Upload and reference resources in Dataphin - Dataphin - Alibaba Cloud Documentation Center

Upload resource files such as JAR, JSON, and Python to Dataphin and reference them in compute tasks.

Create resources

On the Dataphin homepage, click Development in the top menu bar.
Open the New Resource dialog box:

Select Project (in Dev-Prod mode, also select the environment) > click Resource > click the icon.

In the New Resource dialog box, set the parameters:

Parameter	Description
Type	Select the file Type: file (.xls, .xlsx, .doc, .docx, .txt, .csv), archive (.zip, .tgz, .tar.gz, .tar, .jar), jar, Python, cplus, and others. file: Text files used for task data. archive: Compressed packages used for task dependencies. Python: Python files used as task dependencies. cplus: C++ source files for registering UDFs in Impala tasks.
Name	Enter the resource Name. Rules: Must be unique within the project and cannot start with a number. Can contain numbers, letters, underscores (_), hyphens (-), and periods (.). Must be 3 to 200 characters long.
Description	A brief description of the resource, up to 1000 characters.
Upload File	Upload File matching the selected Type.
Compute Type	Select where to upload the resource. Compute engine: Uploads the resource to the project's compute engine (such as MaxCompute), typically for UDF creation. For JAR files, the system runs `add jar xxx.jar`. No affiliated engine: Uploads the resource only to Dataphin. Suitable for file and archive types.
Select Directory	Select the storage directory. To create a directory, enter its name and select a location.

Click Submit.
In the Submit Remarks dialog box, enter remarks (up to 128 characters).
Click Confirm And Submit.

Note
In Dev-Prod mode, you must publish resources to the production environment. Manage publishing tasks.

Reference resources

Reference methods

After uploading, copy the reference statement for use in compute tasks:
Or add @resource_reference{"resource name"} directly in your compute task code.

How resource references work

@resource_reference{"resource name"} defines the resource file path as a variable. Example:

@resource_reference{"dataphin.xls"} # Reference dataphin.xls resource

# Import the xlrd module.
import xlrd
wb = xlrd.open_workbook('dataphince.xls') # Open excel
sh = wb.sheet_by_name('Sheet1') # Locate the worksheet by workbook

# Traverse excel and print all data
for i in range(sh.nrows):
    print(sh.row_values(i))

@resource_reference{"dataphin.xls"} is equivalent to dataphin.xls = "/path/to/dataphin.xls". Use dataphin.xls as the file path in your code.