Pass in argument of directory for find command to ignore in script

When I call the script (via commandline) I'd like to pass in an optional list of directories for the find command to ignore. This is what I have but my output includes data from files in the directory that is supposed to be ignored. This script works as expected without a directory passed in to ignore.

#!/usr/bin/env bash

# Ensure the script is called with the correct number of arguments
if [ "$#" -lt 2 ]; then
    echo "Usage: $0 <directory> <output_filename> [directories_to_ignore...]"
    exit 1
fi

# Convert the directory to an absolute path
directory=$(realpath "$1")
output_file=$2
shift 2  # Shift the arguments so we can access any ignored directories

# Collect directories to ignore (if provided)
ignore_dirs=()
for dir in "$@"; do
    ignore_dirs+=("-path" "$dir" -prune -o)
done

temp_file=$(mktemp)

# Recursively find all parquet files and save their absolute paths to a temp file
find "$directory" "${ignore_dirs[@]}" -type f -exec file "{}" \; | ug -i -e 'apache parquet' | awk -F: '{print $1}' >> "$temp_file"

# Declare an associative array to store unique MD5s
declare -A md5_dict

# Iterate over each file in the temp file
while IFS= read -r parquet_file; do
    # Extract just the filename from the absolute path
    filename=$(basename "$parquet_file")

    # Search the file for the regex pattern and loop through each match
    matches=$(pqrs cat $parquet_file | sed '/^#\{40,\}/d' | rg -o "md5=\w{32}" | sed 's/md5=//' | sort | uniq)
    
    # Write each unique match on a new line in the output file with the filename prefixed
    while IFS= read -r match; do
        # Extract the actual hash value from the match (remove "md5=")
        hash_value=$(echo "$match")

        # Check if the MD5 is already in the dictionary
        if [[ -z "${md5_dict[$hash_value]}" ]]; then
            # If not in the dictionary, add it and write to the output file
            md5_dict["$hash_value"]="$filename"
            echo "$filename,$match" >> "$output_file"
        fi
    done <<< "$matches"
done < "$temp_file"

# Clean up the temporary file
rm "$temp_file"

echo "Search complete. Results saved to $output_file."

Edited to upload correct version of the script.

UPDATE: Here is the directory structure I'm working with

.
├── data
│   ├── outputFiles
│   └── parquetFiles
│       ├── 2023-08-06-07-00_1691276460134_127.0.0.1
│       ├── 2023-08-06-07-01_1691276522106_127.0.0.1
│       ├── ignoreThisDir
│       │   ├── 2023-08-06-07-02_1691276582057_127.0.0.1
│       │   └── testFileNottoSearch.txt
│       └── testFiletoSearch.txt
├── db
│   └── docker-compose.yaml
├── dehasher.sh
├── find_pg_files.sh
├── source
└── test.sh

When I run the script I expect it to scrape values out of the two files in data/parquetFiles and ignore the file in data/parquetFiles/ignoreThisDir. When I run the script and check the output file using ripgrep rg "2023-08-06-07-02_1691276582057" ouput there are over 200 matches.