Querying a DataFrame using python-oracledb from the video "Python DataFrames with Oracle Database for Analysis and AI" https://youtu.be/0BJNlbh71LY For other examples, see https://github.com/oracle/python-oracledb/tree/main/samples

pandas-new.py

# pandas-new.py
#
# For other python-oracledb DataFrame examples, see https://github.com/oracle/python-oracledb/tree/main/samples

import os

import pyarrow
import oracledb

un = os.environ.get('ORACLE_USERNAME')
pw = os.environ.get('ORACLE_PASSWORD')
cs = os.environ.get('ORACLE_DSN')

connection = oracledb.connect(user=un, password=pw, dsn=cs)

#-------------------------------------------------------------------------------

sql = """select employee_id, department_id, salary
         from employees
         where department_id < :id
         order by department_id"""

# The new, fast way to get a DataFrame
odf = connection.fetch_df_all(
    statement=sql,
    parameters=[80],
    arraysize=1000
)
# Use PyArrow to do zero-copy conversion to our chosen package's DataFrame
df = pyarrow.table(odf).to_pandas()

#-------------------------------------------------------------------------------

print()

# Sum of salaries in each department
deptsum = df.groupby('DEPARTMENT_ID').sum()
print(deptsum)

print()