Avoid Fixed Offset Timezones: Use zoneinfo for Accurate DST Handling

[ShreyTiwari]

Hi there! 👋

I noticed that the codebase uses fixed offset timezones when creating timezone-aware datetime objects. This can lead to issues with daylight saving time (DST) changes, causing inconsistencies and potential software failures. Can we consider a more dynamic approach to handling time zones?

CodeQL Alerts:

Here are the specific instances CodeQL flagged:

1. Location:

   CARD/regression/main.py

   Line 409 in ebe2f2a

   timestr = datetime.now(timezone(timedelta(hours=-6))).strftime("%Y%m%d-%H%M%S-%f") # US Central time

Explanation

Issue:

• Fixed offset timezones, such as timezone(timedelta(hours=-6)), do not account for daylight saving time changes. This can lead to inaccurate times during DST transitions, which may cause software errors or inconsistencies.

Potential Solution:

• Use the zoneinfo module to handle timezone-aware datetime objects, which dynamically adjust for DST changes. This ensures your datetimes reflect accurate local times throughout the year.

Example Solution:

from datetime import datetime
from zoneinfo import ZoneInfo

# Using zoneinfo for US Central Time
timestr = datetime.now(ZoneInfo('America/Chicago')).strftime("%Y%m%d-%H%M%S-%f")

# Correctly accounts for DST transitions
print(f"Dynamic CST with zoneinfo: {timestr}")

Action Required:

• Replace fixed offset timezone usage with zoneinfo.ZoneInfo to ensure accurate handling of time zones, including daylight saving time adjustments.

References:

• Python zoneinfo Documentation

Thank you for your efforts in maintaining this project! 🌟

Best,
Shrey

[XzwHan]

Hi @ShreyTiwari, thank you very much for your detailed explanation and proposed solution! We are currently working on updates to improve the codebase’s practicality and robustness, and we’ll take your suggestions into consideration as part of this process.

[ShreyTiwari]

You're welcome!