[ClubElo, FotMob, SoFIFA, FBREF] Fixing and unifying TEAMNAME_REPLACE…

…MENTS loading (#755)

This commit fixes how alternative team names are handled in methods using a `team` parameter. These methods should allow providing both the team name that is used by the data source and the standardized team name in the teamname_replacements.json file.

Example: 
The teamname_replacements.json file sets "Manchester City" as the default name. However, ClubElo uses "Man City". Hence, the `ClubElo.read_team_history` function should work with both `team="Man City"` and `team="Manchester City"`.

---------

Co-authored-by: root <[email protected]>
Co-authored-by: Pieter Robberechts <[email protected]>

CONTRIBUTING.rst

@@ -92,6 +92,7 @@ Install the package with development requirements:
 .. code:: console
 
    $ poetry install
+   $ poetry self add poetry-plugin-export
 
 You can now run an interactive Python session.
 

soccerdata/_common.py

@@ -22,7 +22,7 @@
 from packaging import version
 from selenium.common.exceptions import JavascriptException, WebDriverException
 
-from ._config import DATA_DIR, LEAGUE_DICT, MAXAGE, logger
+from ._config import DATA_DIR, LEAGUE_DICT, MAXAGE, TEAMNAME_REPLACEMENTS, logger
 
 
 class SeasonCode(Enum):
@@ -684,6 +684,61 @@ def make_game_id(row: pd.Series) -> str:
     return game_id
 
 
+def add_alt_team_names(team: Union[str, list[str]]) -> set[str]:
+    """Add a set of alternative team names for a standardized team name.
+
+    If a standardized team name is given, add the set of alternative
+    names used by the data sources. If a non-standardized name is given,
+    a set only containing the given name is returned.
+
+    Parameters
+    ----------
+    team : str or list of str
+        The team name(s) to consider.
+
+    Returns
+    -------
+    set of str
+        A set contraining the given team name(s) and alternative names.
+    """
+    teams = [team] if isinstance(team, str) else team
+
+    alt_teams = set()
+    for team in teams:
+        for alt_name, norm_name in TEAMNAME_REPLACEMENTS.items():
+            if norm_name == team:
+                alt_teams.add(alt_name)
+        alt_teams.add(team)
+    return alt_teams
+
+
+def add_standardized_team_name(team: Union[str, list[str]]) -> set[str]:
+    """Add the standardized team name for a non-standardized team name.
+
+    If a non-standardized team name is given, add the standardized
+    name. If a standardized name is given, a set only containing the given
+    name is returned.
+
+    Parameters
+    ----------
+    team : str or list of str
+        The team name(s) to consider.
+
+    Returns
+    -------
+    set of str
+        A set contraining the given team name(s) and standardized names.
+    """
+    teams = [team] if isinstance(team, str) else team
+    std_teams = set()
+    for team in teams:
+        for alt_name, norm_name in TEAMNAME_REPLACEMENTS.items():
+            if alt_name == team:
+                std_teams.add(norm_name)
+        std_teams.add(team)
+    return std_teams
+
+
 def standardize_colnames(df: pd.DataFrame, cols: Optional[list[str]] = None) -> pd.DataFrame:
     """Convert DataFrame column names to snake case."""
 

soccerdata/clubelo.py

@@ -8,7 +8,7 @@
 import pandas as pd
 from unidecode import unidecode
 
-from ._common import BaseRequestsReader, standardize_colnames
+from ._common import BaseRequestsReader, add_alt_team_names, standardize_colnames
 from ._config import DATA_DIR, NOCACHE, NOSTORE, TEAMNAME_REPLACEMENTS
 
 CLUB_ELO_DATADIR = DATA_DIR / "ClubElo"
@@ -148,12 +148,8 @@ def read_team_history(
         -------
         pd.DataFrame
         """
-        teams_to_check = [k for k, v in TEAMNAME_REPLACEMENTS.items() if v == team]
-        teams_to_check.append(team)
-
-        for i, _ in enumerate(teams_to_check):
-            teams_to_check[i] = unidecode(teams_to_check[i])
-            teams_to_check[i] = re.sub(r"[\s']", "", teams_to_check[i])
+        teams_to_check = add_alt_team_names(team)
+        teams_to_check = {re.sub(r"[\s']", "", unidecode(team)) for team in teams_to_check}
 
         for _team in teams_to_check:
             filepath = self.data_dir / f"{_team}.csv"

soccerdata/fbref.py

@@ -9,7 +9,13 @@
 import pandas as pd
 from lxml import etree, html
 
-from ._common import BaseRequestsReader, SeasonCode, make_game_id, standardize_colnames
+from ._common import (
+    BaseRequestsReader,
+    SeasonCode,
+    add_alt_team_names,
+    make_game_id,
+    standardize_colnames,
+)
 from ._config import DATA_DIR, NOCACHE, NOSTORE, TEAMNAME_REPLACEMENTS, logger
 
 FBREF_DATADIR = DATA_DIR / "FBref"
@@ -411,14 +417,7 @@ def read_team_match_stats(  # noqa: C901
         df_teams = self.read_team_season_stats()
 
         if team is not None:
-            # get alternative names of the specified team(s)
-            teams = [team] if isinstance(team, str) else team
-            teams_to_check = []
-            for team in teams:
-                for alt_name, norm_name in TEAMNAME_REPLACEMENTS.items():
-                    if norm_name == team:
-                        teams_to_check.append(alt_name)
-            teams_to_check.append(team)
+            teams_to_check = add_alt_team_names(team)
 
             # select requested teams
             iterator = df_teams.loc[df_teams.index.isin(teams_to_check, level=2), :]

soccerdata/fotmob.py

@@ -9,7 +9,7 @@
 import pandas as pd
 import requests
 
-from ._common import BaseRequestsReader, make_game_id
+from ._common import BaseRequestsReader, add_standardized_team_name, make_game_id
 from ._config import DATA_DIR, NOCACHE, NOSTORE, TEAMNAME_REPLACEMENTS, logger
 
 FOTMOB_DATADIR = DATA_DIR / "FotMob"
@@ -327,7 +327,7 @@ def read_schedule(self, force_cache: bool = False) -> pd.DataFrame:
         df[["home_score", "away_score"]] = df["status.scoreStr"].str.split("-", expand=True)
         return df.set_index(["league", "season", "game"]).sort_index()[cols]
 
-    def read_team_match_stats(  # noqa: C901
+    def read_team_match_stats(
         self,
         stat_type: str = "Top stats",
         opponent_stats: bool = True,
@@ -378,13 +378,8 @@ def read_team_match_stats(  # noqa: C901
 
         if team is not None:
             # get alternative names of the specified team(s)
-            teams = [team] if isinstance(team, str) else team
-            teams_to_check = []
-            for team in teams:
-                for alt_name, norm_name in TEAMNAME_REPLACEMENTS.items():
-                    if norm_name == team:
-                        teams_to_check.append(alt_name)
-                teams_to_check.append(team)
+            teams_to_check = add_standardized_team_name(team)
+
             # select requested teams
             iterator = df_complete.loc[
                 (

soccerdata/sofifa.py

@@ -10,7 +10,7 @@
 import pandas as pd
 from lxml import html
 
-from ._common import BaseRequestsReader, standardize_colnames
+from ._common import BaseRequestsReader, add_standardized_team_name, standardize_colnames
 from ._config import DATA_DIR, NOCACHE, NOSTORE, TEAMNAME_REPLACEMENTS, logger
 
 SO_FIFA_DATADIR = DATA_DIR / "SoFIFA"
@@ -240,14 +240,7 @@ def read_players(self, team: Optional[Union[str, list[str]]] = None) -> pd.DataF
         df_teams = self.read_teams()
 
         if team is not None:
-            # get alternative names of the specified team(s)
-            teams = [team] if isinstance(team, str) else team
-            teams_to_check = []
-            for team in teams:
-                for alt_name, norm_name in TEAMNAME_REPLACEMENTS.items():
-                    if norm_name == team:
-                        teams_to_check.append(alt_name)
-            teams_to_check.append(team)
+            teams_to_check = add_standardized_team_name(team)
 
             # select requested teams
             iterator = df_teams.loc[df_teams.team.isin(teams_to_check), :]

tests/appdata/config/teamname_replacements.json

@@ -1,3 +1,6 @@
 {
-  "Manchester City": ["Man City"]
+  "Manchester City": ["Man City"],
+  "Olympique Marseille": ["Marseille"],
+  "Valencia CF": ["Valencia"],
+  "FC Bayern Munich": ["FC Bayern München"]
 }

tests/test_FBref.py

@@ -59,6 +59,19 @@ def test_read_team_match_stats(fbref_ligue1: FBref, stat_type: str) -> None:
     assert isinstance(fbref_ligue1.read_team_match_stats(stat_type), pd.DataFrame)
 
 
+def test_read_team_match_stats_alt_names(fbref_ligue1: FBref) -> None:
+    # Test with FBref team name
+    assert isinstance(
+        fbref_ligue1.read_team_match_stats(stat_type="schedule", team="Olympique Marseille"),
+        pd.DataFrame,
+    )
+    # Test with standardized team name
+    assert isinstance(
+        fbref_ligue1.read_team_match_stats(stat_type="schedule", team="Marseille"),
+        pd.DataFrame,
+    )
+
+
 @pytest.mark.parametrize(
     "stat_type",
     [

tests/test_FotMob.py

@@ -28,3 +28,16 @@ def test_read_team_match_stats(fotmob_laliga: FotMob, stat_type: str) -> None:
     assert isinstance(
         fotmob_laliga.read_team_match_stats(stat_type, team="Valencia"), pd.DataFrame
     )
+
+
+@pytest.mark.fails_gha()
+def test_read_team_match_stats_alt_names(fotmob_laliga: FotMob) -> None:
+    # Test with Fotmob team name
+    assert isinstance(
+        fotmob_laliga.read_team_match_stats(stat_type="Top stats", team="Valencia"), pd.DataFrame
+    )
+    # Test with standardized team name
+    assert isinstance(
+        fotmob_laliga.read_team_match_stats(stat_type="Top stats", team="Valencia CF"),
+        pd.DataFrame,
+    )

tests/test_SoFIFA.py

@@ -5,6 +5,16 @@
 from soccerdata.sofifa import SoFIFA
 
 
+def test_read_players(sofifa_bundesliga: SoFIFA) -> None:
+    """It should use the replacement names from teamname_replacements.json."""
+    assert isinstance(sofifa_bundesliga.read_players(team="FC Bayern München"), pd.DataFrame)
+
+
+def test_read_players_replacement(sofifa_bundesliga: SoFIFA) -> None:
+    """It should use the replacement names from teamname_replacements.json."""
+    assert isinstance(sofifa_bundesliga.read_players(team="FC Bayern Munich"), pd.DataFrame)
+
+
 def test_read_team_ratings(sofifa_bundesliga: SoFIFA) -> None:
     """It should return a dataframe with the team ratings."""
     assert isinstance(sofifa_bundesliga.read_team_ratings(), pd.DataFrame)

tests/test_common.py

@@ -11,6 +11,8 @@
 from soccerdata._common import (
     BaseRequestsReader,
     SeasonCode,
+    add_alt_team_names,
+    add_standardized_team_name,
     make_game_id,
     standardize_colnames,
 )
@@ -108,6 +110,23 @@ def test_make_game_id():
     assert game_id == "1993-07-30 Barcelona-Real Madrid"
 
 
+# add_alt_team_names
+
+
+def test_add_alt_team_names():
+    # "Valencia" is replaced by "Valencia CF"
+    assert add_alt_team_names("Valencia CF") == {"Valencia", "Valencia CF"}
+    # "Real Madrid" is not replaced
+    assert add_alt_team_names("Real Madrid") == {"Real Madrid"}
+
+
+def test_add_standardize_team_name():
+    # "Valencia" is replaced by "Valencia CF"
+    assert add_standardized_team_name("Valencia") == {"Valencia", "Valencia CF"}
+    # "Real Madrid" is not replaced
+    assert add_standardized_team_name("Real Madrid") == {"Real Madrid"}
+
+
 # standardize_colnames