Assembling Parts (SDK for Python)

Function

This API assembles the uploaded parts to compete the multipart upload. Before performing this operation, you cannot download the uploaded data. When assembling parts, you need to copy the additional message header information recorded during the multipart upload initiation to the object metadata. Such information is processed the same way the information in a common object upload is processed. In the case of assembling parts concurrently, last write wins is applied, but the time of last write is defined as the time when a multipart upload was initiated.

As long as the multipart upload is not aborted, all uploaded parts occupy the space. However, after you assembled the specified parts, those uploaded but not assembled will be deleted to free up space.

When assembling parts, OBS creates an object by putting part numbers in ascending order. If any object metadata is provided in the initiation of the multipart upload, OBS will associate the metadata with the object. After the multipart upload is complete, the parts will no longer exist. A part assembling request must contain the upload ID, part numbers, and a list of corresponding ETag values. In response to the request, the ETag that uniquely identifies the assembled parts is contained. This ETag is not the MD5 hash value of the entire object.

Restrictions

To assemble parts, you must be the bucket owner or have the required permission (obs:object:PutObject in IAM or PutObject in a bucket policy). For details, see Introduction to OBS Access Control, IAM Custom Policies, and Configuring an Object Policy.
The mapping between OBS regions and endpoints must comply with what is listed in Regions and Endpoints.
After a multipart upload is complete, the uploaded parts that are not assembled will be automatically deleted and cannot be recovered. Before assembling parts, use the API for listing uploaded parts to check all parts to ensure that no part is missed.

Method

ObsClient.completeMultipartUpload(bucketName, objectKey, uploadId, completeMultipartUploadRequest, extensionHeaders)

Request Parameters

**Table 1** List of request parameters
Parameter	Type	Mandatory (Yes/No)	Description
bucketName	str	Yes	Explanation: Bucket name Restrictions: A bucket name must be unique across all accounts and regions. A bucket name: Must be 3 to 63 characters long and start with a digit or letter. Lowercase letters, digits, hyphens (-), and periods (.) are allowed. Cannot be formatted as an IP address. Cannot start or end with a hyphen (-) or period (.). Cannot contain two consecutive periods (..), for example, my..bucket. Cannot contain periods (.) and hyphens (-) adjacent to each other, for example, my-.bucket or my.-bucket. If you repeatedly create buckets of the same name in the same region, no error will be reported and the bucket attributes comply with those set in the first creation request. Default value: None
objectKey	str	Yes	Explanation: Object name. An object is uniquely identified by an object name in a bucket. An object name is a complete path that does not contain the bucket name. For example, if the address for accessing the object is examplebucket.obs.ap-southeast-1.myhuaweicloud.com/folder/test.txt, the object name is folder/test.txt. Value range: The value must contain 1 to 1,024 characters. Default value: None NOTE: The object URL is in the following format: https://Bucket name.Domain name/Folder directory level/Object name. If this object is stored in the root directory of the bucket, its URL does not contain the folder directory level.
uploadId	str	Yes	Explanation: Multipart upload ID, for example, 000001648453845DBB78F2340DD460D8 Value range: The value must contain 32 characters. Default value: None
completeMultipartUploadRequest	CompleteMultipartUploadRequest	Yes	Explanation: Request parameters for assembling parts. For details, see Table 2.
extensionHeaders	dict	No	Explanation: Extension headers. Value range: See User-defined Header (SDK for Python). Default value: None
encoding_type	str	No	Explanation: Encoding type for key in the response. If key in the response contains control characters that are not supported by the XML 1.0 standard, you can specify this parameter to encode key. Value range: url Default value: None. If you leave this parameter blank, encoding is not applied to Key.

**Table 2** CompleteMultipartUploadRequest
Parameter	Type	Mandatory (Yes/No)	Description
parts	list of CompletePart	Yes	Explanation: List of parts to be assembled. For details, see Table 3.

**Table 3** CompletePart
Parameter	Type	Mandatory (Yes/No)	Description
partNum	int	Yes	Explanation: Part number Value range: [1,10000] Default value: None
etag	str	Yes	Explanation: ETag of a part. It is calculated by encoding the 128-bit MD5 value of the part using Base64. Value range: The value must contain 32 characters. Default value: None

Responses

**Table 4** List of returned results
Type	Description
GetResult	Explanation: SDK common results

**Table 5** GetResult
Parameter	Type	Description
status	int	Explanation: HTTP status code Value range: A status code is a group of digits ranging from 2xx (indicating successes) to 4xx or 5xx (indicating errors). It indicates the status of a response. For more information, see Status Code. Default value: None
reason	str	Explanation: Reason description. Default value: None
errorCode	str	Explanation: Error code returned by the OBS server. If the value of status is less than 300, this parameter is left blank. Default value: None
errorMessage	str	Explanation: Error message returned by the OBS server. If the value of status is less than 300, this parameter is left blank. Default value: None
requestId	str	Explanation: Request ID returned by the OBS server Default value: None
indicator	str	Explanation: Error indicator returned by the OBS server. Default value: None
hostId	str	Explanation: Requested server ID. If the value of status is less than 300, this parameter is left blank. Default value: None
resource	str	Explanation: Error source (a bucket or an object). If the value of status is less than 300, this parameter is left blank. Default value: None
header	list	Explanation: Response header list, composed of tuples. Each tuple consists of two elements, respectively corresponding to the key and value of a response header. Default value: None
body	object	Explanation: Result content returned after the operation is successful. If the value of status is larger than 300, the value of body is null. The value varies with the API being called. For details, see Bucket-Related APIs (SDK for Python) and Object-Related APIs (SDK for Python). Default value: None

**Table 6** GetResult.body
GetResult.body Type	Description
CompleteMultipartUploadResponse	Explanation: Response to the request for assembling parts. For details, see Table 7.

**Table 7** CompleteMultipartUploadResponse
Parameter	Type	Description
etag	str	Explanation: The ETag that uniquely identifies the object after its parts were assembled, calculated based on the ETag of each part. Restrictions: If an object is encrypted using server-side encryption, the ETag is not the MD5 value of the object. Value range: The value must contain 32 characters. Default value: None
bucket	str	Explanation: Bucket in which parts are assembled Restrictions: A bucket name must be unique across all accounts and regions. A bucket name: Must be 3 to 63 characters long and start with a digit or letter. Lowercase letters, digits, hyphens (-), and periods (.) are allowed. Cannot be formatted as an IP address. Cannot start or end with a hyphen (-) or period (.). Cannot contain two consecutive periods (..), for example, my..bucket. Cannot contain periods (.) and hyphens (-) adjacent to each other, for example, my-.bucket or my.-bucket. If you repeatedly create buckets of the same name in the same region, no error will be reported and the bucket attributes comply with those set in the first creation request. Default value: None
key	str	Explanation: Object name obtained after part assembling. An object is uniquely identified by an object name in a bucket. An object name is a complete path that does not contain the bucket name. For example, if the address for accessing the object is examplebucket.obs.ap-southeast-1.myhuaweicloud.com/folder/test.txt, the object name is folder/test.txt. Value range: The value must contain 1 to 1,024 characters. Default value: None
location	str	Explanation: URL of the generated object after part assembling Example: https://example-Bucket.obs.regions.myhuaweicloud.com/example-Object Default value: None
versionId	str	Explanation: Version ID of the object obtained after part assembling Value range: The value must contain 32 characters. Default value: None
sseKms	str	Explanation: SSE-KMS algorithm Value range: kms Default value: None
sseKmsKey	str	Explanation: ID of the KMS master key when SSE-KMS is used Value range: Valid value formats are as follows: regionID:domainID:key/key_id key_id In the preceding formats: regionID indicates the ID of the region where the key is used. You can obtain it from Regions and Endpoints. domainID indicates the ID of the account that the key is for. To obtain it, see How Do I Get My Account ID and IAM User ID? (SDK for Python) key_id indicates the ID of the key created on Data Encryption Workshop (DEW). To obtain it, see Viewing a CMK. Default value: If this parameter is not specified, the default master key will be used. If there is no such a default master key, OBS will create one and use it by default.
sseC	str	Explanation: SSE-C algorithm Value range: AES256 Default value: None
sseCKeyMd5	str	Explanation: MD5 value of the key for encrypting objects when SSE-C is used. This value is used to check whether any error occurs during the transmission of the key. Restrictions: The value is encrypted by MD5 and then encoded by Base64, for example, 4XvB3tbNTN+tIEVa0/fGaQ==. Default value: None
objectUrl	str	Explanation: Full path to the obtained object after part assembling Default value: None
encoding_type	str	Explanation: Encoding type for Key in the response. If Key in the response contains control characters that are not supported by the XML 1.0 standard, you can specify this parameter to encode Key. Value range: url Default value: None. If you leave this parameter blank, encoding is not applied to Key.

Code Examples

This example assembles the parts previously uploaded to a bucket.

    
     
       
       
         from obs import ObsClient
from obs import CompleteMultipartUploadRequest, CompletePart
import os
import traceback

# Obtain an AK and SK pair using environment variables or import the AK and SK pair in other ways. Using hard coding may result in leakage.
# Obtain an AK and SK pair on the management console. For details, see https://support.huaweicloud.com/intl/en-us/usermanual-ca/ca_01_0003.html.
ak = os.getenv("AccessKeyID")
sk = os.getenv("SecretAccessKey")
# (Optional) If you use a temporary AK and SK pair and a security token to access OBS, obtain them from environment variables.
# security_token = os.getenv("SecurityToken")
# Set server to the endpoint corresponding to the bucket. CN-Hong Kong is used here as an example. Replace it with the one in use.
server = "https://obs.ap-southeast-1.myhuaweicloud.com" 

# Create an obsClient instance.
# If you use a temporary AK and SK pair and a security token to access OBS, you must specify security_token when creating an instance.
obsClient = ObsClient(access_key_id=ak, secret_access_key=sk, server=server)
try:
    # Specify the list of parts to be assembled by configuring completeMultipartUploadRequest. Each part contains partNum and etag.
    part1 = CompletePart(partNum=1, etag='etag1')
    part2 = CompletePart(partNum=2, etag='etag2')
    completeMultipartUploadRequest = CompleteMultipartUploadRequest(parts=[part1, part2])

    bucketName = "examplebucket"
    objectKey = "objectname"
    # Specify the ID of the multipart upload.
    uploadId = "your uploadid"
    # Assemble the parts uploaded to the bucket.
    resp = obsClient.completeMultipartUpload(bucketName, objectKey, uploadId, completeMultipartUploadRequest, encoding_type='url')

    # If status code 2xx is returned, the API is called successfully. Otherwise, the API call fails.
    if resp.status < 300:
        print('Complete Multipart Upload Succeeded')
        print('requestId:', resp.requestId)
        print('etag:', resp.body.etag)
        print('bucket:', resp.body.bucket)
        print('key:', resp.body.key)
        print('location:', resp.body.location)
        print('versionId:', resp.body.versionId)
    else:
        print('Complete Multipart Upload Failed')
        print('requestId:', resp.requestId)
        print('errorCode:', resp.errorCode)
        print('errorMessage:', resp.errorMessage)
except:
    print('Complete Multipart Upload Failed')
    print(traceback.format_exc())

        

      

    
   