Release v4.4 (#93)

* Adds play-by-play stats scraping for pages like https://www.basketball-reference.com/boxscores/pbp/201810160BOS.html (#86) * Rewrites internal writing interface (#89) * Updates `README` to reflect new `play_by_play` client method and adds `Contributors` section (#94)
jaebradley · Aug 13, 2019 · b3627f7 · b3627f7
2 parents 947eb40 + f41e796
commit b3627f7
Show file tree

Hide file tree

Showing 24 changed files with 1,062 additions and 301 deletions.
diff --git a/README.md b/README.md
@@ -133,6 +133,26 @@ client.players_advanced_season_totals(season_end_year=2018)
 # The players_advanced_season_totals method also supports all output behavior previously described
 ```
 
+### Get play-by-play data for a game
+
+The structure of the API is due to the unique URL pattern that Basketball Reference has for getting play-by-play data, 
+which depends on the date of the game and the home team.
+
+Example: `https://www.basketball-reference.com/boxscores/pbp/201810160BOS.html`
+
+```python
+from basketball_reference_web_scraper import client
+from basketball_reference_web_scraper.data import Team
+
+# Get play-by-play data for 2018-10-16 game played at the Boston Celtics
+play_by_play = client.play_by_play(
+    home_team=Team.BOSTON_CELTICS,
+    year=2018,
+    month=10,
+    day=16,
+)
+```
+
 ## Development
 
 There are currently two supported major versions - `V3` and `V4`.
@@ -142,3 +162,7 @@ when making changes.
 
 `master` will reflect the latest major version branch.
 
+## Contributors
+
+Thanks to [@DaiJunyan](https://github.com/DaiJunyan) for their contributions!
+
diff --git a/basketball_reference_web_scraper/client.py b/basketball_reference_web_scraper/client.py
@@ -1,14 +1,15 @@
 import requests
 
 from basketball_reference_web_scraper import http_client
-
 from basketball_reference_web_scraper.errors import InvalidSeason, InvalidDate
-from basketball_reference_web_scraper.output import box_scores_to_csv, schedule_to_csv, players_season_totals_to_csv, players_advanced_season_totals_to_csv, team_box_scores_to_csv
 from basketball_reference_web_scraper.output import output
-from basketball_reference_web_scraper.json_encoders import BasketballReferenceJSONEncoder
+from basketball_reference_web_scraper.writers import CSVWriter, RowFormatter, \
+    BOX_SCORE_COLUMN_NAMES, SCHEDULE_COLUMN_NAMES, PLAYER_SEASON_TOTALS_COLUMN_NAMES, \
+    PLAYER_ADVANCED_SEASON_TOTALS_COLUMN_NAMES, TEAM_BOX_SCORES_COLUMN_NAMES, PLAY_BY_PLAY_COLUMN_NAMES
 
 
-def player_box_scores(day, month, year, output_type=None, output_file_path=None, output_write_option=None, json_options=None):
+def player_box_scores(day, month, year, output_type=None, output_file_path=None, output_write_option=None,
+                      json_options=None):
     try:
         values = http_client.player_box_scores(day=day, month=month, year=year)
     except requests.exceptions.HTTPError as http_error:
@@ -21,13 +22,16 @@ def player_box_scores(day, month, year, output_type=None, output_file_path=None,
         output_type=output_type,
         output_file_path=output_file_path,
         output_write_option=output_write_option,
-        csv_writer=box_scores_to_csv,
-        encoder=BasketballReferenceJSONEncoder,
+        csv_writer=CSVWriter(
+            column_names=BOX_SCORE_COLUMN_NAMES,
+            row_formatter=RowFormatter(data_field_names=BOX_SCORE_COLUMN_NAMES)
+        ),
         json_options=json_options,
     )
 
 
-def season_schedule(season_end_year, output_type=None, output_file_path=None, output_write_option=None, json_options=None):
+def season_schedule(season_end_year, output_type=None, output_file_path=None, output_write_option=None,
+                    json_options=None):
     try:
         values = http_client.season_schedule(season_end_year)
     except requests.exceptions.HTTPError as http_error:
@@ -41,13 +45,16 @@ def season_schedule(season_end_year, output_type=None, output_file_path=None, ou
         output_type=output_type,
         output_file_path=output_file_path,
         output_write_option=output_write_option,
-        csv_writer=schedule_to_csv,
-        encoder=BasketballReferenceJSONEncoder,
+        csv_writer=CSVWriter(
+            column_names=SCHEDULE_COLUMN_NAMES,
+            row_formatter=RowFormatter(data_field_names=SCHEDULE_COLUMN_NAMES)
+        ),
         json_options=json_options,
     )
 
 
-def players_season_totals(season_end_year, output_type=None, output_file_path=None, output_write_option=None, json_options=None):
+def players_season_totals(season_end_year, output_type=None, output_file_path=None, output_write_option=None,
+                          json_options=None):
     try:
         values = http_client.players_season_totals(season_end_year)
     except requests.exceptions.HTTPError as http_error:
@@ -60,13 +67,16 @@ def players_season_totals(season_end_year, output_type=None, output_file_path=No
         output_type=output_type,
         output_file_path=output_file_path,
         output_write_option=output_write_option,
-        csv_writer=players_season_totals_to_csv,
-        encoder=BasketballReferenceJSONEncoder,
+        csv_writer=CSVWriter(
+            column_names=PLAYER_SEASON_TOTALS_COLUMN_NAMES,
+            row_formatter=RowFormatter(data_field_names=PLAYER_SEASON_TOTALS_COLUMN_NAMES)
+        ),
         json_options=json_options,
     )
 
 
-def players_advanced_season_totals(season_end_year, output_type=None, output_file_path=None, output_write_option=None, json_options=None):
+def players_advanced_season_totals(season_end_year, output_type=None, output_file_path=None, output_write_option=None,
+                                   json_options=None):
     try:
         values = http_client.players_advanced_season_totals(season_end_year)
     except requests.exceptions.HTTPError as http_error:
@@ -79,13 +89,16 @@ def players_advanced_season_totals(season_end_year, output_type=None, output_fil
         output_type=output_type,
         output_file_path=output_file_path,
         output_write_option=output_write_option,
-        csv_writer=players_advanced_season_totals_to_csv,
-        encoder=BasketballReferenceJSONEncoder,
+        csv_writer=CSVWriter(
+            column_names=PLAYER_ADVANCED_SEASON_TOTALS_COLUMN_NAMES,
+            row_formatter=RowFormatter(data_field_names=PLAYER_ADVANCED_SEASON_TOTALS_COLUMN_NAMES)
+        ),
         json_options=json_options,
     )
 
 
-def team_box_scores(day, month, year, output_type=None, output_file_path=None, output_write_option=None, json_options=None):
+def team_box_scores(day, month, year, output_type=None, output_file_path=None, output_write_option=None,
+                    json_options=None):
     try:
         values = http_client.team_box_scores(day=day, month=month, year=year)
     except requests.exceptions.HTTPError as http_error:
@@ -98,7 +111,31 @@ def team_box_scores(day, month, year, output_type=None, output_file_path=None, o
         output_type=output_type,
         output_file_path=output_file_path,
         output_write_option=output_write_option,
-        csv_writer=team_box_scores_to_csv,
-        encoder=BasketballReferenceJSONEncoder,
+        csv_writer=CSVWriter(
+            column_names=TEAM_BOX_SCORES_COLUMN_NAMES,
+            row_formatter=RowFormatter(data_field_names=TEAM_BOX_SCORES_COLUMN_NAMES)
+        ),
+        json_options=json_options,
+    )
+
+
+def play_by_play(home_team, day, month, year, output_type=None, output_file_path=None, output_write_option=None,
+                 json_options=None):
+    try:
+        values = http_client.play_by_play(home_team=home_team, day=day, month=month, year=year)
+    except requests.exceptions.HTTPError as http_error:
+        if http_error.response.status_code == requests.codes.not_found:
+            raise InvalidDate(day=day, month=month, year=year)
+        else:
+            raise http_error
+    return output(
+        values=values,
+        output_type=output_type,
+        output_file_path=output_file_path,
+        output_write_option=output_write_option,
+        csv_writer=CSVWriter(
+            column_names=PLAY_BY_PLAY_COLUMN_NAMES,
+            row_formatter=RowFormatter(data_field_names=PLAY_BY_PLAY_COLUMN_NAMES)
+        ),
         json_options=json_options,
     )
diff --git a/basketball_reference_web_scraper/data.py b/basketball_reference_web_scraper/data.py
@@ -74,6 +74,11 @@ class Position(Enum):
     GUARD = "GUARD"
 
 
+class PeriodType(Enum):
+    QUARTER = "QUARTER"
+    OVERTIME = "OVERTIME"
+
+
 TEAM_ABBREVIATIONS_TO_TEAM = {
     'ATL': Team.ATLANTA_HAWKS,
     'BOS': Team.BOSTON_CELTICS,
@@ -116,6 +121,9 @@ class Position(Enum):
     'VAN': Team.VANCOUVER_GRIZZLIES,
 }
 
+TEAM_TO_TEAM_ABBREVIATION = {v: k for k, v in TEAM_ABBREVIATIONS_TO_TEAM.items()}
+TEAM_TO_TEAM_ABBREVIATION[Team.CHARLOTTE_HORNETS] = "CHO"
+
 TEAM_NAME_TO_TEAM = {
     "ATLANTA HAWKS": Team.ATLANTA_HAWKS,
     "BOSTON CELTICS": Team.BOSTON_CELTICS,

diff --git a/basketball_reference_web_scraper/http_client.py b/basketball_reference_web_scraper/http_client.py
@@ -1,12 +1,14 @@
 import requests
 
+from basketball_reference_web_scraper.data import TEAM_TO_TEAM_ABBREVIATION
 from basketball_reference_web_scraper.errors import InvalidDate
-from basketball_reference_web_scraper.parsers.box_scores.players import parse_player_box_scores
 from basketball_reference_web_scraper.parsers.box_scores.games import parse_game_url_paths
+from basketball_reference_web_scraper.parsers.play_by_play import parse_play_by_plays
+from basketball_reference_web_scraper.parsers.box_scores.players import parse_player_box_scores
 from basketball_reference_web_scraper.parsers.box_scores.teams import parse_team_totals
-from basketball_reference_web_scraper.parsers.schedule import parse_schedule, parse_schedule_for_month_url_paths
-from basketball_reference_web_scraper.parsers.players_season_totals import parse_players_season_totals
 from basketball_reference_web_scraper.parsers.players_advanced_season_totals import parse_players_advanced_season_totals
+from basketball_reference_web_scraper.parsers.players_season_totals import parse_players_season_totals
+from basketball_reference_web_scraper.parsers.schedule import parse_schedule, parse_schedule_for_month_url_paths
 
 BASE_URL = 'https://www.basketball-reference.com'
 
@@ -108,3 +110,17 @@ def team_box_scores(day, month, year):
         for game_url_path in game_url_paths
         for box_score in team_box_score(game_url_path=game_url_path)
     ]
+
+
+def play_by_play(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=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]
+    )
+    response = requests.get(url=url)
+    response.raise_for_status()
+    return parse_play_by_plays(response.content, home_team)
diff --git a/basketball_reference_web_scraper/json_encoders.py b/basketball_reference_web_scraper/json_encoders.py
@@ -1,6 +1,6 @@
 from datetime import datetime
-from json import JSONEncoder
 from enum import Enum
+from json import JSONEncoder
 
 
 class BasketballReferenceJSONEncoder(JSONEncoder):