transtractor

The Transtractor

Universal PDF bank statement parsing

The Transaction Extractor, or 'Transtractor', aspires to be a universal library for extracting transaction data from PDF bank statements. Key features:

• Written in Rust (fast)
• Python API (user friendly)
• AI-free (lightweight)
• Rules-based extraction (100% predictable and accurate)

Installation

Compile from source

• Install Rust: Download and install Rust from rustup.rs

• Install Maturin: Install the Python build tool for Rust extensions

  pip install maturin
  

• Build and install Transtractor: Clone the repository and build

  git clone https://github.com/gravytoast/transtractor.git
  cd transtractor
  maturin develop --release
  

Basic usage (Python)

• Import and initialise the parser

  from transtractor import Parser
  
  parser = Parser()
  

• Convert PDF to CSV: All CSV files are written in a standard format

  parser.parse('statement.pdf').to_csv('statement.csv')
  

• Convert PDF to DataFrame: Load into a DataFrame for analysis

  import pandas as pd
  
  data = parser.parse('statement.pdf').to_pandas_dict()
  df = pd.DataFrame(data)
  

Supported statements

See the documentation for a current list of supported statements. You may also create your own parsing configuration files by following these instructions and loading it by:

from transtractor import Parser

parser = Parser()
parser.load('my_config.json')
parser.parse('statement.pdf').to_csv('statement.csv')

Contributions

New and well-tested configuration files are especially welcome. Please submit a pull request with them add to the python/transtractor/configs directory, or email to [e-mail].