jaebradley · jaebradley · Jan 4, 2025 · Jan 5, 2025
diff --git a/basketball_reference_web_scraper/data.py b/basketball_reference_web_scraper/data.py
@@ -54,6 +54,50 @@ class Team(Enum):
     VANCOUVER_GRIZZLIES = "VANCOUVER GRIZZLIES"
     WASHINGTON_BULLETS = "WASHINGTON BULLETS"
 
+class TeamAbbreviation(Enum):
+    ATL = "ATL"
+    BOS = "BOS"
+    BRK = "BRK"
+    CHI = "CHI"
+    CHO = "CHO"
+    CLE = "CLE"
+    DAL = "DAL"
+    DEN = "DEN"
+    DET = "DET"
+    GSW = "GSW"
+    HOU = "HOU"
+    IND = "IND"
+    LAC = "LAC"
+    LAL = "LAL"
+    MEM = "MEM"
+    MIA = "MIA"
+    MIL = "MIL"
+    MIN = "MIN"
+    NOP = "NOP"
+    NYK = "NYK"
+    OKC = "OKC"
+    ORL = "ORL"
+    PHI = "PHI"
+    PHO = "PHO"
+    POR = "POR"
+    SAC = "SAC"
+    SAS = "SAS"
+    TOR = "TOR"
+    UTA = "UTA"
+    WAS = "WAS"
+
+    # DEPRECATED TEAMS
+    KCK = "KCK"
+    NJN = "NJN"
+    NOH = "NOH"
+    NOK = "NOK"
+    CHA = "CHA"
+    CHH = "CHH"
+    SEA = "SEA"
+    STL = "STL"
+    VAN = "VAN"
+    WSB = "WSB"
+
 
 class OutputType(Enum):
     JSON = "JSON"

diff --git a/basketball_reference_web_scraper/http_service.py b/basketball_reference_web_scraper/http_service.py
@@ -1,11 +1,16 @@
+import datetime
+
 import requests
 from lxml import html
 
-from basketball_reference_web_scraper.data import TEAM_TO_TEAM_ABBREVIATION, TeamTotal, PlayerData
+from basketball_reference_web_scraper.data import TeamTotal, PlayerData
 from basketball_reference_web_scraper.errors import InvalidDate, InvalidPlayerAndSeason
 from basketball_reference_web_scraper.html import DailyLeadersPage, PlayerSeasonBoxScoresPage, PlayerSeasonTotalTable, \
     PlayerAdvancedSeasonTotalsTable, PlayByPlayPage, SchedulePage, BoxScoresPage, DailyBoxScoresPage, SearchPage, \
     PlayerPage, StandingsPage
+from basketball_reference_web_scraper.models.calculators import calculate_team_abbreviation
+from basketball_reference_web_scraper.serialization.urls.models import PlayByPlayURLData
+from basketball_reference_web_scraper.serialization.urls.serializers import DEFAULT_PLAY_BY_PLAY_URL_SERIALIZER
 
 
 class HTTPService:
@@ -26,7 +31,7 @@ def standings(self, season_end_year):
 
         page = StandingsPage(html=html.fromstring(response.content))
         return self.parser.parse_division_standings(standings=page.division_standings.eastern_conference_table.rows) + \
-               self.parser.parse_division_standings(standings=page.division_standings.western_conference_table.rows)
+            self.parser.parse_division_standings(standings=page.division_standings.western_conference_table.rows)
 
     def player_box_scores(self, day, month, year):
         url = '{BASE_URL}/friv/dailyleaders.cgi?month={month}&day={day}&year={year}'.format(
@@ -65,7 +70,8 @@ def regular_season_player_box_scores(self, player_identifier, season_end_year, i
         if page.regular_season_box_scores_table is None:
             raise InvalidPlayerAndSeason(player_identifier=player_identifier, season_end_year=season_end_year)
 
-        return self.parser.parse_player_season_box_scores(box_scores=page.regular_season_box_scores_table.rows, include_inactive_games=include_inactive_games)
+        return self.parser.parse_player_season_box_scores(box_scores=page.regular_season_box_scores_table.rows,
+                                                          include_inactive_games=include_inactive_games)
 
     def playoff_player_box_scores(self, player_identifier, season_end_year, include_inactive_games=False):
         # Makes assumption that basketball reference pattern of breaking out player pathing using first character of
@@ -86,15 +92,19 @@ def playoff_player_box_scores(self, player_identifier, season_end_year, include_
         if page.playoff_box_scores_table is None:
             raise InvalidPlayerAndSeason(player_identifier=player_identifier, season_end_year=season_end_year)
 
-        return self.parser.parse_player_season_box_scores(box_scores=page.playoff_box_scores_table.rows, include_inactive_games=include_inactive_games)
+        return self.parser.parse_player_season_box_scores(box_scores=page.playoff_box_scores_table.rows,
+                                                          include_inactive_games=include_inactive_games)
 
     def play_by_play(self, home_team, day, month, year):
-        add_0_if_needed = lambda s: "0" + s if len(s) == 1 else s
-
-        # the hard-coded `0` in the url assumes we always take the first match of the given date and team.
-        url = "{BASE_URL}/boxscores/pbp/{year}{month}{day}0{team_abbr}.html".format(
-            BASE_URL=HTTPService.BASE_URL, year=year, month=add_0_if_needed(str(month)), day=add_0_if_needed(str(day)),
-            team_abbr=TEAM_TO_TEAM_ABBREVIATION[home_team]
+        try:
+            date = datetime.date(year=year, month=month, day=day)
+        except ValueError:
+            raise InvalidDate(day=day, month=month, year=year)
+        url = DEFAULT_PLAY_BY_PLAY_URL_SERIALIZER.serialize(
+            value=PlayByPlayURLData(
+                date=date,
+                team_abbreviation=calculate_team_abbreviation(team=home_team, date=date)
+            )
         )
         response = requests.get(url=url)
         response.raise_for_status()

diff --git a/basketball_reference_web_scraper/models/__init__.py b/basketball_reference_web_scraper/models/__init__.py
diff --git a/basketball_reference_web_scraper/models/calculators.py b/basketball_reference_web_scraper/models/calculators.py
@@ -0,0 +1,15 @@
+import datetime
+
+from basketball_reference_web_scraper.data import TeamAbbreviation, Team
+from basketball_reference_web_scraper.models.teams import ABBREVIATIONS_BY_TEAM
+
+# The Charlotte Hornets' (with Basketball Reference abbreviation CHH) last game was May 15th, 2002
+# https://www.basketball-reference.com/teams/CHH/2002_games.html
+DATE_OF_CHARLOTTE_HORNETS_LAST_GAME = datetime.date(year=2002, month=5, day=15)
+
+
+def calculate_team_abbreviation(team: Team, date: datetime.date) -> TeamAbbreviation:
+    if team is Team.CHARLOTTE_HORNETS and date <= DATE_OF_CHARLOTTE_HORNETS_LAST_GAME:
+        return TeamAbbreviation.CHH
+
+    return ABBREVIATIONS_BY_TEAM.get(team)
diff --git a/basketball_reference_web_scraper/models/teams.py b/basketball_reference_web_scraper/models/teams.py
@@ -0,0 +1,49 @@
+from basketball_reference_web_scraper.data import Team, TeamAbbreviation
+
+TEAMS_BY_ABBREVIATION: dict[TeamAbbreviation, Team] = {
+    TeamAbbreviation.ATL: Team.ATLANTA_HAWKS,
+    TeamAbbreviation.BOS: Team.BOSTON_CELTICS,
+    TeamAbbreviation.BRK: Team.BROOKLYN_NETS,
+    TeamAbbreviation.CHI: Team.CHICAGO_BULLS,
+    TeamAbbreviation.CHO: Team.CHARLOTTE_HORNETS,
+    TeamAbbreviation.CLE: Team.CLEVELAND_CAVALIERS,
+    TeamAbbreviation.DAL: Team.DALLAS_MAVERICKS,
+    TeamAbbreviation.DEN: Team.DENVER_NUGGETS,
+    TeamAbbreviation.DET: Team.DETROIT_PISTONS,
+    TeamAbbreviation.GSW: Team.GOLDEN_STATE_WARRIORS,
+    TeamAbbreviation.HOU: Team.HOUSTON_ROCKETS,
+    TeamAbbreviation.IND: Team.INDIANA_PACERS,
+    TeamAbbreviation.LAC: Team.LOS_ANGELES_CLIPPERS,
+    TeamAbbreviation.LAL: Team.LOS_ANGELES_LAKERS,
+    TeamAbbreviation.MEM: Team.MEMPHIS_GRIZZLIES,
+    TeamAbbreviation.MIA: Team.MIAMI_HEAT,
+    TeamAbbreviation.MIL: Team.MILWAUKEE_BUCKS,
+    TeamAbbreviation.MIN: Team.MINNESOTA_TIMBERWOLVES,
+    TeamAbbreviation.NOP: Team.NEW_ORLEANS_PELICANS,
+    TeamAbbreviation.NYK: Team.NEW_YORK_KNICKS,
+    TeamAbbreviation.OKC: Team.OKLAHOMA_CITY_THUNDER,
+    TeamAbbreviation.ORL: Team.ORLANDO_MAGIC,
+    TeamAbbreviation.PHI: Team.PHILADELPHIA_76ERS,
+    TeamAbbreviation.PHO: Team.PHOENIX_SUNS,
+    TeamAbbreviation.POR: Team.PORTLAND_TRAIL_BLAZERS,
+    TeamAbbreviation.SAC: Team.SACRAMENTO_KINGS,
+    TeamAbbreviation.SAS: Team.SAN_ANTONIO_SPURS,
+    TeamAbbreviation.TOR: Team.TORONTO_RAPTORS,
+    TeamAbbreviation.UTA: Team.UTAH_JAZZ,
+    TeamAbbreviation.WAS: Team.WASHINGTON_WIZARDS,
+
+    # DEPRECATED TEAMS
+    TeamAbbreviation.KCK: Team.KANSAS_CITY_KINGS,
+    TeamAbbreviation.NJN: Team.NEW_JERSEY_NETS,
+    TeamAbbreviation.NOH: Team.NEW_ORLEANS_HORNETS,
+    TeamAbbreviation.NOK: Team.NEW_ORLEANS_OKLAHOMA_CITY_HORNETS,
+    TeamAbbreviation.CHA: Team.CHARLOTTE_BOBCATS,
+    TeamAbbreviation.CHH: Team.CHARLOTTE_HORNETS,
+    TeamAbbreviation.SEA: Team.SEATTLE_SUPERSONICS,
+    TeamAbbreviation.STL: Team.ST_LOUIS_HAWKS,
+    TeamAbbreviation.VAN: Team.VANCOUVER_GRIZZLIES,
+    TeamAbbreviation.WSB: Team.WASHINGTON_BULLETS,
+}
+
+ABBREVIATIONS_BY_TEAM: dict[Team, TeamAbbreviation] = {v: k for k, v in TEAMS_BY_ABBREVIATION.items()}
+ABBREVIATIONS_BY_TEAM[Team.CHARLOTTE_HORNETS] = TeamAbbreviation.CHO
diff --git a/basketball_reference_web_scraper/serialization/__init__.py b/basketball_reference_web_scraper/serialization/__init__.py
diff --git a/basketball_reference_web_scraper/serialization/urls/__init__.py b/basketball_reference_web_scraper/serialization/urls/__init__.py
diff --git a/basketball_reference_web_scraper/serialization/urls/models.py b/basketball_reference_web_scraper/serialization/urls/models.py
@@ -0,0 +1,10 @@
+import dataclasses
+import datetime
+
+from basketball_reference_web_scraper.data import TeamAbbreviation
+
+
+@dataclasses.dataclass(frozen=True)
+class PlayByPlayURLData:
+    date: datetime.date
+    team_abbreviation: TeamAbbreviation
diff --git a/basketball_reference_web_scraper/serialization/urls/serializers.py b/basketball_reference_web_scraper/serialization/urls/serializers.py
@@ -0,0 +1,43 @@
+from datetime import date
+from urllib.parse import urljoin, SplitResult
+
+from basketball_reference_web_scraper.data import TeamAbbreviation
+from basketball_reference_web_scraper.serialization.urls.models import PlayByPlayURLData
+
+
+# TODO: @jaebradley once Python 3.7 support is deprecated, add a Serializer Protocol
+class DateSerializer:
+    def serialize(self, value: date) -> str:
+        return value.strftime("%Y%m%d")
+
+
+class TeamAbbreviationSerializer:
+    def serialize(self, value: TeamAbbreviation) -> str:
+        return value.name
+
+
+class PlayByPlayURLSerializer:
+    def __init__(self, date_serializer, team_abbreviation_serializer):
+        self._date_serializer = date_serializer
+        self._team_abbreviation_serializer = team_abbreviation_serializer
+
+    def serialize(self, value: PlayByPlayURLData) -> str:
+        # From https://docs.python.org/3.9/library/urllib.parse.html#urllib.parse.urlsplit
+        return SplitResult(
+            scheme="https",
+            netloc="www.basketball-reference.com",
+            path=urljoin(
+                "/boxscores/",
+                urljoin(
+                    "pbp/",
+                    f"{self._date_serializer.serialize(value.date)}0{self._team_abbreviation_serializer.serialize(value.team_abbreviation)}.html")),
+            query="",
+            fragment=""
+        ).geturl()
+
+
+# TODO: @jaebradley create a proper singleton metaclass
+DEFAULT_PLAY_BY_PLAY_URL_SERIALIZER = PlayByPlayURLSerializer(
+    date_serializer=DateSerializer(),
+    team_abbreviation_serializer=TeamAbbreviationSerializer()
+)
diff --git a/tests/integration/client/test_play_by_play.py b/tests/integration/client/test_play_by_play.py
@@ -169,11 +169,25 @@ def test_non_unicode_matches(self, m):
         self.assertEqual(len(plays), 509)
 
 
-class TestErrorCases(TestCase):
+class Test20010109CHH(TestCase):
+    def setUp(self):
+        with open(os.path.join(
+                os.path.dirname(__file__),
+                "../files/play_by_play/200101090CHH.html",
+        ), 'r') as file_input: self._html = file_input.read();
 
     @requests_mock.Mocker()
-    def test_get_play_by_play_for_day_that_does_not_exist(self, m):
-        m.get('https://www.basketball-reference.com/boxscores/pbp/201801-10MIL.html', text="Not found", status_code=404)
+    def test_charlotte_hornets(self, m):
+        m.get('https://www.basketball-reference.com/boxscores/pbp/200101090CHH.html', text=self._html, status_code=200)
+        plays = play_by_play(home_team=Team.CHARLOTTE_HORNETS, day=9, month=1, year=2001)
+        self.assertIsNotNone(plays)
+        self.assertEqual(len(plays), 576)
+
+
+class TestErrorCases(TestCase):
+
+    @requests_mock.Mocker(real_http=False)
+    def test_get_play_by_play_for_day_that_does_not_exist(self, _):
         self.assertRaisesRegex(
             InvalidDate,
             "Date with year set to 2018, month set to 1, and day set to -1 is invalid",

diff --git a/tests/integration/files/play_by_play/200101090CHH.html b/tests/integration/files/play_by_play/200101090CHH.html
diff --git a/tests/unit/models/__init__.py b/tests/unit/models/__init__.py
diff --git a/tests/unit/models/test_calculators.py b/tests/unit/models/test_calculators.py
@@ -0,0 +1,18 @@
+import datetime
+import unittest
+
+from basketball_reference_web_scraper.data import TeamAbbreviation, Team
+from basketball_reference_web_scraper.models.calculators import calculate_team_abbreviation
+
+
+class TestCalculatingTeamAbbreviations(unittest.TestCase):
+    def test_charlotte_hornets_on_may_15_2002(self):
+        self.assertEqual(
+            calculate_team_abbreviation(team=Team.CHARLOTTE_HORNETS, date=datetime.date(year=2002, month=5, day=15)),
+            TeamAbbreviation.CHH
+        )
+
+    def test_charlotte_hornets_on_may_16_2002(self):
+        self.assertEqual(
+            calculate_team_abbreviation(team=Team.CHARLOTTE_HORNETS, date=datetime.date(year=2002, month=5, day=16)),
+            TeamAbbreviation.CHO)
diff --git a/tests/unit/serialization/__init__.py b/tests/unit/serialization/__init__.py
diff --git a/tests/unit/serialization/urls/__init__.py b/tests/unit/serialization/urls/__init__.py
diff --git a/tests/unit/serialization/urls/test_serializers.py b/tests/unit/serialization/urls/test_serializers.py
@@ -0,0 +1,41 @@
+import unittest
+from datetime import date
+from unittest import mock
+
+from basketball_reference_web_scraper.data import TeamAbbreviation
+from basketball_reference_web_scraper.serialization.urls.models import PlayByPlayURLData
+from basketball_reference_web_scraper.serialization.urls.serializers import PlayByPlayURLSerializer, \
+    DEFAULT_PLAY_BY_PLAY_URL_SERIALIZER
+
+
+class TestPlayByPlayURLSerializer(unittest.TestCase):
+    def setUp(self):
+        self.serializer = PlayByPlayURLSerializer(
+            date_serializer=mock.Mock(name="mock date serializer",
+                                      serialize=mock.Mock(return_value="some serialized date")),
+            team_abbreviation_serializer=mock.Mock(name="mock team abbreviation serializer", serialize=mock.Mock(
+                return_value="some serialized team abbreviation"))
+        )
+
+        super().setUp()
+
+    def test_output(self):
+        self.assertEqual(
+            "https://www.basketball-reference.com/boxscores/pbp/some serialized date0some serialized team abbreviation.html",
+            self.serializer.serialize(value=PlayByPlayURLData(
+                date=date(year=2001, month=1, day=1),
+                team_abbreviation=TeamAbbreviation.BOS
+            ))
+        )
+
+
+class TestDefaultPlayByPlayURLSerializer(unittest.TestCase):
+    def test_output(self):
+        self.assertEqual(
+            "https://www.basketball-reference.com/boxscores/pbp/200101090CHH.html",
+            DEFAULT_PLAY_BY_PLAY_URL_SERIALIZER.serialize(
+                value=PlayByPlayURLData(
+                    date=date(year=2001, month=1, day=9),
+                    team_abbreviation=TeamAbbreviation.CHH
+                ))
+        )