Store data in OneLake

Every Fabric item has access to OneLake storage, which provides a secure, scalable way to store files associated with your item. This guide explains how to upload files to your Fabric item using the FabricPlatformAPIClient and the underlying OneLakeClient.

Understanding item storage in Fabric

Each item in Fabric has its own dedicated storage area in OneLake. This storage is organized into folders, with the main folders being:

• Files - For storing general files and documents
• Tables - For storing table data

Prerequisites

Before uploading files to your item, you need:

• A valid Fabric workspace
• An existing item where you want to upload the file
• Appropriate permissions to write to the item

Creating the FabricPlatformAPIClient

First, create a FabricPlatformAPIClient instance:

import { getWorkloadClient } from "../controller/WorkloadClient";
import { FabricPlatformAPIClient } from "../clients/FabricPlatformAPIClient";

// Create client using the current user's context
const fabricClient = FabricPlatformAPIClient.create(getWorkloadClient());

Example: Uploading a text file to an item

Here's how to upload a text file to an item using the OneLake client:

async function uploadTextFileToItem(
  workspaceId: string, 
  itemId: string, 
  fileName: string, 
  content: string
) {
  try {
    // Get the FabricPlatformAPIClient
    const fabricClient = FabricPlatformAPIClient.create(getWorkloadClient());
    
    // Access the OneLake client
    const oneLakeClient = fabricClient.oneLake;
    
    // Generate the file path in OneLake for this item
    // This follows the pattern: workspaceId/itemId/Files/fileName
    const filePath = oneLakeClient.constructor.getFilePath(workspaceId, itemId, fileName);
    
    // Write the text content to the file
    await oneLakeClient.writeFileAsText(filePath, content);
    
    console.log(`Successfully uploaded ${fileName} to item ${itemId}`);
    return true;
  } catch (error) {
    console.error("Error uploading file to item:", error);
    throw error;
  }
}

Example: Uploading a binary file to an item

For binary files like images or PDFs, you need to convert the file to base64 first:

async function uploadBinaryFileToItem(
  workspaceId: string,
  itemId: string,
  fileName: string,
  fileData: ArrayBuffer // Binary file data
) {
  try {
    const fabricClient = FabricPlatformAPIClient.create(getWorkloadClient());
    const oneLakeClient = fabricClient.oneLake;
    
    // Convert binary data to base64
    const base64Content = arrayBufferToBase64(fileData);
    
    // Generate the file path
    const filePath = oneLakeClient.constructor.getFilePath(workspaceId, itemId, fileName);
    
    // Write the binary content to the file
    await oneLakeClient.writeFileAsBase64(filePath, base64Content);
    
    console.log(`Successfully uploaded binary file ${fileName} to item ${itemId}`);
    return true;
  } catch (error) {
    console.error("Error uploading binary file to item:", error);
    throw error;
  }
}

// Helper function to convert ArrayBuffer to base64
function arrayBufferToBase64(buffer: ArrayBuffer): string {
  let binary = '';
  const bytes = new Uint8Array(buffer);
  const len = bytes.byteLength;
  
  for (let i = 0; i < len; i++) {
    binary += String.fromCharCode(bytes[i]);
  }
  
  return btoa(binary);
}

Example: Uploading a file from the browser

If you're building a web interface, you can use this function to handle file uploads from a file input:

async function handleFileUpload(
  workspaceId: string,
  itemId: string,
  fileInputElement: HTMLInputElement
) {
  if (!fileInputElement.files || fileInputElement.files.length === 0) {
    console.warn("No file selected");
    return false;
  }
  
  const file = fileInputElement.files[0];
  const fileName = file.name;
  
  try {
    // Read the file as ArrayBuffer
    const fileBuffer = await readFileAsArrayBuffer(file);
    
    // Upload based on file type
    if (file.type.startsWith('text/')) {
      // For text files, convert to string and upload as text
      const textDecoder = new TextDecoder();
      const textContent = textDecoder.decode(fileBuffer);
      
      return await uploadTextFileToItem(workspaceId, itemId, fileName, textContent);
    } else {
      // For binary files, upload as base64
      return await uploadBinaryFileToItem(workspaceId, itemId, fileName, fileBuffer);
    }
  } catch (error) {
    console.error("Error processing file upload:", error);
    throw error;
  }
}

// Helper function to read file as ArrayBuffer
function readFileAsArrayBuffer(file: File): Promise<ArrayBuffer> {
  return new Promise((resolve, reject) => {
    const reader = new FileReader();
    reader.onload = () => resolve(reader.result as ArrayBuffer);
    reader.onerror = reject;
    reader.readAsArrayBuffer(file);
  });
}

Managing files in your item

Once files are uploaded, you can also:

Check if a file exists

async function checkFileExists(workspaceId: string, itemId: string, fileName: string) {
  const fabricClient = FabricPlatformAPIClient.create(getWorkloadClient());
  const filePath = fabricClient.oneLake.constructor.getFilePath(workspaceId, itemId, fileName);
  
  return await fabricClient.oneLake.checkIfFileExists(filePath);
}

Read file content

async function readTextFile(workspaceId: string, itemId: string, fileName: string) {
  const fabricClient = FabricPlatformAPIClient.create(getWorkloadClient());
  const filePath = fabricClient.oneLake.constructor.getFilePath(workspaceId, itemId, fileName);
  
  return await fabricClient.oneLake.readFileAsText(filePath);
}

Delete a file

async function deleteFile(workspaceId: string, itemId: string, fileName: string) {
  const fabricClient = FabricPlatformAPIClient.create(getWorkloadClient());
  const filePath = fabricClient.oneLake.constructor.getFilePath(workspaceId, itemId, fileName);
  
  await fabricClient.oneLake.deleteFile(filePath);
  console.log(`File ${fileName} deleted successfully`);
}

Best practices for file uploads

1. Use appropriate file formats: Consider the purpose of the file and use formats that are widely supported.
2. Handle errors gracefully: Always include error handling for network issues or permission problems.
3. Validate file sizes: Large files may take longer to upload and process.
4. Check permissions: Ensure the user has appropriate permissions before attempting uploads.
5. Use file prefixes or folders: For complex items with many files, consider organizing them in subfolders.

Using the OneLakeClientItemWrapper

For simplified access to item files, you can use the OneLakeClientItemWrapper:

async function uploadFileWithItemWrapper(item, fileName, content) {
  const fabricClient = FabricPlatformAPIClient.create(getWorkloadClient());
  
  // Create a wrapper for simpler access to this specific item
  const itemWrapper = fabricClient.oneLake.createItemWrapper({
    workspaceId: item.workspaceId,
    itemId: item.id
  });
  
  // Upload directly to the item (no need to specify paths)
  await itemWrapper.writeFileAsText(fileName, content);
  
  // Read the file back
  const fileContent = await itemWrapper.readFileAsText(fileName);
  
  console.log(`File uploaded and read back: ${fileContent.substring(0, 50)}...`);
}

This wrapper simplifies file operations by automatically handling the full path construction.