#!/bin/bash

# Generate EuroVoc dictionary
python get_eurovoc_dict.py

# Create output directory if it doesn't exist
mkdir -p outs_id/

# Process .jsonl.gz files in outs_id/
for input_file in outs/*.jsonl.gz; do
    filename=$(basename "$input_file" .gz)
    temp_file="outs_id_correct/${filename}.tmp"
    output_file="outs_id_correct/${filename}.gz"

    echo "Processing $filename..."

    gunzip -c "$input_file" | python add_id.py /dev/stdin "$temp_file"

    if [ -s "$temp_file" ]; then
        gzip "$temp_file" -c > "$output_file"
        rm -f "$temp_file"
        echo "✓ Saved to $output_file"
    else
        echo "✗ Error: temp file is empty"
        rm -f "$temp_file"
    fi
done

echo "All files processed and compressed!"