Text Extraction Tool - Sensible Test

Overview

This tool extracts text based on positional and alignment criteria from standardized text outputs, such as PDFs. It supports two main extraction methods: Label Extraction and Row Extraction, allowing users to specify an anchor and direction for precise text extraction.

Setup

To set up the Text Extraction Tool, follow these steps:

1. Ensure Node.js is Installed:

   • Visit the Node.js official website and download the installer for your operating system. It's recommended to install the LTS version.
   • Follow the installation prompts to install Node.js and npm.
   • Verify the installation by running node -v and npm -v in your terminal or command prompt.

2. Clone the Repository:

   • Clone the project repository to your local machine using Git or download the ZIP file and extract it.

3. Install Dependencies:

   • Navigate to the project directory in your terminal or command prompt.
   • Run npm install to install the project's dependencies as listed in package.json. This command will install TypeScript, Jest for testing, and other necessary packages.

Compile and Run

• Compile TypeScript:

  • The project uses TypeScript, which needs to be compiled to JavaScript before execution. You can compile the TypeScript files by running npm run start. This script compiles the TypeScript files and then executes the dist/index.js file.

• Start the Application:

  • Use the npm start command to compile TypeScript files and run the application. This command is defined in package.json under the scripts section and performs both compilation and execution in one step.

Testing

• Run npm test to perform jest tests.

Input Configuration

Edit input.json to switch between label and row extraction methods. The structure for each method is defined as follows:

Label Extraction Type

export interface Label {
  id: "label";
  position: "right" | "left" | "above" | "below";
  textAlignment: "right" | "left";
  anchor: string;
}

Row Extraction Type

export interface Row {
  id: "row";
  position: "right" | "left";
  tiebreaker: number | "last";
  anchor: string;
}

Example input.json:

{
  "method": "label",
  "position": "above",
  "textAlignment": "left",
  "anchor": "Example Anchor"
}

Types Explained

• Label: Targets text adjacent to an anchor.
• Row: Focuses on text at the same vertical level as the anchor.

Functionality

• Filtering: Narrows down lines based on their spatial relation to the anchor.
• Sorting: Orders filtered lines by proximity to the anchor, considering alignment.
• Extraction: Returns the line that best matches the criteria.

Remember to adjust input.json for different documents and extraction needs.