setup.sh

#!/bin/bash
# setup.sh - Setup script for Hugging Face Spaces

set -e  # Exit on any error

echo "🚀 Setting up OCR dependencies for Hugging Face Spaces..."

# Update package list
echo "📦 Updating package list..."
apt-get update -qq

# Install system dependencies for OCR
echo "🔧 Installing system dependencies..."
apt-get install -y -qq tesseract-ocr tesseract-ocr-ben tesseract-ocr-eng poppler-utils

# Verify Tesseract installation
echo "✅ Verifying Tesseract installation..."
tesseract --version

# Check available languages
echo "🌐 Available Tesseract languages:"
tesseract --list-langs

# Create tessdata directory if it doesn't exist
TESSDATA_DIR="/usr/share/tesseract-ocr/4.00/tessdata"
if [ ! -d "$TESSDATA_DIR" ]; then
    echo "📁 Creating tessdata directory..."
    mkdir -p "$TESSDATA_DIR"
fi

# Download additional language data if needed
echo "📥 Ensuring language data is available..."

# Bengali language support
if [ ! -f "$TESSDATA_DIR/ben.traineddata" ]; then
    echo "🇧🇩 Installing Bengali language data..."
    wget -q -O "$TESSDATA_DIR/ben.traineddata" \
        https://github.com/tesseract-ocr/tessdata/raw/main/ben.traineddata || \
        echo "⚠️ Warning: Could not download Bengali language data"
fi

# English language support
if [ ! -f "$TESSDATA_DIR/eng.traineddata" ]; then
    echo "🇺🇸 Installing English language data..."
    wget -q -O "$TESSDATA_DIR/eng.traineddata" \
        https://github.com/tesseract-ocr/tessdata/raw/main/eng.traineddata || \
        echo "⚠️ Warning: Could not download English language data"
fi

# Set permissions
chmod -R 755 "$TESSDATA_DIR" 2>/dev/null || true

echo "🎉 Setup completed successfully!"
echo "📊 Final verification:"
tesseract --list-langs | grep -E "(ben|eng)" || echo "⚠️ Some languages may not be available"