Dataset Management

This guide is for developers and explains the Dataset feature of the EO Platform. A dataset is a logical collection of metadata. You can use it to organize, manage, and retrieve multiple pieces of data related to a project, such as archiving images from the same region or time series into a single dataset.

1. Prerequisites

Before you begin the integration, please ensure that the following environment is ready:

• Account Permissions: You need access and operational rights for the "Dataset Mgt" module.
• Input Data: The images, vectors, and other data to be added to the dataset should already exist in the "Data Storage" module.
• Running Services: Ensure that data_management_service is running correctly.

---

2. Core Workflow

2.1 Create a Dataset

1. Navigate to the Dataset Mgt page from the left-side navigation bar.
2. Click the Add Dataset button in the upper-right corner of the page to open the creation dialog.
3. Fill in the form fields:

Field	Required	Description
DataSet Name	Yes	The display name of the dataset, ≤ 50 characters.
Select Tag	No	Associate one or more tags with the dataset for easy categorization and retrieval.
Description	No	A detailed description of the dataset.
Cover	No	Upload an image as the dataset's cover to enhance recognition.

4. Click Save to complete the creation by calling POST /dataset/create.

2.2 View the Dataset List

• After successful creation, the new dataset will appear in the list, sorted by creation time in descending order by default.
• The list page displays key information about the datasets, such as name, cover, tags, description, creator, and creation time.
• You can quickly filter by name using the search box at the top or browse using the pagination controls.
• The list data is provided by POST /dataset/query/page.

2.3 View Details and Manage Metadata

1. Click on the name of any dataset in the list page to go to its details page.
2. The top of the details page displays basic information about the dataset, such as its name, Client ID, Secret Key, description, and capture time.
3. The All Metadata area below lists all the metadata (data files) that have been associated with this dataset.
   • Here you can view the status of the associated data (publishing, published, failed, etc.).
   • You can use the buttons on the right to edit, share, delete, etc., individual metadata items.
4. Click the Add Metadata button to select one or more data files from "Data Storage" and associate them with the current dataset by calling POST /metadata/create.

---

3. API Quick Index

Capability	API	Description
Create Dataset	POST /dataset/create	Creates a new dataset.
Paginated Query	POST /dataset/query/page	Retrieves a list of datasets, with support for pagination and filtering.
Query Details	GET /dataset/detail	Retrieves detailed information about a single dataset and its associated metadata list.
Update Dataset	POST /dataset/update	Modifies the name, description, cover, etc., of a dataset.
Delete Dataset	DELETE /dataset/delete	Deletes a dataset (does not delete its associated metadata files).
Associate Metadata	POST /metadata/create	Associates a data file from "Data Storage" with a dataset.

---

4. Debugging and Troubleshooting

• Creation Failed: Please check if the DataSet Name is filled in and if the network request is normal.
• Details Page Fails to Load: Confirm that the dataset ID is correct and that the GET /dataset/detail API can return data.
• Failed to Associate Metadata: Check if the metadata to be associated exists in "Data Storage" and confirm that the current account has permission to access this data.
• API 401/403: Check the Token in the Authorization request header and confirm that the account has dataset management permissions.