Archives

An archive is a collection of resources stored inside an archive file. The archiving format chosen for this purpose is ZIP. Archive resources are rooted in the /v1/archives/{org_label}/{project_label}/ collection.

Each archive…

• belongs to a project identifier by the label {project_label}
• inside an organization identifier by the label {org_label}

Lifecycle

Contrarily to the rest of the platform resources, archives are not persisted resources, given their nature. Therefore there are no update, tag or deprecation operations available on archive resources.

An archive resource will be automatically erased from the system after certain after certain time. This time is configurable (config property plugins.archive.ephemeral.ttl) and it defaults to 5 hours.

Payload

{
    "resources" : [
        {
            "@type": "Resource",
            "resourceId": "{resource_id}",
            "project": "{project}",
            "path": "{path}",
            "originalSource": "{originalSource}",
            "format": "{format}",
            "rev": "{rev}",
            "tag": "{tag}"
        },
        {
            "@type": "File",
            "resourceId": "{resource_id}",
            "project": "{project}",
            "path": "{path}",
            "rev": "{rev}",
            "tag": "{tag}"
        },
        {
            "@type": "FileSelf",
            "value": "{file_self}",
            "path": "{path}",
        },
        {...}       
    ]
}

where…

• {resource_id}: Iri - the @id value of the resource to be added to the archive.
• {file_self}: Uri - the _self value of the file to be added to the archive. This is the same as the http location of the file in the Delta API.
• {project}: String - the project (in the format ‘myorg/myproject’) where the specified resource belongs. This field is optional. It defaults to the current project.
• {path}: Path - the relative path on the archive where this resource is going to stored
  • Optional / defaults to {project}/{resourceId}.json for a Resource type and {project}/{filename} for File type.
  • The provided filename should not exceed 99 characters.
• {originalSource}: Boolean - flag to decide the whether to fetch the original payload or its compacted form.
  • Only allowed for Resource type
  • Optional and defaults to false
  • Can not be present at the same time as format field.
• {format}: String - the format we expect for the resource in the archive.
  • Only allowed for Resource type
  • Optional and defaults to compacted
  • Accepts the following values: source (to get the original payload), annotated-source (to get the original payload with metadata), compacted, expanded, n-triples, dot
  • Can not be present at the same time as originalSource field.
• {rev}: Int - the revision of the resource. This field is optional. It defaults to the latest revision.
• {tag}: String - the tag of the resource. This field is optional. This field cannot be present at the same time as rev field.

In order to decide whether we want to select a resource or a file, the @type discriminator is used with the following possibilities:

• Resource: targets a resource
• File: targets a file using its project and id
• FileSelf: targets a file using its address (_self)

Create using POST

This endpoint is used to describe the archive and to subsequently consume it.

POST /v1/archives/{org_label}/{project_label}

The json payload:

• If the @id value is found on the payload, this @id will be used.
• If the @id value is not found on the payload, an @id will be generated as follows: base:{UUID}. The base is the prefix defined on the resource’s project ({project_label}).

The response will be an HTTP 303 Location redirect, which will point to the url where to consume the archive (ZIP).

The following diagram can help to understand the HTTP exchange

Example

The following example shows how to create an archive containing 3 files. 2 of them are resources and the other is a file. As a response, the ZIP file will be offered.

Request

copysourcecurl -L \
     -X POST \
     -H "Content-Type: application/json" -H "Accept: */*" \
     "http://localhost:8080/v1/archives/myorg/myproject" \
     -d '{
    "resources" : [
        {
            "@type": "Resource",
            "resourceId": "http://localhost:8080/resource1",
            "rev": 2
        },
        {
            "@type": "Resource",
            "resourceId": "http://localhost:8080/resource2",
            "project": "myorg/myproject2",
            "originalSource": false
        },
        {
            "@type": "File",
            "resourceId": "http://localhost:8080/resource2",
            "project": "myorg/myproject2",
            "path": "my/custom/path/resource2.json"
        }
    ]
}
'

Payload

copysource{
    "resources" : [
        {
            "@type": "Resource",
            "resourceId": "http://localhost:8080/resource1",
            "rev": 2
        },
        {
            "@type": "Resource",
            "resourceId": "http://localhost:8080/resource2",
            "project": "myorg/myproject2",
            "format": "expanded"
        },
        {
            "@type": "File",
            "resourceId": "http://localhost:8080/resource2",
            "project": "myorg/myproject2",
            "path": "my/custom/path/resource2.json"
        }
    ]
}

Create using PUT

This alternative endpoint to create an archive is useful in case you want to split the creation of the archive resource and the consumption of it.

It can also be useful in cases where one user wants to create the definition of the archive and share the link with another user who then is going to consume it.

PUT /v1/archives/{org_label}/{project_label}/{archive_id}

Example

Request

copysourcecurl -X POST \
     -H "Content-Type: application/json" \
     "http://localhost:8080/v1/archives/myorg/myproject/myarchive" \
     -d '{
    "resources" : [
        {
            "@type": "Resource",
            "resourceId": "http://localhost:8080/resource1",
            "rev": 2
        },
        {
            "@type": "Resource",
            "resourceId": "http://localhost:8080/resource2",
            "project": "myorg/myproject2",
            "originalSource": false
        },
        {
            "@type": "File",
            "resourceId": "http://localhost:8080/resource2",
            "project": "myorg/myproject2",
            "path": "my/custom/path/resource2.json"
        }
    ]
}
'

Payload

copysource{
    "resources" : [
        {
            "@type": "Resource",
            "resourceId": "http://localhost:8080/resource1",
            "rev": 2
        },
        {
            "@type": "Resource",
            "resourceId": "http://localhost:8080/resource2",
            "project": "myorg/myproject2",
            "format": "expanded"
        },
        {
            "@type": "File",
            "resourceId": "http://localhost:8080/resource2",
            "project": "myorg/myproject2",
            "path": "my/custom/path/resource2.json"
        }
    ]
}

Response

copysource{
  "@context": [
    "https://bluebrain.github.io/nexus/contexts/archives-metadata.json",
    "https://bluebrain.github.io/nexus/contexts/metadata.json"
  ],
  "@id": "http://localhost:8080/v1/archives/myorg/myproj/myarchive",
  "@type": "Archive",
  "_self": "http://localhost:8080/v1/archives/myorg/myproj/myarchive",
  "_constrainedBy" : "https://bluebrain.github.io/nexus/schemas/archives.json",
  "_project": "http://localhost:8080/v1/projects/myorg/myproj",
  "_rev": 1,
  "_deprecated": false,
  "_createdAt": "2021-05-17T14:54:42.939Z",
  "_createdBy": "http://localhost:8080/v1/realms/myrealm/users/john",
  "_updatedAt": "2021-05-17T14:54:42.939Z",
  "_updatedBy": "http://localhost:8080/v1/realms/myrealm/users/john"
}

Note that if the payload contains an @id different from the {archive_id}, the request will fail.

Fetch

When fetching an archive, the response format can be chosen through HTTP content negotiation. In order to fetch the archive metadata, the client can use any of the following MIME types. However, in order to fetch the archive content, the HTTP Accept header should be provided as application/zip.

@@@

When downloading the archive, it is possible to ignore resources that does not exist by appending the query parameter ignoreNotFound=true to the fetch link.

GET /v1/archives/{org_label}/{project_label}/{archive_id}
GET /v1/archives/{org_label}/{project_label}/{archive_id}?ignoreNotFound=true

Example

Request (ZIP)

copysourcecurl "http://localhost:8080/v1/archives/myorg/myproject/myarchive" \
     -H "Accept: application/x-zip" \
     -o output.zip

Request (metadata)

copysourcecurl "http://localhost:8080/v1/archives/myorg/myproject/myarchive" \
     -H "Accept: application/json"

Response

copysource{
  "@context": [
    "https://bluebrain.github.io/nexus/contexts/archives.json",
    "https://bluebrain.github.io/nexus/contexts/metadata.json"
  ],
  "@id": "http://localhost:8080/v1/archives/myorg/myproj/myarchive",
  "@type": "Archive",
  "resources" : [
    {
      "@type": "Resource",
      "resourceId": "http://localhost:8080/resource1",
      "rev": 2
    },
    {
      "@type": "Resource",
      "resourceId": "http://localhost:8080/resource2",
      "project": "myorg/myproject2",
      "format": "expanded"
    },
    {
      "@type": "File",
      "resourceId": "http://localhost:8080/resource2",
      "project": "myorg/myproject2",
      "path": "my/custom/path/resource2.json"
    }
  ],
  "_self": "http://localhost:8080/v1/archives/myorg/myproj/myarchive",
  "_constrainedBy": "https://bluebrain.github.io/nexus/schemas/archives.json",
  "_project": "http://localhost:8080/v1/projects/myorg/myproj",
  "_rev": 1,
  "_deprecated": false,
  "_createdAt": "2021-05-17T14:54:42.939Z",
  "_createdBy": "http://localhost:8080/v1/realms/myrealm/users/john",
  "_updatedAt": "2021-05-17T14:54:42.939Z",
  "_updatedBy": "http://localhost:8080/v1/realms/myrealm/users/john",
  "_expiresInSeconds": 17530
}