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIMBJHN2O62W8IELQ%2F20180312%2Feu-west-3%2Fs3%2Faws4_request&X-Amz-Date=20180312T110328Z&X-Amz-Expires=900&X-Amz-SignedHeaders=host&X-Amz-Signature=8413facb20ff33a49a147a0b4abcff4c8487cc33ee1f7e450c46e8f695569dbd"
{
  "message": "202 Accepted"
}

Export status

Get the status of export.

GET /projects/:id/export
AttributeTypeRequiredDescription
idinteger or stringyesThe ID or URL-encoded path of the project owned by the authenticated user.
curl --header "PRIVATE-TOKEN: <your_access_token>" \
  "https://gitlab.example.com/api/v4/projects/1/export"

Status can be one of:

  • none: No exports queued, started, finished, or being regenerated.
  • queued: The request for export is received, and is in the queue to be processed.
  • started: The export process has started and is in progress. It includes:
    • The process of exporting.
    • Actions performed on the resulting file, such as sending an email notifying the user to download the file, or uploading the exported file to a web server.
  • finished: After the export process has completed and the user has been notified.
  • regeneration_in_progress: An export file is available to download, and a request to generate a new export is in process.

_links are only present when export has finished.

created_at is the project create timestamp, not the export start time.

{
  "id": 1,
  "description": "Itaque perspiciatis minima aspernatur corporis consequatur.",
  "name": "Gitlab Test",
  "name_with_namespace": "Gitlab Org / Gitlab Test",
  "path": "gitlab-test",
  "path_with_namespace": "gitlab-org/gitlab-test",
  "created_at": "2017-08-29T04:36:44.383Z",
  "export_status": "finished",
  "_links": {
    "api_url": "https://gitlab.example.com/api/v4/projects/1/export/download",
    "web_url": "https://gitlab.example.com/gitlab-org/gitlab-test/download_export"
  }
}

Export download

Download the finished export.

GET /projects/:id/export/download
AttributeTypeRequiredDescription
idinteger or stringyesThe ID or URL-encoded path of the project owned by the authenticated user.
curl --header "PRIVATE-TOKEN: <your_access_token>" --remote-header-name \
     --remote-name "https://gitlab.example.com/api/v4/projects/5/export/download"
ls *export.tar.gz
2017-12-05_22-11-148_namespace_project_export.tar.gz

Import a file

Requirement for Maintainer role instead of Developer role introduced in GitLab 16.0 and backported to GitLab 15.11.1 and GitLab 15.10.5.

POST /projects/import
AttributeTypeRequiredDescription
filestringyesThe file to be uploaded.
pathstringyesName and path for new project.
namestringnoThe name of the project to be imported. Defaults to the path of the project if not provided.
namespaceinteger or stringnoThe ID or path of the namespace to import the project to. Defaults to the current user’s namespace.

Requires at least the Maintainer role on the destination group to import to.
override_paramsHashnoSupports all fields defined in the Project API.
overwritebooleannoIf there is a project with the same path the import overwrites it. Defaults to false.

The override parameters passed take precedence over all values defined inside the export file.

To upload a file from your file system, use the --form argument. This causes cURL to post data using the header Content-Type: multipart/form-data. The file= parameter must point to a file on your file system and be preceded by @. For example:

curl --request POST --header "PRIVATE-TOKEN: <your_access_token>" --form "path=api-project" \
     --form "file=@/path/to/file" "https://gitlab.example.com/api/v4/projects/import"

cURL doesn’t support posting a file from a remote server. This example imports a project using Python’s open method:

import requests

url =  'https://gitlab.example.com/api/v4/projects/import'
files = { "file": open("project_export.tar.gz", "rb") }
data = {
    "path": "example-project",
    "namespace": "example-group"
}
headers = {
    'Private-Token': "<your_access_token>"
}

requests.post(url, headers=headers, data=data, files=files)
{
  "id": 1,
  "description": null,
  "name": "api-project",
  "name_with_namespace": "Administrator / api-project",
  "path": "api-project",
  "path_with_namespace": "root/api-project",
  "created_at": "2018-02-13T09:05:58.023Z",
  "import_status": "scheduled",
  "correlation_id": "mezklWso3Za",
  "failed_relations": []
}
note
The maximum import file size can be set by the Administrator. It defaults to 0 (unlimited). As an administrator, you can modify the maximum import file size. To do so, use the max_import_size option in the Application settings API or the Admin Area. Default modified from 50 MB to 0 in GitLab 13.8.

Import a file from a remote object storage

Introduced in GitLab 13.12 in Beta with a flag named import_project_from_remote_file. Enabled by default.

On self-managed GitLab, by default this feature is available. To hide the feature, an administrator can disable the feature flag named import_project_from_remote_file. On GitLab.com, this feature is available.
POST /projects/remote-import
AttributeTypeRequiredDescription
pathstringyesName and path for the new project.
urlstringyesURL for the file to import.
namestringnoThe name of the project to import. If not provided, defaults to the path of the project.
namespaceinteger or stringnoThe ID or path of the namespace to import the project to. Defaults to the current user’s namespace.
overwritebooleannoWhether to overwrite a project with the same path when importing. Defaults to false.
override_paramsHashnoSupports all fields defined in the Project API.

The passed override parameters take precedence over all values defined in the export file.

curl --request POST \
  --header "PRIVATE-TOKEN: <your_access_token>" \
  --header "Content-Type: application/json" \
  --url "https://gitlab.example.com/api/v4/projects/remote-import" \
  --data '{"url":"https://remoteobject/file?token=123123","path":"remote-project"}'
{
  "id": 1,
  "description": null,
  "name": "remote-project",
  "name_with_namespace": "Administrator / remote-project",
  "path": "remote-project",
  "path_with_namespace": "root/remote-project",
  "created_at": "2018-02-13T09:05:58.023Z",
  "import_status": "scheduled",
  "correlation_id": "mezklWso3Za",
  "failed_relations": [],
  "import_error": null
}

The Content-Length header must return a valid number. The maximum file size is 10 GB. The Content-Type header must be application/gzip.

Import a file from AWS S3

Version history
POST /projects/remote-import-s3
AttributeTypeRequiredDescription
access_key_idstringyes AWS S3 access key ID.
bucket_namestringyes AWS S3 bucket name where the file is stored.
file_keystringyes AWS S3 file key to identify the file.
pathstringyesThe full path of the new project.
regionstringyes AWS S3 region name where the file is stored.
secret_access_keystringyes AWS S3 secret access key.
namestringnoThe name of the project to import. If not provided, defaults to the path of the project.
namespaceinteger or stringnoThe ID or path of the namespace to import the project to. Defaults to the current user’s namespace.

The passed override parameters take precedence over all values defined in the export file.

curl --request POST \
  --url "https://gitlab.example.com/api/v4/projects/remote-import-s3" \
  --header "PRIVATE-TOKEN: <your gitlab access key>" \
  --header 'Content-Type: application/json' \
  --data '{
  "name": "Sample Project",
  "path": "sample-project",
  "region": "<Your S3 region name>",
  "bucket_name": "<Your S3 bucket name>",
  "file_key": "<Your S3 file key>",
  "access_key_id": "<Your AWS access key id>",
  "secret_access_key": "<Your AWS secret access key>"
}'

This example imports from an Amazon S3 bucket, using a module that connects to Amazon S3:

import requests
from io import BytesIO

s3_file = requests.get(presigned_url)

url =  'https://gitlab.example.com/api/v4/projects/import'
files = {'file': ('file.tar.gz', BytesIO(s3_file.content))}
data = {
    "path": "example-project",
    "namespace": "example-group"
}
headers = {
    'Private-Token': "<your_access_token>"
}

requests.post(url, headers=headers, data=data, files=files)
{
  "id": 1,
  "description": null,
  "name": "Sample project",
  "name_with_namespace": "Administrator / sample-project",
  "path": "sample-project",
  "path_with_namespace": "root/sample-project",
  "created_at": "2018-02-13T09:05:58.023Z",
  "import_status": "scheduled",
  "correlation_id": "mezklWso3Za",
  "failed_relations": [],
  "import_error": null
}

Import status

Get the status of an import.

GET /projects/:id/import
AttributeTypeRequiredDescription
idinteger or stringyesThe ID or URL-encoded path of the project owned by the authenticated user.
curl --header "PRIVATE-TOKEN: <your_access_token>" \
  "https://gitlab.example.com/api/v4/projects/1/import"

Status can be one of:

  • none
  • scheduled
  • failed
  • started
  • finished

If the status is failed, it includes the import error message under import_error. If the status is failed, started or finished, the failed_relations array might be populated with any occurrences of relations that failed to import due to either:

  • Unrecoverable errors.
  • Retries were exhausted. A typical example: query timeouts.
note
An element’s id field in failed_relations references the failure record, not the relation.
note
The failed_relations array is capped to 100 items.
{
  "id": 1,
  "description": "Itaque perspiciatis minima aspernatur corporis consequatur.",
  "name": "Gitlab Test",
  "name_with_namespace": "Gitlab Org / Gitlab Test",
  "path": "gitlab-test",
  "path_with_namespace": "gitlab-org/gitlab-test",
  "created_at": "2017-08-29T04:36:44.383Z",
  "import_status": "started",
  "import_type": "github",
  "correlation_id": "mezklWso3Za",
  "failed_relations": [
    {
      "id": 42,
      "created_at": "2020-04-02T14:48:59.526Z",
      "exception_class": "RuntimeError",
      "exception_message": "A failure occurred",
      "source": "custom error context",
      "relation_name": "merge_requests",
      "line_number": 0
    }
  ]
}

When importing from GitHub, the a stats field lists how many objects were already fetched from GitHub and how many were already imported:

{
  "id": 1,
  "description": "Itaque perspiciatis minima aspernatur corporis consequatur.",
  "name": "Gitlab Test",
  "name_with_namespace": "Gitlab Org / Gitlab Test",
  "path": "gitlab-test",
  "path_with_namespace": "gitlab-org/gitlab-test",
  "created_at": "2017-08-29T04:36:44.383Z",
  "import_status": "started",
  "import_type": "github",
  "correlation_id": "mezklWso3Za",
  "failed_relations": [
    {
      "id": 42,
      "created_at": "2020-04-02T14:48:59.526Z",
      "exception_class": "RuntimeError",
      "exception_message": "A failure occurred",
      "source": "custom error context",
      "relation_name": "merge_requests",
      "line_number": 0
    }
  ],
  "stats": {
    "fetched": {
      "diff_note": 19,
      "issue": 3,
      "label": 1,
      "note": 3,
      "pull_request": 2,
      "pull_request_merged_by": 1,
      "pull_request_review": 16
    },
    "imported": {
      "diff_note": 19,
      "issue": 3,
      "label": 1,
      "note": 3,
      "pull_request": 2,
      "pull_request_merged_by": 1,
      "pull_request_review": 16
    }
  }
}