footballdata-api 0.2.0 → 0.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 16f099d030a007993b7289956e920fd89baff6abc506feeca3dad792453efe76
-  data.tar.gz: 8f3d2df27a3dc12cc15fa2a8ee0dcd42a51ed793261a2ab2795020147596e7f6
+  metadata.gz: a0fe438df785fdc33e14a57346b2ec720029a97e34ae4a5eaeddf4e7fa4d8cda
+  data.tar.gz: e4eaf71567d20a666525139d94dbe4323fb2c8773e245b67454281e6740a9cb4
 SHA512:
-  metadata.gz: cca1d63da2e1823a3c9cc2636953da2f691101c32b63f5004c5349338d9dd75099ac993786e038cf3e2ea8a41f3ac1ce7510cdf148f1f13a98d360203613cf99
-  data.tar.gz: 6030f680eb2e5dadd269ee78bd778c1b465914702e20fe98e55e0967346feb67a6b74d8a965dea73ce62b9c8f457ba1de811d02e26a603b90ff4d3a160b8ee99
+  metadata.gz: 4081619b75253f9ad57b77ef6232e262b37af91a6e154e3832274520d7363a150a38ddf5a6596d12c5e28646419e530c569a902a74b807f78c987de02104e92f
+  data.tar.gz: 2639c457fea6f4c4a7cae24196f8133e72ccfa80c433963726b17b498f7e2fcb34dfbe7ae64afb78c76ebc5ff120905cc2f71b3ceec05deb5a535140b6761faa

data/CHANGELOG.md

@@ -1,4 +1,4 @@
-### 0.2.0
+### 0.3.0
 
 ### 0.0.1 / 2024-07-03
 

data/Manifest.txt

@@ -6,7 +6,6 @@ bin/fbdat
 lib/footballdata.rb
 lib/footballdata/convert.rb
 lib/footballdata/download.rb
-lib/footballdata/generator.rb
 lib/footballdata/leagues.rb
 lib/footballdata/mods.rb
 lib/footballdata/prettyprint.rb

data/Rakefile

@@ -19,15 +19,15 @@ Hoe.spec 'footballdata-api' do
 
   self.extra_deps = [
     ['tzinfo'],
-    ['season-formats'], 
-    ['webget'],    
+    ['season-formats'],
+    ['webget'],
     ['cocos'],  ## later pull in with sportsdb-writers
   ]
 
   self.licenses = ['Public Domain']
 
   self.spec_extras = {
-   required_ruby_version: '>= 2.2.2'
+   required_ruby_version: '>= 3.1.0'
   }
 
 end

data/bin/fbdat

@@ -12,17 +12,20 @@ load_env   ## use dotenv (.env)
 Webcache.root =  if File.exist?( '/sports/cache' )
                      puts "  setting web cache to >/sports/cache<"
                      '/sports/cache'
-                 else 
+                 else
                      './cache'
                  end
 
-## note -  free tier (tier one) plan - 10 requests/minute   
+## note -  free tier (tier one) plan - 10 requests/minute
 ##              (one request every 6 seconds 6*10=60 secs)
 ##     10 API calls per minute max.
 ##  note - default sleep (delay in secs) is 3 sec(s)
 Webget.config.sleep  = 10
 
 
+Footballdata.config.convert.out_dir = '/sports/cache.api.fbdat'   if File.exist?( '/sports/cache.api.fbdat' )
+
+
 
 
 require 'optparse'
@@ -32,17 +35,28 @@ require 'optparse'
 module Footballdata
 def self.main( args=ARGV )
 
-opts = {}
+opts = {
+   cached:  false,
+   convert: true,
+}
+
 parser = OptionParser.new do |parser|
-  parser.banner = "Usage: #{$PROGRAM_NAME} [options]"
+  parser.banner = "Usage: #{$PROGRAM_NAME} [options] [args]"
 
   parser.on( "--cache", "--cached", "--offline",
-               "use cached data in #{Webcache.root}" ) do |cached|
+               "use cached data in #{Webcache.root} - default is (#{opts[:cached]})" ) do |cached|
     opts[:cached] = cached
   end
+
+  parser.on( "--no-convert",
+               "turn off conversion to .csv in #{Footballdata.config.convert.out_dir} - default is (#{!opts[:convert]})" ) do |convert|
+    opts[:convert] = !convert
+  end
 end
 parser.parse!( args )
 
+
+
 puts "OPTS:"
 p opts
 puts "ARGV:"
@@ -51,23 +65,23 @@ p args
 
 ## try special args
 
-if ['plan', 'plans', 
-    'comp', 'comps'].include?(args[0]) 
+if ['plan', 'plans',
+    'comp', 'comps'].include?(args[0])
 
     if opts[:cached]
         ## do nothing
-    else    
+    else
       Metal.competitions( auth: true )   ## get free tier (TIER_ONE) with auth (token)
     end
 
   url = Metal.competitions_url
   pp url
   #=> "http://api.football-data.org/v4/competitions"
-   
+
   data = Webcache.read_json( url )
   pp data
-    
-  comps = data['competitions']  
+
+  comps = data['competitions']
   comps.each do |rec|
       print "==> "
       print "#{rec['area']['name']} (#{rec['area']['code']}) - "
@@ -75,12 +89,12 @@ if ['plan', 'plans',
       print "#{rec['plan']} #{rec['type']}, "
       print "#{rec['numberOfAvailableSeasons']} season(s)"
       print "\n"
-  
-      print "     #{rec['currentSeason']['startDate']} - #{rec['currentSeason']['endDate']} "  
+
+      print "     #{rec['currentSeason']['startDate']} - #{rec['currentSeason']['endDate']} "
       print "@ #{rec['currentSeason']['currentMatchday']}"
       print "\n"
   end
-  
+
   puts "  #{comps.size} competition(s)"
   exit 0
 end
@@ -100,9 +114,9 @@ end
 ## todo - add more date offsets - t+2,t+3,t+4, etc.
 
 date =  if ['y', 'yesterday', 't-1', '-1'].include?( args[0] )
-            Date.today-1       
+            Date.today-1
         elsif ['t', 'tomorrow', 't+1', '1', '+1'].include?( args[0] )
-            Date.today+1        
+            Date.today+1
         elsif ['m', 'match', 'matches', 'today'].include?( args[0] || 'today' )  ## make default - why? why not?
             Date.today
         else
@@ -110,7 +124,7 @@ date =  if ['y', 'yesterday', 't-1', '-1'].include?( args[0] )
         end
 
 if date
-  if opts[:cached] 
+  if opts[:cached]
       ## do nothing
   else
      Metal.todays_matches( date )
@@ -124,7 +138,7 @@ if date
     ## only print competition header if different from last match
     comp = fmt_competition( rec )
     puts comp   if comp != last_comp
-    
+
     puts fmt_match( rec )
 
     last_comp = comp
@@ -145,40 +159,48 @@ end
 ##   club cups intl
 ##    CL 2023  - (uefa/european) champions league
 ##    CLI 2204 - (south american) copa libertadores
-##   club leagues 
+##   club leagues
 ##    PL 2024  - england premiere league
 
+##
+## note - only use "generic" uniform league codes for now!!
 
-league_code       = args[0] || 'PL'
+league_code       = (args[0] || 'eng.1').downcase
+metal_league_code  = nil  ##  todo - find a better name
+                          ##  use internal_league_code or such - why? why not?
 
 ###  convenience helpers - lets you use eng.1, euro, etc.
 ## check if mapping for league_code
-if LEAGUES.has_key?( league_code.downcase )
-  league_code = LEAGUES[ league_code.downcase ]
+if LEAGUES.has_key?( league_code )
+  metal_league_code = LEAGUES[ league_code ]
 else
-   ## assume  "native" code
-   ##  always upcase e.g. pl => PL etc.
-   league_code = league_code.upcase
+  puts "!! ERROR - no code/mapping found for league >#{league_code}<"
+  puts "     mappings include:"
+  pp LEAGUES
+  exit 1
 end
 
 
-season            = Season( args[1] ||  
-                             (league_code == 'EC' ? '2024' : '2024/25'))
+season            = Season( args[1] ||
+                             (['euro',
+                               'copa.l',
+                               'br.1'].include?(league_code) ? '2024' : '2024/25'))
 
 season_start_year  = season.start_year   ## use year - why? why not?
 
-pp league_code, season_start_year
+pp [metal_league_code, season_start_year]
 
 if opts[:cached]
     ## do nothing
 else
     ## download dataset(s)
     ## try download
-    Metal.matches( league_code, 
-                   season_start_year )
-end    
+    ##   note:  include teams (for convert) for now too!!
+    Metal.teams( metal_league_code, season_start_year )
+    Metal.matches( metal_league_code, season_start_year )
+end
 
-url = Metal.competition_matches_url( league_code, 
+url = Metal.competition_matches_url( metal_league_code,
                                       season_start_year )
 pp url
 #=> "http://api.football-data.org/v4/competitions/EC/matches?season=2024"
@@ -188,6 +210,13 @@ data = Webcache.read_json( url )
 
 pp_matches( data )
 
+
+if opts[:convert]
+  puts "==> converting to .csv"
+  convert( league: league_code, season: season )
+end
+
+
 end  # def self.main
 end  # module Footballdata
 

data/lib/footballdata/convert.rb

@@ -5,30 +5,113 @@ module Footballdata
 
 
   TIMEZONES = {
-    'eng.1' => 'Europe/London',  
+    'eng.1' => 'Europe/London',
     'eng.2' => 'Europe/London',
-  
+
     'es.1'  => 'Europe/Madrid',
-  
+
     'de.1'  => 'Europe/Berlin',
-    'fr.1'  => 'Europe/Paris', 
+    'fr.1'  => 'Europe/Paris',
     'it.1'  => 'Europe/Rome',
     'nl.1'  => 'Europe/Amsterdam',
-  
-    'pt.1'  => 'Europe/Lisbon',   
+
+    'pt.1'  => 'Europe/Lisbon',
+
+    ## for champs default for not to cet (central european time) - why? why not?
+    'uefa.cl'  => 'Europe/Paris',
+    'euro'     => 'Europe/Paris',
 
     ## todo/fix - pt.1
     ##  one team in madeira!!! check for different timezone??
-    ##  CD Nacional da Madeira 
+    ##  CD Nacional da Madeira
 
     'br.1'  => 'America/Sao_Paulo',
     ## todo/fix - brazil has 4 timezones
     ##           really only two in use for clubs
     ##             west and east (amazonas et al)
     ##           for now use west for all - why? why not?
+    'copa.l'  => 'America/Sao_Paulo',
   }
 
 
+
+
+def self.convert_score( score )
+  ## duration: REGULAR  · PENALTY_SHOOTOUT  · EXTRA_TIME
+  ft, ht, et, pen = ["","","",""]
+
+  if score['duration'] == 'REGULAR'
+    ft  = "#{score['fullTime']['home']}-#{score['fullTime']['away']}"
+    ht  = "#{score['halfTime']['home']}-#{score['halfTime']['away']}"
+  elsif score['duration'] == 'EXTRA_TIME'
+    et  =  "#{score['regularTime']['home']+score['extraTime']['home']}"
+    et << "-"
+    et << "#{score['regularTime']['away']+score['extraTime']['away']}"
+
+    ft =  "#{score['regularTime']['home']}-#{score['regularTime']['away']}"
+    ht =  "#{score['halfTime']['home']}-#{score['halfTime']['away']}"
+  elsif score['duration'] == 'PENALTY_SHOOTOUT'
+    if score['extraTime']
+      ## quick & dirty hack - calc et via regulartime+extratime
+      pen = "#{score['penalties']['home']}-#{score['penalties']['away']}"
+      et  = "#{score['regularTime']['home']+score['extraTime']['home']}"
+      et << "-"
+      et << "#{score['regularTime']['away']+score['extraTime']['away']}"
+
+      ft = "#{score['regularTime']['home']}-#{score['regularTime']['away']}"
+      ht = "#{score['halfTime']['home']}-#{score['halfTime']['away']}"
+    else  ### south american-style (no extra time)
+        ## quick & dirty hacke - calc ft via fullTime-penalties
+        pen =  "#{score['penalties']['home']}-#{score['penalties']['away']}"
+        ft  =  "#{score['fullTime']['home']-score['penalties']['home']}"
+        ft << "-"
+        ft << "#{score['fullTime']['away']-score['penalties']['away']}"
+        ht  = "#{score['halfTime']['home']}-#{score['halfTime']['away']}"
+    end
+  else
+    puts "!! unknown score duration:"
+    pp score
+    exit 1
+  end
+
+  [ft,ht,et,pen]
+end
+
+
+#######
+##  map round-like to higher-level stages
+STAGES = {
+  'REGULAR_SEASON'          => ['Regular'],
+
+  'PRELIMINARY_ROUND'       => ['Qualifying', 'Preliminary Round' ],
+  'PRELIMINARY_SEMI_FINALS' => ['Qualifying', 'Preliminary Semifinals' ],
+  'PRELIMINARY_FINAL'       => ['Qualifying', 'Preliminary Final' ],
+  '1ST_QUALIFYING_ROUND'    => ['Qualifying', 'Qual. Round 1' ],
+  '2ND_QUALIFYING_ROUND'    => ['Qualifying', 'Qual. Round 2' ],
+  '3RD_QUALIFYING_ROUND'    => ['Qualifying', 'Qual. Round 3' ],
+  'QUALIFICATION_ROUND_1'   => ['Qualifying', 'Qual. Round 1' ],
+  'QUALIFICATION_ROUND_2'   => ['Qualifying', 'Qual. Round 2' ],
+  'QUALIFICATION_ROUND_3'   => ['Qualifying', 'Qual. Round 3' ],
+  'ROUND_1'                 => ['Qualifying', 'Round 1'],    ##  use Qual. Round 1 - why? why not?
+  'ROUND_2'                 => ['Qualifying', 'Round 2'],
+  'ROUND_3'                 => ['Qualifying', 'Round 3'],
+  'PLAY_OFF_ROUND'          => ['Qualifying', 'Playoff Round'],
+  'PLAYOFF_ROUND_1'         => ['Qualifying', 'Playoff Round 1'],
+
+  'LEAGUE_STAGE'            => ['League'],
+  'GROUP_STAGE'             => ['Group'],
+  'PLAYOFFS'                => ['Playoffs'],
+
+  'ROUND_OF_16'             => ['Finals',     'Round of 16'],
+  'LAST_16'                 => ['Finals',     'Round of 16'],  ## use Last 16 - why? why not?
+  'QUARTER_FINALS'          => ['Finals',     'Quarterfinals'],
+  'SEMI_FINALS'             => ['Finals',     'Semifinals'],
+  'FINAL'                   => ['Finals',     'Final'],
+}
+
+
+
+
 def self.convert( league:, season: )
 
   ### note/fix: cl (champions league for now is a "special" case)
@@ -48,14 +131,14 @@ def self.convert( league:, season: )
   data           = Webcache.read_json( matches_url )
   data_teams     = Webcache.read_json( teams_url )
 
-  
+
   ## check for time zone
   tz_name = TIMEZONES[ league.downcase ]
   if tz_name.nil?
     puts "!! ERROR - sorry no timezone configured for league #{league}"
     exit 1
   end
-  
+
   tz  = TZInfo::Timezone.get( tz_name )
   pp tz
 
@@ -67,9 +150,9 @@ def self.convert( league:, season: )
      h
   end
 
-  pp teams_by_name.keys
+  ## pp teams_by_name.keys
+
 
-  
 
 mods = MODS[ league.downcase ] || {}
 
@@ -81,22 +164,62 @@ teams = Hash.new( 0 )
 
 # stat  =  Stat.new
 
-#  track stati counts 
-stati = Hash.new(0)
+  ## track stage, match status et
+  stats = { 'status'    => Hash.new(0),
+            'stage'     => Hash.new(0),
+           }
+
+
 
 
 matches = data[ 'matches']
 matches.each do |m|
   # stat.update( m )
 
-  team1 = m['homeTeam']['name']
-  team2 = m['awayTeam']['name']
+  ## use ? or N.N. or ? for nil  - why? why not?
+  team1 = m['homeTeam']['name'] || 'N.N.'
+  team2 = m['awayTeam']['name'] || 'N.N.'
 
   score = m['score']
 
 
+  stage_key = m['stage']
+
+  stats['stage'][ stage_key ] += 1   ## track stage counts
+
+  ## map stage to stage + round
+  stage, stage_round  =  STAGES[ stage_key ]
+
+  if stage.nil?
+      puts "!! ERROR - no stage mapping found for stage >#{stage_key}<"
+      exit 1
+  end
+
+  matchday_num = m['matchday']
+  matchday_num = nil   if matchday_num == 0   ## change 0 to nil (empty) too
+
+  if stage_round.nil?  ## e.g. Regular, League, Group, Playoffs
+     ## keep/assume matchday number is matchday .e.g
+     ##   matchday 1, 2 etc.
+     matchday = matchday_num.to_s
+  else
+    ## note - if matchday defined; assume leg e.g. 1|2
+    ##     skip if different than one or two for now
+    matchday = String.new
+    matchday << stage_round
+    matchday << " | Leg #{matchday_num}"  if matchday_num &&
+                                         (matchday_num == 1 || matchday_num == 2)
+  end
+
+
+
+  group = m['group'] || ''
+  ## GROUP_A
+  ##  shorten group to A|B|C etc.
+  group = group.sub( /^GROUP_/, '' )
+
+
 
-  if m['stage'] == 'REGULAR_SEASON'
     teams[ team1 ] += 1
     teams[ team2 ] += 1
 
@@ -107,31 +230,36 @@ matches.each do |m|
     end
 
 
+   ## auto-fix copa.l 2024
+   ##  !! ERROR: unsupported match status >IN_PLAY< - sorry:
+   if m['status'] == 'IN_PLAY' &&
+      team1 == 'Club Aurora' && team2 == 'FBC Melgar'
+        m['status'] = 'FINISHED'
+   end
+
 
     comments = ''
     ft       = ''
     ht       = ''
+    et       = ''
+    pen      = ''
 
-    stati[m['status']] += 1  ## track stati counts for logs
+    stats['status'][m['status']]  += 1  ## track status counts
 
     case m['status']
     when 'SCHEDULED', 'TIMED'   ## , 'IN_PLAY'
       ft = ''
       ht = ''
     when 'FINISHED'
-      ## todo/fix: assert duration == "REGULAR"
-      assert( score['duration'] == 'REGULAR', 'score.duration REGULAR expected' ) 
-      ft = "#{score['fullTime']['home']}-#{score['fullTime']['away']}"
-      ht = "#{score['halfTime']['home']}-#{score['halfTime']['away']}"
+      ft, ht, et, pen = convert_score( score )
     when 'AWARDED'
-      ## todo/fix: assert duration == "REGULAR"
-      assert( score['duration'] == 'REGULAR', 'score.duration REGULAR expected' ) 
+      assert( score['duration'] == 'REGULAR', 'score.duration REGULAR expected' )
       ft = "#{score['fullTime']['home']}-#{score['fullTime']['away']}"
       ft << ' (*)'
       ht = ''
       comments = 'awarded'
     when 'CANCELLED'
-      ## note cancelled might have scores!!
+      ## note cancelled might have scores!! -- add/fix later!!!
       ##   ht only or ft+ht!!!  (see fr 2021/22)
       ft = '(*)'
       ht = ''
@@ -157,55 +285,36 @@ matches.each do |m|
     ## utc   = ## tz_utc.strptime( m['utcDate'], '%Y-%m-%dT%H:%M:%SZ' )
     ##  note:  DateTime.strptime  is supposed to be unaware of timezones!!!
     ##            use to parse utc
-    utc = DateTime.strptime( m['utcDate'], '%Y-%m-%dT%H:%M:%SZ' ).to_time.utc 
+    utc = DateTime.strptime( m['utcDate'], '%Y-%m-%dT%H:%M:%SZ' ).to_time.utc
     assert( utc.strftime( '%Y-%m-%dT%H:%M:%SZ' ) == m['utcDate'], 'utc time mismatch' )
-    
+
     local = tz.to_local( utc )
-   
+
 
     ## do NOT add time if status is SCHEDULED
     ##                        or POSTPONED for now
     ##   otherwise assume time always present - why? why not?
-  
+
+
 
     ## todo/fix: assert matchday is a number e.g. 1,2,3, etc.!!!
-    recs << [m['matchday'].to_s,   ## note: convert integer to string!!!
+    recs << [stage,
+             group,
+             matchday,
              local.strftime( '%Y-%m-%d' ),
              ['SCHEDULED','POSTPONED'].include?( m['status'] ) ? '' : local.strftime( '%H:%M' ),
-             local.strftime( '%Z / UTC%z' ), 
+             local.strftime( '%Z / UTC%z' ),
              team1,
              ft,
              ht,
              team2,
+             et,
+             pen,
              comments,
              ## add more columns e.g. utc date, status
              m['status'],  # e.g. FINISHED, TIMED, etc.
              m['utcDate'],
             ]
-
-
-    print '%2s' % m['matchday']
-    print ' - '
-    print '%-26s' % team1
-    print '  '
-    print ft
-    print ' '
-    print "(#{ht})"    unless ht.empty?
-    print '  '
-    print '%-26s' % team2
-    print '  '
-    print comments
-    print ' | '
-    ## print date.to_date  ## strip time
-    print utc.strftime( '%a %b %-d %Y' )
-    print ' -- '
-    print utc
-    print "\n"
-  else
-    puts "!!! unexpected stage:"
-    puts "-- skipping #{m['stage']}"
-    # exit 1
-  end
 end # each match
 
 
@@ -227,8 +336,8 @@ dates = "#{start_date.strftime('%b %-d')} - #{end_date.strftime('%b %-d')}"
 
 buf = ''
 buf << "#{season.key} (#{dates}) - "
-buf << "#{teams.keys.size} clubs, "
-# buf << "#{stat[:regular_season][:matches]} matches, "
+buf << "#{teams.keys.size} teams, "
+buf << "#{recs.size} matches"
 # buf << "#{stat[:regular_season][:goals]} goals"
 buf << "\n"
 
@@ -248,7 +357,7 @@ puts buf
 
    File.open( './logs.txt', 'a:utf-8' ) do |f|
      f.write "====  #{league} #{season.key}  =============\n"
-     f.write "  match stati: #{stati.inspect}\n"
+     f.write "  #{stats.inspect}\n"
    end
 
 =begin
@@ -283,26 +392,30 @@ puts buf
 
 
 ## reformat date / beautify e.g. Sat Aug 7 1993
-recs = recs.map do |rec| 
-           rec[1] = Date.strptime( rec[1], '%Y-%m-%d' ).strftime( '%a %b %-d %Y' ) 
+recs = recs.map do |rec|
+           rec[3] = Date.strptime( rec[3], '%Y-%m-%d' ).strftime( '%a %b %-d %Y' )
            rec
        end
 
+## pp recs
+
+## check if all status colums
+###  are FINISHED
+###   if yes, set all to empty (for vacuum)
+
+if stats['status'].keys.size == 1 && stats['status'].keys[0] == 'FINISHED'
+  recs = recs.map { |rec| rec[-2] = ''; rec }
+end
+
+if stats['stage'].keys.size == 1 && stats['stage'].keys[0] == 'REGULAR_SEASON'
+  recs = recs.map { |rec| rec[0] = ''; rec }
+end
+
+
+recs, headers = vacuum( recs )
+
+
 
-headers = [
-  'Matchday',
-  'Date',
-  'Time',
-  'Timezone',  ## move back column - why? why not?
-  'Team 1',
-  'FT',
-  'HT',
-  'Team 2',
-  'Comments',
-  ##
-  'Status', # e.g. 
-  'UTC',    # date utc
-]
 
 ## note: change season_key from 2019/20 to 2019-20  (for path/directory!!!!)
   write_csv( "#{config.convert.out_dir}/#{season.to_path}/#{league.downcase}.csv",
@@ -319,14 +432,89 @@ teams.each do |name, count|
     print " › #{rec['area']['name']}"
     print "  - #{rec['address']}"
   else
-    puts "!! ERROR  - no team record found in teams.json for #{name}"
-    exit 1
+    if name == 'N.N.'
+      ## ignore missing record
+    else
+      puts "!! ERROR  - no team record found in teams.json for #{name}"
+      exit 1
+    end
   end
   print "\n"
 end
 
 ## pp stat
 end   # method convert
+
+
+
+MAX_HEADERS = [
+  'Stage',    # 0
+  'Group',    # 1
+  'Matchday', # 2
+  'Date',     # 3
+  'Time',     # 4
+  'Timezone',   # 5 ## move back column - why? why not?
+  'Team 1',     # 6
+  'FT',         # 7
+  'HT',         # 8
+  'Team 2',     # 9
+  'ET',         # 10     # extra: incl. extra time
+  'P',          # 11     # extra: incl. penalties
+  'Comments',   # 12
+  'Status',     # 13 / -2  # e.g.
+  'UTC',        # 14 / -1 # date utc
+]
+
+MIN_HEADERS = [   ## always keep even if all empty
+'Date',
+'Team 1',
+'FT',
+'Team 2'
+]
+
+
+
+def self.vacuum( rows, headers: MAX_HEADERS, fixed_headers: MIN_HEADERS )
+  ## check for unused columns and strip/remove
+  counter = Array.new( MAX_HEADERS.size, 0 )
+  rows.each do |row|
+     row.each_with_index do |col, idx|
+       counter[idx] += 1  unless col.nil? || col.empty?
+     end
+  end
+
+  ## pp counter
+
+  ## check empty columns
+  headers       = []
+  indices       = []
+  empty_headers = []
+  empty_indices = []
+
+  counter.each_with_index do |num, idx|
+     header = MAX_HEADERS[ idx ]
+     if num > 0 || (num == 0 && fixed_headers.include?( header ))
+       headers << header
+       indices << idx
+     else
+       empty_headers << header
+       empty_indices << idx
+     end
+  end
+
+  if empty_indices.size > 0
+    rows = rows.map do |row|
+             row_vacuumed = []
+             row.each_with_index do |col, idx|
+               ## todo/fix: use values or such??
+               row_vacuumed << col   unless empty_indices.include?( idx )
+             end
+             row_vacuumed
+         end
+    end
+
+  [rows, headers]
+end
 end #  module Footballdata
 
 

data/lib/footballdata/mods.rb

@@ -6,6 +6,8 @@ module Footballdata
   # Cardiff City FC | Cardiff  › Wales  - Cardiff City Stadium, Leckwith Road Cardiff CF11 8AZ
   # AS Monaco FC | Monaco  › Monaco     - Avenue des Castellans Monaco 98000
 
+
+
   MODS = {
     'br.1' => {
        'América FC' => 'América Mineiro',  # in year 2018 ??

data/lib/footballdata/prettyprint.rb

@@ -17,7 +17,7 @@ def self.fmt_competition( rec )
    buf << "#{rec['competition']['name']} (#{rec['competition']['code']}) -- "
    buf << "#{rec['area']['name']} (#{rec['area']['code']}) "
    buf << "#{rec['competition']['type']} "
-   buf << "#{rec['season']['startDate']} - #{rec['season']['endDate']} "  
+   buf << "#{rec['season']['startDate']} - #{rec['season']['endDate']} "
    buf << "@ #{rec['season']['currentMatchday']}"
    buf << "\n"
 
@@ -26,12 +26,12 @@ end
 
 def self.fmt_match( rec )
    buf = String.new
- 
+
     ## -- todo - make sure / assert it's always utc - how???
     ## utc   = ## tz_utc.strptime( m['utcDate'], '%Y-%m-%dT%H:%M:%SZ' )
     ##  note:  DateTime.strptime  is supposed to be unaware of timezones!!!
     ##            use to parse utc
-    utc = DateTime.strptime( rec['utcDate'], '%Y-%m-%dT%H:%M:%SZ' ).to_time.utc 
+    utc = DateTime.strptime( rec['utcDate'], '%Y-%m-%dT%H:%M:%SZ' ).to_time.utc
     assert( utc.strftime( '%Y-%m-%dT%H:%M:%SZ' ) == rec['utcDate'], 'utc time mismatch' )
 
     status = rec['status']
@@ -50,7 +50,7 @@ def self.fmt_match( rec )
      team1 = rec['homeTeam']['name'] ?
                   "#{rec['homeTeam']['name']} (#{rec['homeTeam']['tla']})" : '?'
      team2 = rec['awayTeam']['name'] ?
-                  "#{rec['awayTeam']['name']} (#{rec['awayTeam']['tla']})" : '?'     
+                  "#{rec['awayTeam']['name']} (#{rec['awayTeam']['tla']})" : '?'
      buf << '%22s' % team1
      buf << " - "
      buf << '%-22s' % team2
@@ -58,7 +58,7 @@ def self.fmt_match( rec )
 
      stage = rec['stage']
      group = rec['group']
-     
+
      buf << "#{rec['matchday']} - #{stage} "
      buf << "/ #{group}  "  if group
      buf << "\n"
@@ -66,49 +66,31 @@ def self.fmt_match( rec )
      buf << "  "
      buf << '%-20s' % rec['score']['duration']
      buf << ' '*24
-     
+
      duration = rec['score']['duration']
      assert( %w[REGULAR
                 EXTRA_TIME
                 PENALTY_SHOOTOUT
                ].include?( duration ), "unknown duration - #{duration}" )
 
-     score = String.new 
-
-     if duration == 'PENALTY_SHOOTOUT'
-        if rec['score']['extraTime'] 
-          ## quick & dirty hack - calc et via regulartime+extratime 
-          score << "#{rec['score']['penalties']['home']}-#{rec['score']['penalties']['away']} pen. "
-          score << "#{rec['score']['regularTime']['home']+rec['score']['extraTime']['home']}"
-          score << "-"
-          score << "#{rec['score']['regularTime']['away']+rec['score']['extraTime']['away']}"
-          score << " a.e.t. "
-          score << "(#{rec['score']['regularTime']['home']}-#{rec['score']['regularTime']['away']},"
-          score << "#{rec['score']['halfTime']['home']}-#{rec['score']['halfTime']['away']})"
-        else  ### south american-style (no extra time)
-            ## quick & dirty hacke - calc ft via fullTime-penalties
-            score << "#{rec['score']['penalties']['home']}-#{rec['score']['penalties']['away']} pen. "
-            score << "(#{rec['score']['fullTime']['home']-rec['score']['penalties']['home']}"
-            score << "-"
-            score << "#{rec['score']['fullTime']['away']-rec['score']['penalties']['away']},"
-            score << "#{rec['score']['halfTime']['home']}-#{rec['score']['halfTime']['away']})"  
-        end
-     elsif  duration == 'EXTRA_TIME'
-          score << "#{rec['score']['regularTime']['home']+rec['score']['extraTime']['home']}"
-          score << "-"
-          score << "#{rec['score']['regularTime']['away']+rec['score']['extraTime']['away']}"
-          score << " a.e.t. "
-          score << "(#{rec['score']['regularTime']['home']}-#{rec['score']['regularTime']['away']},"
-          score << "#{rec['score']['halfTime']['home']}-#{rec['score']['halfTime']['away']})"          
-     elsif  duration == 'REGULAR'
-          if rec['score']['fullTime']['home'] && rec['score']['fullTime']['away']
-            score << "#{rec['score']['fullTime']['home']}-#{rec['score']['fullTime']['away']} "
-            score << "(#{rec['score']['halfTime']['home']}-#{rec['score']['halfTime']['away']})"
-          end          
+
+     ft, ht, et, pen = convert_score( rec['score'] )
+     score = String.new
+     if !pen.empty?
+       if et.empty? ### south american-style (no extra time)
+         score << "#{pen} pen. "
+         score << "(#{ft}, #{ht})"
+       else
+         score << "#{pen} pen. "
+         score << "#{et} a.e.t. "
+         score << "(#{ft}, #{ht})"
+      end
+     elsif !et.empty?
+       score << "#{et} a.e.t. "
+       score << "(#{ft}, #{ht})"
      else
-        raise ArgumentError, "unexpected/unknown score duration #{rec['score']['duration']}"
+       score << "#{ft} (#{ht})"
      end
- 
 
      buf << score
      buf << "\n"
@@ -118,14 +100,14 @@ end
 
 def self.pp_matches( data )
 
-  ## track match status and score duration  
+  ## track match status and score duration
   stats = { 'status'    => Hash.new(0),
             'duration'  => Hash.new(0),
             'stage'     => Hash.new(0),
             'group'     => Hash.new(0),
            }
 
-  first = Date.strptime( data['resultSet']['first'], '%Y-%m-%d' ) 
+  first = Date.strptime( data['resultSet']['first'], '%Y-%m-%d' )
   last  = Date.strptime( data['resultSet']['last'], '%Y-%m-%d' )
 
   diff = (last - first).to_i  # note - returns rational number (e.g. 30/1)
@@ -142,16 +124,16 @@ def self.pp_matches( data )
 
     ## track stats
     status = rec['status']
-    stats['status'][status] += 1 
+    stats['status'][status] += 1
 
     stage = rec['stage']
-    stats['stage'][stage] += 1  
+    stats['stage'][stage] += 1
 
      group = rec['group']
-     stats['group'][group] += 1  if group 
+     stats['group'][group] += 1  if group
 
      duration = rec['score']['duration']
-     stats['duration'][duration] += 1       
+     stats['duration'][duration] += 1
   end
 
   print "   #{data['resultSet']['played']}/#{data['resultSet']['count']} matches"
@@ -182,7 +164,7 @@ def self.fmt_count( h, sort: false )
                             end
     end
     pairs = pairs.map { |name,count| "#{name} (#{count})" }
-    pairs.join( ' · ' )                                                                 
+    pairs.join( ' · ' )
 end
 
 

data/lib/footballdata/version.rb

@@ -1,7 +1,7 @@
 
 module FootballdataApi
   MAJOR = 0    ## todo: namespace inside version or something - why? why not??
-  MINOR = 2
+  MINOR = 3
   PATCH = 0
   VERSION = [MAJOR,MINOR,PATCH].join('.')
 

data/lib/footballdata.rb

@@ -17,7 +17,7 @@ module Footballdata
         def out_dir()       @out_dir || './o'; end
         def out_dir=(value) @out_dir = value; end
      end
- 
+
     def convert()  @convert ||= Convert.new; end
   end # class Configuration
 
@@ -44,13 +44,6 @@ require_relative 'footballdata/convert'
 require_relative 'footballdata/teams'
 
 
-require_relative 'footballdata/generator'
-
-
-
-### for processing tool
-##   (auto-)add sportdb/writer  (pulls in sportdb/catalogs and gitti)
-##  require 'sportdb/writers'
 
 
 puts FootballdataApi.banner  ## say hello

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: footballdata-api
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.3.0
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-07-07 00:00:00.000000000 Z
+date: 2024-09-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: tzinfo
@@ -119,7 +119,6 @@ files:
 - lib/footballdata.rb
 - lib/footballdata/convert.rb
 - lib/footballdata/download.rb
-- lib/footballdata/generator.rb
 - lib/footballdata/leagues.rb
 - lib/footballdata/mods.rb
 - lib/footballdata/prettyprint.rb
@@ -140,7 +139,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 2.2.2
+      version: 3.1.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="

data/lib/footballdata/generator.rb

@@ -1,33 +0,0 @@
-
-##
-### check - change Generator to Writer 
-##                   and write( league:, season: ) - why? why not?
-
-module Footballdata
-class Generator
-
-
-
-###########    
-## always download for now - why? why not?
-##      support cache - why? why not?
-def generate( league:, season: )
-
-  ## for testing use cached version always - why? why not?    
-  ## step 1 - download
-  Footballdata.schedule( league: league, season: season )
-
-  ## step 2 - convert (to .csv)
-
-  ## todo/fix - convert in-memory and return matches
-  Footballdata.convert( league: league, season: season )
-
-  source_dir = Footballdata.config.convert.out_dir
-
-  Writer.write( league: league, 
-                season: season,
-                source: source_dir )
-end  # def generate
-
-end # class Generator
-end # module Footballdata