special-days

Lookup dates for special events, so far the Super Bowl Sunday and Oscars night. Compatible with the holidays package, so you can use the same date in calendar logic that's available for public holidays for "is today Super Bowl Sunday?" too.

Data is checked daily against Wikidata and, once manually reviewed and published, pip install --upgrade will pull the refreshed known dates.

Status: Alpha. Currently included Super Bowl + Academy Awards (Oscars). The API probably won't change, but it might until the maintenance process has run for a while and I've learned if things need to change.

Install

pip install special-days

Quickstart

Answer "is this day special?" in a holidays-compatible way.

from datetime import date
from special_days import SpecialDays

sd = SpecialDays()                       # all events
sd.get_list(date(2025, 2, 9))            # ['Super Bowl']
sd.get_list(date(2025, 3, 2))            # ['Academy Awards']
sd.get_list(date(2025, 5, 1))            # []

You can combine with the holidays package via union(...):

import holidays
from special_days import SpecialDays, union

combined = union(holidays.US(), SpecialDays())
combined.get_list(date(2025, 2, 9))      # ['Super Bowl']
combined.get_list(date(2025, 7, 4))      # ['Independence Day']

You can exclude some special days in most cases with something like this:

EMOJI = {
    "Independence Day": "🎆",
    "Super Bowl":       "🏈",
}

def specials(d):
    return [(n, EMOJI[n]) for n in combined.get_list(d) if n in EMOJI]

specials(date(2025, 2, 9))               # [('Super Bowl', '🏈')]
specials(date(2025, 7, 4))               # [('Independence Day', '🎆')]
specials(date(2025, 12, 25))             # []

A more full version is in examples/by_date.py.

Two ways to use it

Date-keyed (just like holidays)

A dict-like subclass populated from the included data snapshot.

from datetime import date
from special_days import SuperBowl

sb = SuperBowl()                         # every known Super Bowl date
date(2025, 2, 9) in sb                   # True
sb[date(2025, 2, 9)]                     # 'Super Bowl'
sb.get_list(date(2025, 2, 9))            # ['Super Bowl']
len(sb)                                  # one entry per game, grows each year

# Filter to a subset of years.
SuperBowl(years=[1967, 2024, 2025])
SuperBowl(years=range(2020, 2030))

datetime.datetime values are normalized to date automatically, so you can mix the two in a lookup.

Year-keyed (planner-style)

from datetime import date
from special_days import super_bowl

super_bowl.date(2025)                    # datetime.date(2025, 2, 9)
super_bowl.is_super_bowl_sunday(date(2025, 2, 9))   # True
super_bowl.all_known()                   # {1967: date(1967, 1, 15), ..., 2027: date(2027, 2, 14)}

Display strings

By default the dict value is the constant event name ("Super Bowl", "Academy Awards"). Pass label_with_edition=True for edition-numbered display strings:

SuperBowl(label_with_edition=True)[date(2025, 2, 9)]     # 'Super Bowl LIX'
Oscars(label_with_edition=True)[date(2025, 3, 2)]        # '97th Academy Awards'

Super Bowl 50 (2016) uses the Arabic numeral, matching official NFL branding; every other edition uses Roman. The Academy Awards labeller handles 1930's two ceremonies (the 2nd in April, the 3rd in November) and the subsequent resync at the 6th ceremony in 1934.

Two ceremonies in one year

Most series have at most one installment per year, but not all. 1930 hosted two Academy Awards (the 2nd and 3rd):

from special_days import oscars

oscars.date(1930)                        # datetime.date(1930, 4, 3)  -- first only
oscars.dates(1930)                       # [date(1930, 4, 3), date(1930, 11, 5)]

SpecialDays/Oscars dicts contain both dates and label them distinctly when label_with_edition=True.

Recipes

Next upcoming event from today

is_super_bowl_sunday over a date range, or just iterate dates() / all_known():

from datetime import date, timedelta
from special_days import super_bowl

today = date.today()
upcoming = next(
    d for d in (today + timedelta(days=i) for i in range(365 * 2))
    if super_bowl.is_super_bowl_sunday(d)
)

All special days in a date range

from datetime import date, timedelta
from special_days import SpecialDays

sd = SpecialDays()
start, end = date(2025, 1, 1), date(2025, 12, 31)
days = [d for d in (start + timedelta(days=i)
                    for i in range((end - start).days + 1))
        if d in sd]
# [date(2025, 2, 9), date(2025, 3, 2)]

Just one event

SpecialDays(events=[...]) accepts string names, classes, or already-built instances:

from special_days import SpecialDays, SuperBowl

SpecialDays(events=["super_bowl"])      # by registered name
SpecialDays(events=[SuperBowl])         # by class
SpecialDays(events=[SuperBowl(years=[2024, 2025])])   # pre-filtered instance

Data freshness

The shipped snapshot covers every event in Wikidata at release time — every past game and every officially-announced future game. A daily job rebuilds the snapshot from Wikidata; if anything changed, it opens a PR. Once manually reviewed and merged, a new patch release will be created on PyPI. Run pip install --upgrade special-days to pull the latest.

For maintainers and curious users, the full pipeline is documented in docs/how_it_works.md.

Errors

• super_bowl.date(year) raises KeyError if the year is not in the shipped snapshot. Upgrade the package to pick up newly-announced dates.
• Non-int year arguments raise TypeError immediately.
• SpecialDays(events=[...]) raises ValueError for unknown event strings, listing the valid ones.

License

MIT