csv to trino static query

csv2trinosql.py

import pandas as pd

if __name__ == '__main__':

    # Reload the CSV file due to code execution state reset
    file_path = '01HKPMHCEM0TFYD6KNPM66XHTT.csv'
    df = pd.read_csv(file_path)

    # Construct the SQL query in the desired format with dynamic field names
    sql_query = "SELECT "
    column_names = df.columns.tolist()
    sql_query += ", ".join([f"{col} AS {col}" for col in column_names]) + " FROM (VALUES "
    value_strs = []

    for index, row in df.iterrows():
        row_values = []
        for col in column_names:
            value = row[col]
            # Format the value based on its type (string or numeric)
            if isinstance(value, str):
                # Add '0x' prefix if it's a contract address without the prefix
                if col == 'contract_address' and not value.startswith('0x'):
                    value = f"0x{value}"
                row_values.append(f"'{value}'")
            else:
                row_values.append(str(value))
        value_str = f"({', '.join(row_values)})"
        value_strs.append(value_str)

    sql_query += ",\n             ".join(value_strs) + ")"

    # Limit the output length to avoid excessively long outputs
    sql_query_complete = sql_query
    sql_query_complete[:1000]  # Display the first 1000 characters of the query for preview

    # Output file path
    output_file_path = 'trino_query.sql'

    # Write the SQL query to a file
    with open(output_file_path, 'w') as file:
        file.write(sql_query)