Open Sport(s) Data (Football, Alpine Ski, Formula 1, etc.) Tools & Scripts
Command Line Tools • Scripting • Services (HTTP JSON APIs)
See the football.db League Starter Sample - Mauritius Premier League if you want to start from scratch (zero) with your very own league.
See the football.db Quick Starter Datafile Templates if you want to read in ready-to-use / ready-to-fork dataset packages incl. the English Premier League, the German Bundesliga, the Spanish Primera División and some more.
The sportdb tool lets you read in (parse) datasets (e.g. leagues, clubs, match schedules, etc.) in plain text into your sports SQL database of choice (e.g. SQLite, PostgreSQL, etc.)
SYNOPSIS
sportdb [global options] command [command options] [arguments...]
VERSION
2.0
GLOBAL OPTIONS
-d, --dbpath=PATH - Database path (default: .)
-n, --dbname=NAME - Database name (default: sport.db)
--verbose - (Debug) Show debug messages
--version - Show version
COMMANDS
new, n - Build DB w/ quick starter Datafile templates
build, b - Build DB (download/create/read); use ./Datafile - zips get downloaded to ./tmp
serve, server - Start web service (HTTP JSON API)
MORE COMMANDS
create - Create DB schema
download, dl - Download datasets; use ./Datafile - zips get downloaded to ./tmp
read, r - Read datasets; use ./Datafile - zips required in ./tmp
logs - Show logs
props - Show props
stats - Show stats
test - (Debug) Test command suite
help - Shows a list of commands or help for one command
NAME
new - Build DB w/ quick starter Datafile templates
SYNOPSIS
sportdb [global options] new NAME
EXAMPLES
sportdb new eng2019-20
sportdb new eng
NAME
build - Build DB (download/create/read); use ./Datafile - zips get downloaded to ./tmp
SYNOPSIS
sportdb [global options] build
EXAMPLES
sportdb build
NAME
serve - Start web service (HTTP JSON API)
SYNOPSIS
sportdb [global options] serve
EXAMPLES
sportdb serve
Step 1
Setup the (SQL) database. Let's use and build a single-file SQLite database (from scratch), as an example:
require 'sportdb/readers'
SportDb.connect( adapter: 'sqlite3',
database: './england.db' )
SportDb.create_all ## build database schema (tables, indexes, etc.)Step 2
Let's read in some leagues, seasons, clubs, and match schedules and results.
Let's use the public domain football.db datasets for England (see openfootball/england), as an example:
= English Premier League 2015/16
Matchday 1
[Sat Aug 8]
Manchester United 1-0 Tottenham Hotspur
AFC Bournemouth 0-1 Aston Villa
Everton FC 2-2 Watford FC
Leicester City 4-2 Sunderland AFC
Norwich City 1-3 Crystal Palace
Chelsea FC 2-2 Swansea City
[Sun Aug 9]
Arsenal FC 0-2 West Ham United
Newcastle United 2-2 Southampton FC
Stoke City 0-1 Liverpool FC
[Mon Aug 10]
West Bromwich Albion 0-3 Manchester City
...
(Source: england/2015-16/1-premierleague-i.txt)
and let's try:
## assumes football.db datasets for England in ./england directory
## see github.com/openfootball/england
SportDb.read( './england/2015-16/1-premierleague-i.txt' )
SportDb.read( './england/2015-16/1-premierleague-ii.txt' )
## let's try another season
SportDb.read( './england/2019-20/1-premierleague.txt' )All leagues, seasons, clubs, match days and rounds, match fixtures and results, and more are now in your (SQL) database of choice.
Bonus: As an alternative pass in the "package" directory or a zip archive and let read figure
out what datafiles to read in:
## assumes football.db datasets for England in ./england directory
## see github.com/openfootball/england
SportDb.read( './england' )
## -or- use a zip archive download
SportDb.read( './england.zip' )Step 1
Setup the (SQL) database. Let's use and build a single-file SQLite database (from scratch), as an example:
require 'sportdb/importers'
SportDb.connect( adapter: 'sqlite3',
database: './england.db' )
SportDb.create_all ## build database schema (tables, indexes, etc.)Step 2
Let's use the public domain football.csv datasets for England (see footballcsv/england), as an example:
Round, Date, Team 1, FT, HT, Team 2
1, (Fri) 9 Aug 2019, Liverpool FC, 4-1, 4-0, Norwich City FC
1, (Sat) 10 Aug 2019, West Ham United FC, 0-5, 0-1, Manchester City FC
1, (Sat) 10 Aug 2019, AFC Bournemouth, 1-1, 0-0, Sheffield United FC
1, (Sat) 10 Aug 2019, Burnley FC, 3-0, 0-0, Southampton FC
1, (Sat) 10 Aug 2019, Crystal Palace FC, 0-0, 0-0, Everton FC
1, (Sat) 10 Aug 2019, Watford FC, 0-3, 0-1, Brighton & Hove Albion FC
1, (Sat) 10 Aug 2019, Tottenham Hotspur FC, 3-1, 0-1, Aston Villa FC
1, (Sun) 11 Aug 2019, Leicester City FC, 0-0, 0-0, Wolverhampton Wanderers FC
1, (Sun) 11 Aug 2019, Newcastle United FC, 0-1, 0-0, Arsenal FC
1, (Sun) 11 Aug 2019, Manchester United FC, 4-0, 1-0, Chelsea FC
...
(Source: england/2019-20/eng.1.csv)
and let's try:
## assumes football.csv datasets for England in ./england directory
## see github.com/footballcsv/england
SportDb.read_csv( './england/2019-20/eng.1.csv' )
## let's try another season
SportDb.read_csv( './england/2018-19/eng.1.csv' )
SportDb.read_csv( './england/2018-19/eng.2.csv' )All leagues, seasons, clubs, match days and rounds, match fixtures and results, and more are now in your (SQL) database of choice.
Bonus: Let's import all datafiles for all seasons (from 1888-89 to today) for England, use:
## note: requires a local copy of the football.csv england datasets
## see https://github.com/footballcsv/england
SportDb.read_csv( './england' )
# -or- use a zip archive
SportDb.read_csv( './england.zip' )For now the importers support two flavors.
Alternative 1) One league and season per datafile and
the basename (e.g. eng.1 ) holds the league code
and the directory (e.g. 2019-20) the season.
Matchday, Date, Time, Team 1, FT, Team 2
1, Fri Aug 9 2019, 20:00, Liverpool, 4-1, Norwich City
1, Sat Aug 10 2019, 12:30, West Ham, 0-5, Manchester City
...
Alternative 2) Any or many leagues or seasons per datafile,
for example, week by week (see /updates)
or year by year (see /internationals).
Date, League, Team 1, FT, Team 2
Wed Jun 10 2020, DE 3, SpVgg Unterhaching, 1-3, Eintracht Braunschweig
Thu Jun 11 2020, AT 2, FC Blau-Weiß Linz, 1-2, Austria Klagenfurt
Thu Jun 11 2020, ES 1, Sevilla FC, 2-0, Real Betis
...
Note: For now the convention is that the datafile basename
MUST be all numbers, that is, 0 to 9 (plus - or _) e.g.
01 (as in 2020/01.csv) or 2020 (as in 2000s/2020.csv).
The importers ship with hundreds of zero-config preconfigured code and names for leagues & cups. See the /leagues datasets for all builtin national and international football club leagues & cups from around the world.
Or to query in ruby try:
require `sportdb/config`
LEAGUES = SportDb::Import.catalog.leagues
LEAGUES.find( 'ENG 1' ) #=> Premier League › England
LEAGUES.find( 'EPL' ) #=> Premier League › England
LEAGUES.find( 'ENG 2' ) #=> Championship › England
LEAGUES.find( 'ENG CS' ) #=> Championship › England
LEAGUES.find( 'ES' ) #=> Primera División › Spain
LEAUGES.find( 'ESP 1') #=> Primera División › Spain
...You can run any of the HTTP JSON API (web service) scripts using the sportdb command line tool.
By default the serve command will look for
a script named Service or service.rb (in the working folder, that is, ./). Example:
$ sportdb serve
To run any other script - copy the script into the working folder and pass it along as an argument. Example:
$ sportdb serve starter # note: will (auto-)add the .rb extension or
$ sportdb serve starter.rb
Send them along to the Open Sports & Friends Forum/Mailing List. Thanks!