acb-database-scraping

TODO:

• Change MySQL database credentials to external file and load them as environment variables.

The aim of this project is to collect information about all the games of the ACB Spanish Basketball League from 1994 to 2016. A SQLite database has been used for this purpose.

Why from 1994? In 1994, the ACB changed to the current modern league format, which consists of a regular season and a playoff competition between the best 8.

Requirements

• Python 3
• pyquery
• peewee

Instructions on how to execute

The database can be freely accessed from https://data.world/jgonzalezferrer/acb-1994-2016-spanish-basketball-league-results or https://www.kaggle.com/jgonzalezferrer/acb-spanish-basketball-league-results. However, if you want to execute the code by yourself you can just use the run.py script:

$ python run.py [-r] [-d] [-i] [--start] [first_year] [--end] [last_year]

where:

• -r indicates whether you want to reset the database.
• -dif you want to download locally the games.
• -iif you want to inser the information in the database.
• --start first_year from which season you want to scrap (1994 by default).
• --end last_year until which season you want to scrap (2016 by default).

Therefore, the first time you run the script, you must use run.py -r -d -i.

Content

This dataset includes statistics about the games, teams, players and coaches. It is divided in the following tables:

• Game: basic information about the game such as the venue, the attendance, the kickoff, the involved teams and the final score.
• Participant: a participant is a player, coach or referee that participates in a game. A participant is associated to a game, an actor and a team. Each row contains information about different stats such as number of points, assists or rebounds.
• Actor: an actor represents a player or a coach. It contains personal information about them, such as the height, position or birthday. With this table we can track the different teams that a player has been into.
• Team: this class represents a team.
• TeamName: the name of a team can change between seasons (and even within the same season).

In summation, this database contains the stats from games such as http://www.acb.com/fichas/LACB61295.php

Por hacer:

¿Final 2017 no aparece en tabla games? ¿tratar perdidas de conexion a internet en descargas? ¿Metemos los equipos en la tabla participantes?¿tiene info relevante? ¿Que hacemos con los participantes? def fix_participants(): Participant._fix_acbid('Esteban, Màxim', '2CH') Participant._fix_acbid('Sharabidze, G.', 'Y9G') Participant._fix_participations('Tavares, W.', 'T2Z', 'SHP') Participant._fix_participations('Stobart, Micky', 'B7P', 'FII') Participant._fix_participations('Olaizola, Julen', 'T86', '162') Participant._fix_participations('Izquierdo, Antonio', '773', 'YHK') Arreglar nombres de games --> game_acbid ¿hardcodear o pillar de wikipedia algunos equipos que no tienen año de fundacion? tablas y relaciones porque en game el acbid es un varchar? renombrar a game_acbid? renombrar resto de a game_acbid para no confundir con el id de la tabla que suele ser nombreTabla_id relacionar tabla eventos con game -- por game_id? revisar todos lo atributos de las tablas para que tengan sentido. porque CREATE INDEX game_kickoff_time_idx ON game(kickoff_time)? en tabla event relacionar teamid con algo? en vez del varchar el id numerico del equipo? renombrar todos los atrinutoid a atributo_id