How to preserve timezone when parsing date/time strings with strptime?

strptime

timezone

datetime

date parsing

time conversion

How to preserve timezone when parsing date/time strings with strptime?

Master System Design with Codemia

Enhance your system design skills with over 120 practice problems, detailed solutions, and hands-on exercises.

Start Practicing Learn More

Introduction

Python's datetime.strptime() can parse timezone offsets using the %z directive (e.g., +05:30, -0800, UTC), but it does not support named timezone abbreviations like EST or PST because these are ambiguous. For reliable timezone-aware parsing, use %z for numeric offsets, dateutil.parser.parse() for flexible formats, or datetime.fromisoformat() (Python 3.7+) for ISO 8601 strings. The key is to always produce timezone-aware datetime objects so timezone information is not silently lost.

The Problem: strptime and Timezone Loss

python

1from datetime import datetime
2
3# Parsing WITHOUT timezone — produces a naive datetime
4dt = datetime.strptime("2025-01-15 10:30:00", "%Y-%m-%d %H:%M:%S")
5print(dt.tzinfo)  # None — no timezone information
6
7# The timezone is lost — you cannot tell if this is UTC, EST, or JST

A naive datetime (one with tzinfo=None) has no timezone information. Any timezone context from the original string is lost if you do not parse it.

Using %z for Numeric Offsets

python

1from datetime import datetime
2
3# %z parses +HHMM or +HH:MM offset formats
4dt = datetime.strptime("2025-01-15T10:30:00+0530", "%Y-%m-%dT%H:%M:%S%z")
5print(dt)          # 2025-01-15 10:30:00+05:30
6print(dt.tzinfo)   # UTC+05:30
7
8# Negative offset
9dt2 = datetime.strptime("2025-01-15T10:30:00-0800", "%Y-%m-%dT%H:%M:%S%z")
10print(dt2)         # 2025-01-15 10:30:00-08:00
11
12# UTC with Z suffix (Python 3.7+)
13dt3 = datetime.strptime("2025-01-15T10:30:00Z", "%Y-%m-%dT%H:%M:%SZ")
14print(dt3.tzinfo)  # None — %Z and literal Z do NOT set timezone
15
16# Correct way to parse Z as UTC
17dt3 = datetime.strptime("2025-01-15T10:30:00+0000", "%Y-%m-%dT%H:%M:%S%z")
18print(dt3.tzinfo)  # UTC

The %z directive (lowercase) parses numeric UTC offsets. Note that a literal Z at the end of a format string is matched as a character, not as a timezone.

Using datetime.fromisoformat() (Python 3.7+)

python

1from datetime import datetime
2
3# fromisoformat handles ISO 8601 strings including timezone
4dt = datetime.fromisoformat("2025-01-15T10:30:00+05:30")
5print(dt)          # 2025-01-15 10:30:00+05:30
6print(dt.tzinfo)   # UTC+05:30
7
8# Python 3.11+ also handles Z suffix
9dt2 = datetime.fromisoformat("2025-01-15T10:30:00Z")
10print(dt2)         # 2025-01-15 10:30:00+00:00
11
12# Without timezone — returns naive datetime
13dt3 = datetime.fromisoformat("2025-01-15T10:30:00")
14print(dt3.tzinfo)  # None

fromisoformat() is simpler than strptime() for ISO 8601 strings and handles timezone offsets correctly.

Using dateutil.parser.parse()

python

1from dateutil import parser
2
3# dateutil handles many formats automatically
4dt1 = parser.parse("January 15, 2025 10:30 AM EST")
5print(dt1)  # 2025-01-15 10:30:00-05:00
6
7dt2 = parser.parse("2025-01-15T10:30:00+05:30")
8print(dt2)  # 2025-01-15 10:30:00+05:30
9
10dt3 = parser.parse("Wed, 15 Jan 2025 10:30:00 GMT")
11print(dt3)  # 2025-01-15 10:30:00+00:00
12
13# Specify timezone when the string has none
14from dateutil.tz import gettz
15dt4 = parser.parse("2025-01-15 10:30:00", tzinfos={"EST": gettz("US/Eastern")})
16print(dt4)  # 2025-01-15 10:30:00

Install with pip install python-dateutil. The dateutil parser is the most flexible option, handling named timezones and varied date formats.

Using zoneinfo (Python 3.9+)

python

1from datetime import datetime
2from zoneinfo import ZoneInfo
3
4# Attach a timezone to a naive datetime
5naive = datetime.strptime("2025-01-15 10:30:00", "%Y-%m-%d %H:%M:%S")
6aware = naive.replace(tzinfo=ZoneInfo("America/New_York"))
7print(aware)  # 2025-01-15 10:30:00-05:00
8
9# Convert between timezones
10utc_time = aware.astimezone(ZoneInfo("UTC"))
11print(utc_time)  # 2025-01-15 15:30:00+00:00
12
13tokyo_time = aware.astimezone(ZoneInfo("Asia/Tokyo"))
14print(tokyo_time)  # 2025-01-16 00:30:00+09:00

zoneinfo is the standard library module for IANA timezone names. Use it to attach or convert timezones on parsed datetimes.

Using pytz (Pre-3.9)

python

1import pytz
2from datetime import datetime
3
4# Localize a naive datetime
5eastern = pytz.timezone("US/Eastern")
6naive = datetime.strptime("2025-01-15 10:30:00", "%Y-%m-%d %H:%M:%S")
7aware = eastern.localize(naive)
8print(aware)  # 2025-01-15 10:30:00-05:00
9
10# Convert to UTC
11utc_time = aware.astimezone(pytz.utc)
12print(utc_time)  # 2025-01-15 15:30:00+00:00

For Python < 3.9, use pytz.timezone().localize() instead of replace(tzinfo=...).

Parsing Custom Timezone Abbreviations

python

1from datetime import datetime, timezone, timedelta
2
3# Map abbreviations to offsets manually
4TZ_MAP = {
5    "EST": timezone(timedelta(hours=-5)),
6    "CST": timezone(timedelta(hours=-6)),
7    "MST": timezone(timedelta(hours=-7)),
8    "PST": timezone(timedelta(hours=-8)),
9    "IST": timezone(timedelta(hours=5, minutes=30)),
10}
11
12def parse_with_tz_abbrev(date_str, fmt, tz_abbrev):
13    naive = datetime.strptime(date_str, fmt)
14    return naive.replace(tzinfo=TZ_MAP[tz_abbrev])
15
16dt = parse_with_tz_abbrev("2025-01-15 10:30:00", "%Y-%m-%d %H:%M:%S", "PST")
17print(dt)  # 2025-01-15 10:30:00-08:00

Timezone abbreviations are ambiguous — CST can mean Central Standard Time (US), China Standard Time, or Cuba Standard Time. Map them explicitly when you know the context.

Common Pitfalls

Assuming %Z preserves timezone info: The %Z directive in strptime can match timezone names like UTC or EST in some implementations, but it does not reliably attach timezone info to the result. The parsed datetime may still be naive. Use %z (lowercase) for numeric offsets instead.
Using replace(tzinfo=...) with pytz: datetime.replace(tzinfo=pytz.timezone('US/Eastern')) does not handle DST transitions correctly. Use pytz.timezone('US/Eastern').localize(naive_dt) instead. This issue does not apply to zoneinfo (Python 3.9+), where replace works correctly.
Parsing Z as a timezone: A literal Z at the end of an ISO 8601 string means UTC, but strptime treats it as a literal character, not a timezone. Replace Z with +0000 before parsing, or use fromisoformat() (Python 3.11+).
Mixing naive and aware datetimes: Comparing or subtracting a naive datetime with an aware datetime raises TypeError. Ensure all datetimes in your application are consistently aware or consistently naive.
Trusting timezone abbreviations: CST can mean US Central, China Standard, or Cuba Standard. IST can mean India, Israel, or Ireland. Always use IANA timezone names (e.g., America/Chicago) for unambiguous timezone handling.

Summary

Use %z in strptime() to parse numeric UTC offsets like +0530 or -0800
Use datetime.fromisoformat() (Python 3.7+) for ISO 8601 strings with timezone info
Use dateutil.parser.parse() for flexible parsing of varied date formats with named timezones
Use zoneinfo.ZoneInfo (Python 3.9+) or pytz to attach IANA timezones to naive datetimes
Timezone abbreviations are ambiguous — map them explicitly or use full IANA timezone names
Always produce timezone-aware datetimes to prevent silent timezone loss