Skip to content

Latest commit

 

History

History
108 lines (77 loc) · 4.28 KB

README.md

File metadata and controls

108 lines (77 loc) · 4.28 KB
Raw

English | 中文

@huolala-tech/page-spy-plugin-data-harbor

SDK version SDK size

The DataHarborPlugin can be used in browser environment, providing offline caching and data downloading capabilities.

In the past, remote debugging had a prerequisite that "both the client and debugging server must be online simultaneously." By utilizing the DataHarborPlugin, it internally listens for the "public-data" event (what is the public-data event?), allowing the offline caching of data. This innovation breaks the previous prerequisite of both the client and debugging server needing to be online simultaneously. When issues are identified by the client, testing colleagues can directly export the data, marking a significant departure from the historical requirement of simultaneous online status for both the client and debugging server.

DataHarborPlugin starts collecting data when new PageSpy() is called. The collected data is stored in client-side memory. When the accumulated data volume reaches 10MB (default value, specified by maximum), the data is temporarily stored by generating an Object URL through URL.createObjectUrl() and then new data collection continues.

Definition

import { PageSpyPlugin } from '@huolala-tech/page-spy-types';

type DataType = 'console' | 'network' | 'system' | 'storage' | 'rrweb-event';

interface DataHarborConfig {
  // Specify the maximum bytes of single harbor's container.
  // Default 10 * 1024 * 1024.
  maximum?: number;

  // Specify which types of data should be collected
  caredData?: Record<DataType, boolean>;

  // Customize the log's filename
  filename?: () => string;

  // Customize the "Download Log Data"
  // (Version required: @huolala-tech/page-spy-plugin-data-harbor^1.0.6)
  onDownload?: (data: CacheMessageItem[]) => void;
}

declare class DataHarborPlugin implements PageSpyPlugin {
  constructor(config?: DataHarborConfig);
}

export default DataHarborPlugin;

Usage

Load plugin

  • Options 1: Load with script

    <html>
  <head>
    <!-- 1. Load PageSpy -->
    <script src="https://<your-host>/page-spy/index.min.js"></script>
    <!-- 2. Load the plugin -->
    <script src="https://<your-host>/plugin/data-harbor/index.min.js"></script>
    <!-- 3. Register plugin && Init PageSpy -->
    <script>
      // Register plugin
      window.$harbor = new DataHarborPlugin(config: DataHarborConfig);
      PageSpy.registerPlugin(window.$harbor);
      // Init PageSpy
      window.$pageSpy = new PageSpy();
    </script>
  </head>
</html>

  • Option 2: Import within ESM

    // In your entry file like "main.ts"
import PageSpy from '@huolala-tech/page-spy-browser';
import DataHarborPlugin from '@huolala-tech/page-spy-plugin-data-harbor';

// Register plugin
window.$harbor = new DataHarborPlugin(config: DataHarborConfig);
PageSpy.registerPlugin(window.$harbor);
// Init PageSpy
window.$pageSpy = new PageSpy();

Upload / download data

Upon successful integration, the popup that appears after clicking on the client-side rendering control should include download / upload buttons.

Download

If setting autoRender: false when instantiating PageSpy hides the controls, but you still want the client to be able to upload/download without any action required, you can manipulate it by sending the following code on the debugging panel:

// upload
const url = await window.$harbor.onOfflineLog('upload');

// download
window.$harbor.onOfflineLog('download');

Replay log

Goto the debugger's log list page and click the "Replay log" button, select the file uploaded / downloaded in prev step, now you can use the replay to debug.

Entry