From 0ac6d213c1b9cbc24313b9ac5d442e29f027b798 Mon Sep 17 00:00:00 2001
From: Alex Schofield <git@ajschof.me>
Date: Wed, 19 Feb 2025 02:51:13 +0000
Subject: update logger names

---
 obfuscator/csv_writer.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'obfuscator/csv_writer.py')

diff --git a/obfuscator/csv_writer.py b/obfuscator/csv_writer.py
index aa5ac3f..099e910 100644
--- a/obfuscator/csv_writer.py
+++ b/obfuscator/csv_writer.py
@@ -4,7 +4,7 @@ from typing import List, Dict
 from obfuscator.logger import get_logger
 
 # Create the logger
-logger = get_logger("CSVWriter")
+logger = get_logger("CSVWRITER")
 
 
 def create_byte_stream(data: List[Dict[str, str]]) -> bytes:
-- 
cgit v1.2.3


From 3c4b66e8490c6fdf93fb8fee735d52c76eb2853b Mon Sep 17 00:00:00 2001
From: Alex Schofield <git@ajschof.me>
Date: Wed, 19 Feb 2025 03:22:26 +0000
Subject: remove annoying comments for better readability of code

---
 cli.py                   | 16 ----------------
 obfuscator/csv_reader.py | 19 -------------------
 obfuscator/csv_writer.py |  1 -
 obfuscator/obfuscate.py  |  5 -----
 obfuscator/utils.py      |  1 -
 test/test_csv_reader.py  | 23 -----------------------
 test/test_obfuscator.py  | 15 ---------------
 7 files changed, 80 deletions(-)

(limited to 'obfuscator/csv_writer.py')

diff --git a/cli.py b/cli.py
index d19c18d..c03ffd1 100644
--- a/cli.py
+++ b/cli.py
@@ -7,7 +7,6 @@ from obfuscator.csv_writer import create_byte_stream
 
 
 def main():
-    # Create an argument parser
     parser = argparse.ArgumentParser(
         prog="GDPR-Obfuscator",
         description="Obfuscate sensitive data stored locally or in an AWS environment",
@@ -17,16 +16,10 @@ def main():
         "-v", "--verbose", action="store_true", help="Enable verbose logging"
     )
 
-    # Require user to either choose a local file or an S3 object
-    # The user can only choose one of these options or the program will exit
-    # If not provided, the program will exit
     loc = parser.add_mutually_exclusive_group(required=True)
     loc.add_argument("-l", "--local", help="Local path to file")
     loc.add_argument("-s", "--s3", help="URI path to file stored in S3")
 
-    # Require user to provide a list of PII fields to obfuscate
-    # e.g. --pii name email_address
-    # If not provided, the program will exit
     parser.add_argument(
         "-p",
         "--pii",
@@ -35,23 +28,17 @@ def main():
         help="List of PII fields to obfuscate, separated by spaces",
     )
 
-    # Parse the arguments
     args = parser.parse_args()
 
-    # If the user chose verbose logging, set the logger to debug
     log_level = "DEBUG" if args.verbose else "INFO"
 
-    # Create the logger
     logger = get_logger("CLI", log_level)
 
-    # Create the CSVReader object
     reader = CSVReader(log_level)
 
-    # Read the CSV data based on the user's choice of local or S3
     if args.local and not args.s3:
         logger.debug("User chose to read CSV from local path")
         data = reader.read_local(args.local)
-        # For debug purposes, log the data read from the CSV
         logger.debug("Contents: " + str(data))
     else:
         logger.debug("User chose to read CSV from S3")
@@ -59,13 +46,10 @@ def main():
         data = reader.read_s3(args.s3)
         logger.debug("Contents: " + str(data))
 
-    # Obfuscate the data based on the user's choice of PII fields
     obfuscated_data = obfuscate(data, args.pii)
-    # For debug purposes, log the obfuscated data as JSON for readability
     logger.debug("Obfuscated data (JSON): " + json.dumps(obfuscated_data, indent=4))
     return create_byte_stream(obfuscated_data)
 
 
-# If the script is run directly (as it should be), call the main function
 if __name__ == "__main__":
     main()
diff --git a/obfuscator/csv_reader.py b/obfuscator/csv_reader.py
index f8dd7d3..2b099c8 100644
--- a/obfuscator/csv_reader.py
+++ b/obfuscator/csv_reader.py
@@ -6,12 +6,6 @@ from typing import List, Dict
 from obfuscator.logger import get_logger
 from obfuscator.utils import Utilities
 
-# Putting the CSV reading components into a class may seem like overkill
-# for a simple script, but it allows for better organization and scalability.
-# @staticmethod is used to define the method without an instance of the class
-# being required. The methods could be defined just as functions, and this
-# may still be changed.
-
 
 class CSVReader:
     """
@@ -21,7 +15,6 @@ class CSVReader:
 
     def __init__(self, log_level=None):
         self.log_level = log_level
-        # Create the logger
         self.logger = get_logger("CSVREADER", log_level)
 
     def read_local(self, path) -> List[Dict[str, str]]:
@@ -29,12 +22,8 @@ class CSVReader:
         A method to read a local CSV file and return the data as a list of
         dictionaries.
         """
-        # Log the path of the file being read for debugging
         self.logger.debug(f"Reading local CSV from: {path}")
 
-        # Attempt to read the file and return the data as a list of dictionaries
-        # However, if the file isn't found or there is a generic exception, log
-        # the error and raise an exception
         try:
             with open(path, mode="r", encoding="utf-8") as f:
                 reader = csv.DictReader(f)
@@ -54,7 +43,6 @@ class CSVReader:
         bucket, key = utils.get_s3_path(path)
         self.logger.debug(f"Reading S3 CSV from: {bucket}/{key}")
 
-        # If LOCALSTACK=TRUE, use the localstack endpoint for testing
         if os.getenv("LOCALSTACK", "FALSE").upper() == "TRUE":
             localstack_endpoint = "http://localhost.localstack.cloud:4566"
             self.logger.debug("Using LocalStack endpoint for S3")
@@ -69,15 +57,10 @@ class CSVReader:
             client = boto3.client("s3")
 
         try:
-            # Attempt to read the S3 object and return the data as a list of dictionaries
             response = client.get_object(Bucket=bucket, Key=key)
             self.logger.info("S3 object read successfully")
-            # Read and decode the content
             content = response["Body"].read().decode("utf-8")
-            # Even though the read_string method was only created for testing,
-            # it can be reused here to read and return the CSV data
             return CSVReader.read_string(content)
-        # TODO: Add more specific exceptions to catch
         except Exception as e:
             self.logger.error(f"Error reading S3 object: {e}")
             raise
@@ -87,11 +70,9 @@ class CSVReader:
         A method to read CSV data from a string and return the data as a list
         of dictionaries.
         """
-        # If the content is empty, return an empty list
         if not content.strip():
             return []
 
-        # Treat the string as a file-like object and return as list of dictionaries
         f = io.StringIO(content)
         reader = csv.DictReader(f)
         return [dict(row) for row in reader]
diff --git a/obfuscator/csv_writer.py b/obfuscator/csv_writer.py
index 099e910..56b3f1f 100644
--- a/obfuscator/csv_writer.py
+++ b/obfuscator/csv_writer.py
@@ -3,7 +3,6 @@ import io
 from typing import List, Dict
 from obfuscator.logger import get_logger
 
-# Create the logger
 logger = get_logger("CSVWRITER")
 
 
diff --git a/obfuscator/obfuscate.py b/obfuscator/obfuscate.py
index 4f7e6c1..e964433 100644
--- a/obfuscator/obfuscate.py
+++ b/obfuscator/obfuscate.py
@@ -1,7 +1,6 @@
 from typing import List, Dict
 from obfuscator.logger import get_logger
 
-# Create the logger
 logger = get_logger("OBFUSCATE")
 
 
@@ -12,14 +11,10 @@ def obfuscate(
     A function to obfuscate PII fields in a list of dictionaries, replacing
     sensitive values with a string of asterisks.
     """
-    # If no data is provided, log a message and return an empty list
     if not data:
         logger.info("No valid data was provided to obfuscate")
         return []
 
-    # Obfuscate the PII fields in each record using a list/dict comprehension
-    # This code is good but makes debugging a bit tricky. I may consider
-    # breaking it down into a for loop.
     return [
         {k: ("***" if k in pii_fields else v) for k, v in record.items()}
         for record in data
diff --git a/obfuscator/utils.py b/obfuscator/utils.py
index 81eb04a..f61451b 100644
--- a/obfuscator/utils.py
+++ b/obfuscator/utils.py
@@ -5,7 +5,6 @@ from obfuscator.logger import get_logger
 class Utilities:
 
     def __init__(self, logger=None):
-        # Create the logger
         self.logger = get_logger("UTILITIES", logger)
 
     def get_s3_path(self, uri):
diff --git a/test/test_csv_reader.py b/test/test_csv_reader.py
index e4c135b..0206542 100644
--- a/test/test_csv_reader.py
+++ b/test/test_csv_reader.py
@@ -1,6 +1,3 @@
-# csv_reader.py - tests for read_string and read_s3
-# Author: Alex Schofield
-
 import boto3
 from moto import mock_aws
 from obfuscator.csv_reader import CSVReader
@@ -8,11 +5,6 @@ import pytest
 
 reader = CSVReader()
 
-# CSVREADER: READ_STRING TESTS
-
-# Check if the function can read a CSV string with no content and return
-# an empty list
-
 
 def test_empty_csv_should_return_no_content():
     content = ""
@@ -21,10 +13,6 @@ def test_empty_csv_should_return_no_content():
     assert result == expected
 
 
-# Check if the function can read a CSV string with only a header and return
-# an empty list
-
-
 def test_csv_with_header_only_should_return_no_content():
     content = "student_id,name,course\n"
     result = reader.read_string(content)
@@ -32,10 +20,6 @@ def test_csv_with_header_only_should_return_no_content():
     assert result == expected
 
 
-# Check if the function can read a CSV string with valid data and return
-# a list of dictionaries
-
-
 def test_csv_with_valid_data():
     content = (
         "student_id,name,course\n"
@@ -50,10 +34,6 @@ def test_csv_with_valid_data():
     assert result == expected
 
 
-# Check if the function can read a CSV string with quoted fields and return
-# a list of dictionaries with the quoted fields intact
-
-
 def test_csv_with_quoted_fields_should_run_as_expected():
     content = (
         "student_id,name,course\n"
@@ -68,9 +48,6 @@ def test_csv_with_quoted_fields_should_run_as_expected():
     assert result == expected
 
 
-# CSVREADER: READ_S3 TESTS
-
-
 def setup_s3(s3_client, bucket: str, key: str, content: str):
     s3_client.create_bucket(
         Bucket=bucket,
diff --git a/test/test_obfuscator.py b/test/test_obfuscator.py
index 4f61b16..c77b6b4 100644
--- a/test/test_obfuscator.py
+++ b/test/test_obfuscator.py
@@ -1,8 +1,5 @@
 from obfuscator.obfuscate import obfuscate
 
-# Check if the function does what its supposed to and can obfuscate
-# valid PII fields in a list of dictionaries
-
 
 def test_obfuscate_data_with_valid_pii_fields():
     data = [
@@ -39,11 +36,6 @@ def test_obfuscate_data_with_valid_pii_fields():
     assert result == expected
 
 
-# Check if the function can obfuscate data even when some PII
-# fields are missing from some of the data, returning a list of dictionaries
-# but with the missing PII fields obfuscated and the rest of the data intact
-
-
 def test_obfuscate_data_with_missing_pii_field():
     data = [
         {"student_id": "1234", "name": "John Smith", "course": "Software"},
@@ -69,9 +61,6 @@ def test_obfuscate_data_with_missing_pii_field():
     assert result == expected
 
 
-# Check if the function can handle an empty list of data, returning an empty list
-
-
 def test_obfuscate_data_with_no_data():
     data = []
     pii_fields = ["name", "email_address"]
@@ -81,10 +70,6 @@ def test_obfuscate_data_with_no_data():
     assert result == expected
 
 
-# Check if the function can handle an empty list of PII fields, returning the data as is
-# without mutating it
-
-
 def test_obfuscate_data_with_empty_pii_fields():
     data = [
         {
-- 
cgit v1.2.3


From ad0328b2f292fe438a8a6a1f7ff2d36856dc578d Mon Sep 17 00:00:00 2001
From: Alex Schofield <git@ajschof.me>
Date: Wed, 19 Feb 2025 03:27:54 +0000
Subject: modify logger messages to be more clear

---
 cli.py                   | 10 +++-------
 obfuscator/csv_reader.py |  4 +++-
 obfuscator/csv_writer.py |  4 +---
 obfuscator/obfuscate.py  |  2 +-
 4 files changed, 8 insertions(+), 12 deletions(-)

(limited to 'obfuscator/csv_writer.py')

diff --git a/cli.py b/cli.py
index c03ffd1..40c777f 100644
--- a/cli.py
+++ b/cli.py
@@ -31,23 +31,19 @@ def main():
     args = parser.parse_args()
 
     log_level = "DEBUG" if args.verbose else "INFO"
-
     logger = get_logger("CLI", log_level)
 
     reader = CSVReader(log_level)
 
     if args.local and not args.s3:
-        logger.debug("User chose to read CSV from local path")
+        logger.debug("Read data from local path")
         data = reader.read_local(args.local)
-        logger.debug("Contents: " + str(data))
     else:
-        logger.debug("User chose to read CSV from S3")
-
+        logger.debug("Read data from S3")
         data = reader.read_s3(args.s3)
-        logger.debug("Contents: " + str(data))
 
     obfuscated_data = obfuscate(data, args.pii)
-    logger.debug("Obfuscated data (JSON): " + json.dumps(obfuscated_data, indent=4))
+
     return create_byte_stream(obfuscated_data)
 
 
diff --git a/obfuscator/csv_reader.py b/obfuscator/csv_reader.py
index 2b099c8..3649681 100644
--- a/obfuscator/csv_reader.py
+++ b/obfuscator/csv_reader.py
@@ -45,7 +45,9 @@ class CSVReader:
 
         if os.getenv("LOCALSTACK", "FALSE").upper() == "TRUE":
             localstack_endpoint = "http://localhost.localstack.cloud:4566"
-            self.logger.debug("Using LocalStack endpoint for S3")
+            self.logger.debug(
+                "Using LocalStack endpoint for S3 - ensure LocalStack is running"
+            )
             client = boto3.client(
                 "s3",
                 endpoint_url=localstack_endpoint,
diff --git a/obfuscator/csv_writer.py b/obfuscator/csv_writer.py
index 56b3f1f..de7cd4b 100644
--- a/obfuscator/csv_writer.py
+++ b/obfuscator/csv_writer.py
@@ -8,8 +8,7 @@ logger = get_logger("CSVWRITER")
 
 def create_byte_stream(data: List[Dict[str, str]]) -> bytes:
     if not data:
-        logger.info("No valid data was provided to write")
-        return b""
+        logger.error("Invalid or empty data was provided to write")
 
     output = io.StringIO()
 
@@ -20,6 +19,5 @@ def create_byte_stream(data: List[Dict[str, str]]) -> bytes:
     writer.writerows(data)
 
     csv_string = output.getvalue()
-    logger.debug(f"CSV data: {csv_string}")
 
     return csv_string.encode("utf-8")
diff --git a/obfuscator/obfuscate.py b/obfuscator/obfuscate.py
index e964433..9d43975 100644
--- a/obfuscator/obfuscate.py
+++ b/obfuscator/obfuscate.py
@@ -12,7 +12,7 @@ def obfuscate(
     sensitive values with a string of asterisks.
     """
     if not data:
-        logger.info("No valid data was provided to obfuscate")
+        logger.error("Invalid or empty was provided to obfuscate")
         return []
 
     return [
-- 
cgit v1.2.3


From f1edb55e4f3e2692cb6259cd658c70db6f0cadd4 Mon Sep 17 00:00:00 2001
From: Alex Schofield <git@ajschof.me>
Date: Wed, 19 Feb 2025 03:30:21 +0000
Subject: improve error handling in create_byte_stream()

---
 obfuscator/csv_writer.py | 1 +
 1 file changed, 1 insertion(+)

(limited to 'obfuscator/csv_writer.py')

diff --git a/obfuscator/csv_writer.py b/obfuscator/csv_writer.py
index de7cd4b..2bff6e0 100644
--- a/obfuscator/csv_writer.py
+++ b/obfuscator/csv_writer.py
@@ -9,6 +9,7 @@ logger = get_logger("CSVWRITER")
 def create_byte_stream(data: List[Dict[str, str]]) -> bytes:
     if not data:
         logger.error("Invalid or empty data was provided to write")
+        raise
 
     output = io.StringIO()
 
-- 
cgit v1.2.3


From 422acef7a0762089298e9eae9944877e788fd94d Mon Sep 17 00:00:00 2001
From: Alex Schofield <git@ajschof.me>
Date: Wed, 19 Feb 2025 03:50:09 +0000
Subject: fix some tests (some are still broken)

---
 obfuscator/csv_writer.py | 1 -
 obfuscator/obfuscate.py  | 4 ++--
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'obfuscator/csv_writer.py')

diff --git a/obfuscator/csv_writer.py b/obfuscator/csv_writer.py
index 2bff6e0..de7cd4b 100644
--- a/obfuscator/csv_writer.py
+++ b/obfuscator/csv_writer.py
@@ -9,7 +9,6 @@ logger = get_logger("CSVWRITER")
 def create_byte_stream(data: List[Dict[str, str]]) -> bytes:
     if not data:
         logger.error("Invalid or empty data was provided to write")
-        raise
 
     output = io.StringIO()
 
diff --git a/obfuscator/obfuscate.py b/obfuscator/obfuscate.py
index 73448ce..9cd3a03 100644
--- a/obfuscator/obfuscate.py
+++ b/obfuscator/obfuscate.py
@@ -13,10 +13,10 @@ def obfuscate(
     """
     if not data:
         logger.error("Invalid or empty was provided to obfuscate")
-        raise
+        raise ValueError("Invalid data provided to obfuscate")
     if not pii_fields:
         logger.error("No PII fields provided to obfuscate")
-        raise
+        raise ValueError("No PII fields provided to obfuscate")
 
     return [
         {k: ("***" if k in pii_fields else v) for k, v in record.items()}
-- 
cgit v1.2.3


From 1608d01bb68c1f6292b04c70caa609d34943b371 Mon Sep 17 00:00:00 2001
From: Alex Schofield <git@ajschof.me>
Date: Wed, 19 Feb 2025 15:37:57 +0000
Subject: rename write function & update references

---
 obfuscator/csv_writer.py | 23 -------------------
 obfuscator/write.py      | 23 +++++++++++++++++++
 test/test_csv_writer.py  | 57 ------------------------------------------------
 test/test_write.py       | 57 ++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 80 insertions(+), 80 deletions(-)
 delete mode 100644 obfuscator/csv_writer.py
 create mode 100644 obfuscator/write.py
 delete mode 100644 test/test_csv_writer.py
 create mode 100644 test/test_write.py

(limited to 'obfuscator/csv_writer.py')

diff --git a/obfuscator/csv_writer.py b/obfuscator/csv_writer.py
deleted file mode 100644
index de7cd4b..0000000
--- a/obfuscator/csv_writer.py
+++ /dev/null
@@ -1,23 +0,0 @@
-import csv
-import io
-from typing import List, Dict
-from obfuscator.logger import get_logger
-
-logger = get_logger("CSVWRITER")
-
-
-def create_byte_stream(data: List[Dict[str, str]]) -> bytes:
-    if not data:
-        logger.error("Invalid or empty data was provided to write")
-
-    output = io.StringIO()
-
-    headers = list(data[0].keys())
-
-    writer = csv.DictWriter(output, fieldnames=headers)
-    writer.writeheader()
-    writer.writerows(data)
-
-    csv_string = output.getvalue()
-
-    return csv_string.encode("utf-8")
diff --git a/obfuscator/write.py b/obfuscator/write.py
new file mode 100644
index 0000000..de7cd4b
--- /dev/null
+++ b/obfuscator/write.py
@@ -0,0 +1,23 @@
+import csv
+import io
+from typing import List, Dict
+from obfuscator.logger import get_logger
+
+logger = get_logger("CSVWRITER")
+
+
+def create_byte_stream(data: List[Dict[str, str]]) -> bytes:
+    if not data:
+        logger.error("Invalid or empty data was provided to write")
+
+    output = io.StringIO()
+
+    headers = list(data[0].keys())
+
+    writer = csv.DictWriter(output, fieldnames=headers)
+    writer.writeheader()
+    writer.writerows(data)
+
+    csv_string = output.getvalue()
+
+    return csv_string.encode("utf-8")
diff --git a/test/test_csv_writer.py b/test/test_csv_writer.py
deleted file mode 100644
index eceac28..0000000
--- a/test/test_csv_writer.py
+++ /dev/null
@@ -1,57 +0,0 @@
-import io
-import csv
-from obfuscator.csv_writer import create_byte_stream
-
-
-def csv_bytes_to_list(csv_bytes: bytes):
-    csv_string = csv_bytes.decode("utf-8")
-    f = io.StringIO(csv_string)
-    reader = csv.DictReader(f)
-    return [dict(row) for row in reader]
-
-
-def test_create_byte_stream_valid_data():
-    data = [
-        {"student_id": "1234", "name": "Student 1", "course": "Course 1"},
-        {"student_id": "5678", "name": "Student 2", "course": "Course 2"},
-    ]
-    csv_bytes = create_byte_stream(data)
-    result = csv_bytes_to_list(csv_bytes)
-    assert result == data
-
-
-def test_create_byte_stream_empty_data():
-    csv_bytes = create_byte_stream([])
-    assert csv_bytes == b""
-
-
-def test_create_byte_stream_handles_quoted_fields():
-    data = [
-        {"student_id": "1234", "name": 'Student "One"', "course": "Course, A"},
-        {"student_id": "5678", "name": 'Student "Two"', "course": "Course, B"},
-    ]
-    csv_bytes = create_byte_stream(data)
-    result = csv_bytes_to_list(csv_bytes)
-    assert result == data
-
-
-def test_create_byte_stream_consistent_header_order():
-    data = [
-        {"student_id": "1234", "name": "Alice", "course": "Math"},
-        {"student_id": "5678", "name": "Bob", "course": "Science"},
-    ]
-    csv_bytes = create_byte_stream(data)
-    csv_string = csv_bytes.decode("utf-8")
-    header_line = csv_string.splitlines()[0]
-    expected_header = ",".join(data[0].keys())
-    assert header_line == expected_header
-
-
-def test_create_byte_stream_special_characters():
-    data = [
-        {"student_id": "1234", "name": "Student 1", "course": "Line1\nLine2"},
-        {"student_id": "5678", "name": "Student 2", "course": "Value with, comma"},
-    ]
-    csv_bytes = create_byte_stream(data)
-    result = csv_bytes_to_list(csv_bytes)
-    assert result == data
diff --git a/test/test_write.py b/test/test_write.py
new file mode 100644
index 0000000..eceac28
--- /dev/null
+++ b/test/test_write.py
@@ -0,0 +1,57 @@
+import io
+import csv
+from obfuscator.csv_writer import create_byte_stream
+
+
+def csv_bytes_to_list(csv_bytes: bytes):
+    csv_string = csv_bytes.decode("utf-8")
+    f = io.StringIO(csv_string)
+    reader = csv.DictReader(f)
+    return [dict(row) for row in reader]
+
+
+def test_create_byte_stream_valid_data():
+    data = [
+        {"student_id": "1234", "name": "Student 1", "course": "Course 1"},
+        {"student_id": "5678", "name": "Student 2", "course": "Course 2"},
+    ]
+    csv_bytes = create_byte_stream(data)
+    result = csv_bytes_to_list(csv_bytes)
+    assert result == data
+
+
+def test_create_byte_stream_empty_data():
+    csv_bytes = create_byte_stream([])
+    assert csv_bytes == b""
+
+
+def test_create_byte_stream_handles_quoted_fields():
+    data = [
+        {"student_id": "1234", "name": 'Student "One"', "course": "Course, A"},
+        {"student_id": "5678", "name": 'Student "Two"', "course": "Course, B"},
+    ]
+    csv_bytes = create_byte_stream(data)
+    result = csv_bytes_to_list(csv_bytes)
+    assert result == data
+
+
+def test_create_byte_stream_consistent_header_order():
+    data = [
+        {"student_id": "1234", "name": "Alice", "course": "Math"},
+        {"student_id": "5678", "name": "Bob", "course": "Science"},
+    ]
+    csv_bytes = create_byte_stream(data)
+    csv_string = csv_bytes.decode("utf-8")
+    header_line = csv_string.splitlines()[0]
+    expected_header = ",".join(data[0].keys())
+    assert header_line == expected_header
+
+
+def test_create_byte_stream_special_characters():
+    data = [
+        {"student_id": "1234", "name": "Student 1", "course": "Line1\nLine2"},
+        {"student_id": "5678", "name": "Student 2", "course": "Value with, comma"},
+    ]
+    csv_bytes = create_byte_stream(data)
+    result = csv_bytes_to_list(csv_bytes)
+    assert result == data
-- 
cgit v1.2.3