sportdb-quick 0.0.1 → 0.1.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f2ce8e81bd6aece43872ba2718123dd2b34d6f52807ab8260a2e91f01a8c28ed
-  data.tar.gz: c91175dabbac6e590198df83151ee36d9a9daa853b77f0eedfa229f9db55edd0
+  metadata.gz: 5ee93b658f2f6fce94a131931ea37908a2edbfa19a182ec20072447b57ffc847
+  data.tar.gz: f5fa5ab664d71ea9a2ab8c62e1bd30d42fa678e0b2a7564124623dc5d7cd2963
 SHA512:
-  metadata.gz: 53f5ca14405552fe662d35e6b1d1672afce956f27cc40801265a9f9d44a4d622b320a00a5ba935a04bae0453872ecb0da27b21927b184a87ed2755973eef118b
-  data.tar.gz: a688b66358bf16ec13ccbb90b5e29eaf483ca3e69f64ca4816f7210038383ea002660fa62a8ac971e9764903490757c9a84864bf57075794d8e0a337970a55b4
+  metadata.gz: b0df0b970600ee24a88d09ddffec1caa8ef94cfa75561569638da954357cec2d97ffc7ea566f03e42ec814a5998c99da79f861c54a863a439f1778a3f0426ab0
+  data.tar.gz: 01e1ad205c4d271a68136784a17b8f1b7987bdfa5e9b4ba5f395b42f6b488a53747525ecd7bea1e2c188012219529d7611e7d35f49ec4e5e7d499cc791e4e80a

data/CHANGELOG.md

@@ -1,3 +1,5 @@
+### 0.1.1
+
 ### 0.0.1 / 2024-08-27
 
 * Everything is new. First release.

data/Manifest.txt

@@ -4,10 +4,12 @@ README.md
 Rakefile
 bin/fbt
 lib/sportdb/quick.rb
-lib/sportdb/quick/linter.rb
+lib/sportdb/quick/csv/goal.rb
+lib/sportdb/quick/csv/goal_parser_csv.rb
+lib/sportdb/quick/csv/match_parser_csv.rb
+lib/sportdb/quick/csv/match_status_parser.rb
 lib/sportdb/quick/match_parser.rb
 lib/sportdb/quick/opts.rb
-lib/sportdb/quick/outline_reader.rb
 lib/sportdb/quick/quick_league_outline_reader.rb
 lib/sportdb/quick/quick_match_reader.rb
 lib/sportdb/quick/version.rb

data/README.md

@@ -1,5 +1,24 @@
-# sportdb-quick - football.txt (quick) match parsers and more
+# sportdb-quick - football.txt (quick) match readers and more
 
 
 
 
+## Usage
+
+
+``` ruby
+require 'sportdb/quick'
+
+
+# path = "./euro/2024--germany/euro.txt"
+path =  "./deutschland/2024-25/1-bundesliga.txt"
+
+matches = SportDb::QuickMatchReader.read( path )
+pp matches
+
+#  try json for matches
+data = matches.map {|match| match.as_json }
+pp data
+```
+
+

data/Rakefile

@@ -6,7 +6,7 @@ Hoe.spec 'sportdb-quick' do
 
   self.version = SportDb::Module::Quick::VERSION
 
-  self.summary = "sportdb-quick - football.txt (quick) match parsers and more"
+  self.summary = "sportdb-quick - football.txt (quick) match readers and more"
   self.description = summary
 
   self.urls = { home: 'https://github.com/sportdb/sport.db' }

data/bin/fbt

@@ -27,7 +27,7 @@ require 'optparse'
  args = ARGV
  opts = { debug: false,
           metal: false,
-          quick: false }
+          quick: true }
 
  parser = OptionParser.new do |parser|
   parser.banner = "Usage: #{$PROGRAM_NAME} [options]"
@@ -39,19 +39,19 @@ require 'optparse'
 
   parser.on( "--verbose", "--debug",
                "turn on verbose / debug output (default: #{opts[:debug]})" ) do |debug|
-    opts[:debug] = debug
+    opts[:debug] = true
   end
 
   parser.on( "--metal",
                  "turn off typed parse tree; show to the metal tokens"+
                    " (default: #{opts[:metal]})" ) do |metal|
-    opts[:metal] = metal
+    opts[:metal] = true
   end
 
   parser.on( "--quick",
                 "use quick match reader; output matches in json"+
                    " (default: #{opts[:quick]})" ) do |quick|
-     opts[:quick] = quick
+     opts[:quick] = true
   end
 end
 parser.parse!( args )
@@ -92,9 +92,9 @@ if opts[:quick]
     puts "  #{data.size} match(es)"
   end
 else
-  SportDb::Quick::Linter.debug = true    if opts[:debug]
+  SportDb::Parser::Linter.debug = true    if opts[:debug]
 
-  linter = SportDb::Quick::Linter.new
+  linter = SportDb::Parser::Linter.new
 
   errors = []
 

data/lib/sportdb/quick/csv/goal.rb

@@ -0,0 +1,192 @@
+
+module Sports
+
+## "free-standing" goal event - for import/export  in separate event / goal datafiles
+##   returned by CsvGoalParser and others
+class GoalEvent
+
+  def self.build( row )  ## rename to parse or such - why? why not?
+
+  ## split match_id
+  team_str, more_str   = row['Match'].split( '|' )
+  team1_str, team2_str = team_str.split( ' - ' )
+
+  more_str  = more_str.strip
+  team1_str = team1_str.strip
+  team2_str = team2_str.strip
+
+   # check if more_str is a date otherwise assume round
+   date_fmt = if more_str =~ /^[A-Z]{3} [0-9]{1,2}$/i  ## Apr 4
+                '%b %d'
+              elsif more_str =~ /^[A-Z]{3} [0-9]{1,2} [0-9]{4}$/i  ## Apr 4 2019
+                '%b %d %Y'
+             else
+                nil
+             end
+
+   if date_fmt
+    date  = Date.strptime( more_str, date_fmt )
+    round = nil
+   else
+    date  = nil
+    round = more_str
+   end
+
+
+    values = row['Score'].split('-')
+    values = values.map { |value| value.strip }
+    score1 = values[0].to_i
+    score2 = values[1].to_i
+
+    minute = nil
+    offset = nil
+    if m=%r{([0-9]+)
+              (?:[ ]+
+                   \+([0-9]+)
+                )?
+                ['.]
+          $}x.match( row['Minute'])
+      minute = m[1].to_i
+      offset = m[2] ? m[2].to_i : nil
+    else
+      puts "!! ERROR - unsupported minute (goal) format >#{row['Minute']}<"
+      exit 1
+    end
+
+    attributes = {
+          team1:  team1_str,
+          team2:  team2_str,
+          date:   date,
+          round:  round,
+          score1: score1,
+          score2: score2,
+          minute: minute,
+          offset: offset,
+          player: row['Player'],
+          owngoal: ['(og)', '(o.g.)'].include?( row['Extra']),
+          penalty: ['(pen)', '(pen.)'].include?( row['Extra']),
+          notes:   (row['Notes'].nil? || row['Notes'].empty?) ? nil : row['Notes']
+        }
+
+    new( **attributes )
+  end
+
+
+  ## match id
+  attr_reader  :team1,
+               :team2,
+               :round,  ## optional
+               :date   ## optional
+
+  ## main attributes
+  attr_reader :score1,
+              :score2,
+              :player,
+              :minute,
+              :offset,
+              :owngoal,
+              :penalty,
+              :notes
+
+
+  ## todo/check: or just use match.hash or such if match mapping known - why? why not?
+  def match_id
+    if round
+      "#{@team1} - #{@team2} | #{@round}"
+    else
+      "#{@team1} - #{@team2} | #{@date}"
+    end
+  end
+
+
+  def owngoal?() @owngoal==true; end
+  def penalty?() @penalty==true; end
+
+  def initialize( team1:,
+                  team2:,
+                  round:   nil,
+                  date:    nil,
+                  score1:,
+                  score2:,
+                  player:,
+                  minute:,
+                  offset:  nil,
+                  owngoal: false,
+                  penalty: false,
+                  notes:   nil
+                )
+    @team1   = team1
+    @team2   = team2
+    @round   = round
+    @date    = date
+
+    @score1  = score1
+    @score2  = score2
+    @player  = player
+    @minute  = minute
+    @offset  = offset
+    @owngoal = owngoal
+    @penalty = penalty
+    @notes   = notes
+  end
+
+
+  ## note: lets you use normalize teams or such acts like a Match struct
+  def update( **kwargs )
+    ## todo/fix: use team1_name, team2_name or similar - for compat with db activerecord version? why? why not?
+    @team1 = kwargs[:team1]    if kwargs.has_key? :team1
+    @team2 = kwargs[:team2]    if kwargs.has_key? :team2
+  end
+end  # class GoalEvent
+
+
+### extend "basic" goal struct with goal event build
+class Goal  ### nested (non-freestanding) inside match (match is parent)
+
+  def self.build( events )  ## check/todo - rename to build_from_event/row or such - why? why not?
+    ## build an array of goal structs from (csv) recs
+    recs = []
+
+    last_score1 = 0
+    last_score2 = 0
+
+    events.each do |event|
+
+      if last_score1+1 == event.score1 && last_score2 == event.score2
+        team = 1
+      elsif last_score2+1 == event.score2 && last_score1 == event.score1
+        team = 2
+      else
+        puts "!! ERROR - unexpected score advance (one goal at a time expected):"
+        puts "  #{last_score1}-#{last_score2}=> #{event.score1}-#{event.score2}"
+        exit 1
+      end
+
+      last_score1 = event.score1
+      last_score2 = event.score2
+
+
+      attributes = {
+        score1:  event.score1,
+        score2:  event.score2,
+        team:    team,
+        minute:  event.minute,
+        offset:  event.offset,
+        player:  event.player,
+        owngoal: event.owngoal,
+        penalty: event.penalty,
+        notes:   event.notes
+      }
+
+      recs << new( **attributes )
+    end
+
+    recs
+  end
+end  # class Goal
+
+
+end # module Sports
+
+
+

data/lib/sportdb/quick/csv/goal_parser_csv.rb

@@ -0,0 +1,28 @@
+
+module SportDb
+  class CsvGoalParser
+
+
+  def self.read( path )
+     txt = File.open( path, 'r:utf-8' ) {|f| f.read }   ## note: make sure to use (assume) utf-8
+     parse( txt )
+  end
+
+  def self.parse( txt )
+     new( txt ).parse
+  end
+
+
+  def initialize( txt )
+    @txt = txt
+  end
+
+  def parse
+    rows = parse_csv( @txt )
+    recs = rows.map { |row| Sports::GoalEvent.build( row ) }
+    ## pp recs[0]
+    recs
+  end
+
+  end # class CsvGoalParser
+end # module Sports

data/lib/sportdb/quick/csv/match_parser_csv.rb

@@ -0,0 +1,490 @@
+
+module SportDb
+  class CsvMatchParser
+
+  #############
+  # helpers
+  def self.find_seasons( path, col: 'Season', sep: nil, headers: nil )
+
+    ## check if headers incl. season if yes,has priority over col mapping
+    ##  e.g. no need to specify twice (if using headers)
+    col = headers[:season]    if headers && headers[:season]
+
+    seasons = Hash.new( 0 )   ## default value is 0
+
+    ## todo/fix: yes, use CsvHash.foreach  - why? why not?
+    ##                    use read_csv with block  to switch to foreach!!!!
+    rows = read_csv( path, sep: sep )
+
+    rows.each_with_index do |row,i|
+      puts "[#{i}] " + row.inspect  if i < 2
+
+      season = row[ col ]   ## column name defaults to 'Season'
+      seasons[ season ] += 1
+    end
+
+    pp seasons
+
+    ## note: only return season keys/names (not hash with usage counter)
+    seasons.keys
+  end
+
+
+  ##########
+  # main machinery
+
+  ##  todo/fix: use a generic "global" parse_csv method - why? why not?
+  ## def self.parse_csv( text, sep: ',' )    ## helper -lets you change the csv library in one place if needed/desired
+  ##   ## note:  do NOT symbolize keys - keep them as is!!!!!!
+  ##  ##   todo/fix: move "upstream" and remove symbolize keys too!!! - why? why not?
+  ##   CsvHash.parse( text, sep: sep )
+  ## end
+
+  def self.read( path, headers: nil, filters: nil, converters: nil, sep: nil )
+     txt = File.open( path, 'r:utf-8' ) {|f| f.read }   ## note: make sure to use (assume) utf-8
+     parse( txt, headers: headers,
+                 filters: filters,
+                 converters: converters,
+                 sep: sep )
+  end
+
+  def self.parse( txt, headers: nil, filters: nil, converters: nil, sep: nil )
+     new( txt ).parse( headers: headers,
+                        filters: filters,
+                        converters: converters,
+                        sep: sep )
+  end
+
+
+  def initialize( txt )
+    @txt = txt
+  end
+
+  def parse( headers: nil, filters: nil, converters: nil, sep: nil )
+
+    headers_mapping = {}
+
+    rows = parse_csv( @txt, sep: sep )
+
+    return []   if rows.empty?      ## no rows / empty?
+
+
+    ## fix/todo: use logger!!!!
+    ## pp csv
+
+    if headers   ## use user supplied headers if present
+      headers_mapping = headers_mapping.merge( headers )
+    else
+
+      ## note: returns an array of strings (header names)  - assume all rows have the same columns/fields!!!
+      headers = rows[0].keys
+      pp headers
+
+      # note: greece 2001-02 etc. use HT  -  check CVS reader  row['HomeTeam'] may not be nil but an empty string?
+      #   e.g. row['HomeTeam'] || row['HT'] will NOT work for now
+
+      if find_header( headers, ['Team 1']) && find_header( headers, ['Team 2'])
+         ## assume our own football.csv format, see github.com/footballcsv
+         headers_mapping[:team1]  = find_header( headers, ['Team 1'] )
+         headers_mapping[:team2]  = find_header( headers, ['Team 2'] )
+         headers_mapping[:date]   = find_header( headers, ['Date'] )
+         headers_mapping[:time]   = find_header( headers, ['Time'] )
+
+         ## check for all-in-one full time (ft) and half time (ht9 scores?
+         headers_mapping[:score]  = find_header( headers, ['FT'] )
+         headers_mapping[:scorei] = find_header( headers, ['HT'] )
+
+         headers_mapping[:round]  = find_header( headers, ['Round', 'Matchday'] )
+
+         ## optional headers - note: find_header returns nil if header NOT found
+         header_stage = find_header( headers, ['Stage'] )
+         headers_mapping[:stage]  =  header_stage   if header_stage
+
+         header_group = find_header( headers, ['Group'] )
+         headers_mapping[:group]  =  header_group   if header_group
+
+
+         header_et = find_header( headers, ['ET', 'AET'] )   ## (after) extra time
+         headers_mapping[:score_et] = header_et   if header_et
+
+         header_p  = find_header( headers, ['P', 'PEN'] )    ## penalties
+         headers_mapping[:score_p] = header_p   if header_p
+
+         header_notes = find_header( headers, ['Notes', 'Comments'] )
+         headers_mapping[:notes]  =  header_notes   if header_notes
+
+
+         header_league = find_header( headers, ['League'] )
+         headers_mapping[:league] = header_league   if header_league
+      else
+         ## else try footballdata.uk and others
+         headers_mapping[:team1]  = find_header( headers, ['HomeTeam', 'HT', 'Home'] )
+         headers_mapping[:team2]  = find_header( headers, ['AwayTeam', 'AT', 'Away'] )
+         headers_mapping[:date]   = find_header( headers, ['Date'] )
+         headers_mapping[:time]   = find_header( headers, ['Time'] )
+
+         ## note: FT = Full Time, HG = Home Goal, AG = Away Goal
+         headers_mapping[:score1] = find_header( headers, ['FTHG', 'HG'] )
+         headers_mapping[:score2] = find_header( headers, ['FTAG', 'AG'] )
+
+         ## check for half time scores ?
+         ##  note: HT = Half Time
+         headers_mapping[:score1i] = find_header( headers, ['HTHG'] )
+         headers_mapping[:score2i] = find_header( headers, ['HTAG'] )
+      end
+    end
+
+    pp headers_mapping
+
+    ### todo/fix: check headers - how?
+    ##  if present HomeTeam or HT required etc.
+    ##   issue error/warn is not present
+    ##
+    ## puts "*** !!! wrong (unknown) headers format; cannot continue; fix it; sorry"
+    ##    exit 1
+    ##
+
+    matches = []
+
+    rows.each_with_index do |row,i|
+
+      ## fix/todo: use logger!!!!
+      ## puts "[#{i}] " + row.inspect  if i < 2
+
+
+      ## todo/fix: move to its own (helper) method - filter or such!!!!
+       if filters    ## filter MUST match if present e.g. row['Season'] == '2017/2018'
+         skip = false
+         filters.each do |header, value|
+           if row[ header ] != value   ## e.g. row['Season']
+             skip = true
+             break
+           end
+         end
+         next if skip   ## if header values NOT matching
+       end
+
+
+      ## note:
+      ##   add converters after filters for now (why not before filters?)
+      if converters   ## any converters defined?
+        ## convert single proc shortcut to array with single converter
+        converters = [converters]    if converters.is_a?( Proc )
+
+        ## assumes array of procs
+        converters.each do |converter|
+          row = converter.call( row )
+        end
+      end
+
+
+
+      team1 = row[ headers_mapping[ :team1 ]]
+      team2 = row[ headers_mapping[ :team2 ]]
+
+
+      ## check if data present - if not skip (might be empty row)
+      ##  note:  (old classic) csv reader returns nil for empty fields
+      ##         new modern csv reader ALWAYS returns strings (and empty strings for data not available (n/a))
+      if (team1.nil? || team1.empty?) &&
+         (team2.nil? || team2.empty?)
+        puts "*** WARN: skipping empty? row[#{i}] - no teams found:"
+        pp row
+        next
+      end
+
+      ## remove possible match played counters e.g. (4) (11) etc.
+      team1 = team1.sub( /\(\d+\)/, '' ).strip
+      team2 = team2.sub( /\(\d+\)/, '' ).strip
+
+
+
+      col = row[ headers_mapping[ :time ]]
+
+      if col.nil?
+        time = nil
+      else
+        col = col.strip     # make sure not leading or trailing spaces left over
+
+        if col.empty?
+          col =~ /^-{1,}$/ ||      # e.g.  - or ---
+          col =~ /^\?{1,}$/        # e.g. ? or ???
+          ## note: allow missing / unknown date for match
+          time = nil
+        else
+          if col =~ /^\d{1,2}:\d{2}$/
+            time_fmt = '%H:%M'   # e.g. 17:00 or 3:00
+          elsif col =~ /^\d{1,2}.\d{2}$/
+            time_fmt = '%H.%M'   # e.g. 17:00 or 3:00
+          else
+            puts "*** !!! wrong (unknown) time format >>#{col}<<; cannot continue; fix it; sorry"
+            ## todo/fix: add to errors/warns list - why? why not?
+            exit 1
+          end
+
+          ## todo/check: use date object (keep string?) - why? why not?
+          ##  todo/fix: yes!! use date object!!!! do NOT use string
+          time = Time.strptime( col, time_fmt ).strftime( '%H:%M' )
+        end
+      end
+
+
+
+      col = row[ headers_mapping[ :date ]]
+      col = col.strip   # make sure not leading or trailing spaces left over
+
+      if col.empty? ||
+         col =~ /^-{1,}$/ ||      # e.g.  - or ---
+         col =~ /^\?{1,}$/        # e.g. ? or ???
+          ## note: allow missing / unknown date for match
+          date = nil
+      else
+        ## remove possible weekday or weeknumber  e.g. (Fri) (4) etc.
+        col = col.sub( /\(W?\d{1,2}\)/, '' )  ## e.g. (W11), (4), (21) etc.
+        col = col.sub( /\(\w+\)/, '' )  ## e.g. (Fri), (Fr) etc.
+        col = col.strip   # make sure not leading or trailing spaces left over
+
+        if col =~ /^\d{2}\/\d{2}\/\d{4}$/
+          date_fmt = '%d/%m/%Y'   # e.g. 17/08/2002
+        elsif col =~ /^\d{2}\/\d{2}\/\d{2}$/
+          date_fmt = '%d/%m/%y'   # e.g. 17/08/02
+        elsif col =~ /^\d{4}-\d{1,2}-\d{1,2}$/      ## "standard" / default date format
+          date_fmt = '%Y-%m-%d'   # e.g. 1995-08-04
+        elsif col =~ /^\d{1,2} \w{3} \d{4}$/
+          date_fmt = '%d %b %Y'   # e.g. 8 Jul 2017
+        elsif col =~ /^\w{3} \w{3} \d{1,2} \d{4}$/
+          date_fmt = '%a %b %d %Y'   # e.g. Sat Aug 7 1993
+        else
+          puts "*** !!! wrong (unknown) date format >>#{col}<<; cannot continue; fix it; sorry"
+          ## todo/fix: add to errors/warns list - why? why not?
+          exit 1
+        end
+
+        ## todo/check: use date object (keep string?) - why? why not?
+        ##  todo/fix: yes!! use date object!!!! do NOT use string
+        date = Date.strptime( col, date_fmt ).strftime( '%Y-%m-%d' )
+      end
+
+
+      ##
+      ## todo/fix:  round might not always be just a simple integer number!!!
+      ##             might be text such as Final | Leg 1 or such!!!!
+      round   = nil
+      ## check for (optional) round / matchday
+      if headers_mapping[ :round ]
+        col = row[ headers_mapping[ :round ]]
+        ## todo: issue warning if not ? or - (and just empty string) why? why not
+        ## (old attic) was: round = col.to_i  if col =~ /^\d{1,2}$/     # check format - e.g. ignore ? or - or such non-numbers for now
+
+        ## note: make round always a string for now!!!! e.g. "1", "2" too!!
+        round = if col.nil? || col.empty? || col == '-' || col == 'n/a'
+                 ## note: allow missing round for match / defaults to nil
+                 nil
+                else
+                  col
+                end
+      end
+
+
+      score1  = nil
+      score2  = nil
+      score1i = nil
+      score2i = nil
+
+      ## check for full time scores ?
+      if headers_mapping[ :score1 ] && headers_mapping[ :score2 ]
+        ft = [ row[ headers_mapping[ :score1 ]],
+               row[ headers_mapping[ :score2 ]] ]
+
+        ## todo/fix: issue warning if not ? or - (and just empty string) why? why not
+        score1 = ft[0].to_i  if ft[0] =~ /^\d{1,2}$/
+        score2 = ft[1].to_i  if ft[1] =~ /^\d{1,2}$/
+      end
+
+      ## check for half time scores ?
+      if headers_mapping[ :score1i ] && headers_mapping[ :score2i ]
+        ht = [ row[ headers_mapping[ :score1i ]],
+               row[ headers_mapping[ :score2i ]] ]
+
+        ## todo/fix: issue warning if not ? or - (and just empty string) why? why not
+        score1i = ht[0].to_i  if ht[0] =~ /^\d{1,2}$/
+        score2i = ht[1].to_i  if ht[1] =~ /^\d{1,2}$/
+      end
+
+
+      ## check for all-in-one full time scores?
+      if headers_mapping[ :score ]
+        col = row[ headers_mapping[ :score ]]
+        score = parse_score( col )
+        if score
+          score1 = score[0]
+          score2 = score[1]
+        else
+          puts "!! ERROR - invalid score (ft) format >#{col}<:"
+          pp row
+          exit 1
+        end
+      end
+
+      if headers_mapping[ :scorei ]
+        col = row[ headers_mapping[ :scorei ]]
+        score = parse_score( col )
+        if score
+          score1i = score[0]
+          score2i = score[1]
+        else
+          puts "!! ERROR - invalid score (ht) format >#{col}<:"
+          pp row
+          exit 1
+        end
+      end
+
+      ####
+      ## try optional score - extra time (et) and penalities (p/pen)
+      score1et  = nil
+      score2et  = nil
+      score1p   = nil
+      score2p   = nil
+
+      if headers_mapping[ :score_et ]
+        col = row[ headers_mapping[ :score_et ]]
+        score = parse_score( col )
+        if score
+          score1et = score[0]
+          score2et = score[1]
+        else
+          puts "!! ERROR - invalid score (et) format >#{col}<:"
+          pp row
+          exit 1
+        end
+      end
+
+      if headers_mapping[ :score_p ]
+        col = row[ headers_mapping[ :score_p ]]
+        score = parse_score( col )
+        if score
+          score1p = score[0]
+          score2p = score[1]
+        else
+          puts "!! ERROR - invalid score (p) format >#{col}<:"
+          pp row
+          exit 1
+        end
+      end
+
+
+      ## try some optional headings / columns
+      stage = nil
+      if headers_mapping[ :stage ]
+        col = row[ headers_mapping[ :stage ]]
+        ## todo/fix: check can col be nil e.g. col.nil? possible?
+        stage =  if col.nil? || col.empty? || col == '-' || col == 'n/a'
+                    ## note: allow missing stage for match / defaults to "regular"
+                    nil
+                 elsif col == '?'
+                     ## note: default explicit unknown to unknown for now AND not regular - why? why not?
+                    '?'   ## todo/check: use unkown and NOT ?  - why? why not?
+                 else
+                    col
+                 end
+      end
+
+      group = nil
+      if headers_mapping[ :group ]
+        col = row[ headers_mapping[ :group ]]
+        ## todo/fix: check can col be nil e.g. col.nil? possible?
+        group =  if col.nil? || col.empty? || col == '-' || col == 'n/a'
+                    ## note: allow missing stage for match / defaults to "regular"
+                    nil
+                 else
+                    col
+                 end
+      end
+
+      status = nil    ## e.g. AWARDED, CANCELLED, POSTPONED, etc.
+      if headers_mapping[ :notes ]
+        col = row[ headers_mapping[ :notes ]]
+        ## check for optional (match) status in notes / comments
+        status = if col.nil? || col.empty? || col == '-' || col == 'n/a'
+                   nil
+                 else
+                   StatusParser.parse( col )  # note: returns nil if no (match) status found
+                 end
+      end
+
+
+      league = nil
+      league = row[ headers_mapping[ :league ]]   if headers_mapping[ :league ]
+
+
+      ## puts 'match attributes:'
+      attributes = {
+        date:     date,
+        time:     time,
+        team1:    team1,    team2:    team2,
+        score1:   score1,   score2:   score2,
+        score1i:  score1i,  score2i:  score2i,
+        score1et: score1et, score2et: score2et,
+        score1p:  score1p,  score2p:  score2p,
+        round:    round,
+        stage:    stage,
+        group:    group,
+        status:   status,
+        league:   league
+      }
+      ## pp attributes
+
+      match = Sports::Match.new( **attributes )
+      matches << match
+    end
+
+    ## pp matches
+    matches
+  end
+
+
+  private
+
+  def find_header( headers, candidates )
+     ## todo/fix: use find_first from enumare of similar ?! - why? more idiomatic code?
+
+    candidates.each do |candidate|
+       return candidate   if headers.include?( candidate ) ## bingo!!!
+    end
+    nil  ## no matching header  found!!!
+  end
+
+########
+# more helpers
+#
+
+def parse_score( str )
+  if str.nil?    ## todo/check: remove nil case - possible? - why? why not?
+    [nil,nil]
+  else
+    ## remove (optional single) note/footnote/endnote markers
+    ##  e.g. (*) or (a), (b),
+    ##    or [*], [A], [1], etc.
+    ##  - allow (1) or maybe (*1) in the future - why? why not?
+    str = str.sub( /\( [a-z*] \)
+                        |
+                    \[ [1-9a-z*] \]
+                   /ix, '' ).strip
+
+    if str.empty? || str == '?' || str == '-' || str == 'n/a'
+      [nil,nil]
+    ### todo/check: use regex with named capture groups here - why? why not?
+    elsif str =~ /^\d{1,2}[:-]\d{1,2}$/   ## sanity check scores format
+      score  = str.split( /[:-]/ )
+      [score[0].to_i, score[1].to_i]
+    else
+      nil  ## note: returns nil if invalid / unparseable format!!!
+    end
+  end
+end # method parse_score
+
+
+
+  end # class CsvMatchParser
+end # module Sports
+

data/lib/sportdb/quick/csv/match_status_parser.rb

@@ -0,0 +1,90 @@
+#####################
+# helpers for parsing & finding match status e.g.
+#    - cancelled / canceled
+#    - awarded
+#    - abandoned
+#    - replay
+#    etc.
+
+
+module SportDb
+
+
+  ### todo/fix: move Status inside Match struct - why? why not?
+
+  class Status
+# note: use a class as an "enum"-like namespace for now - why? why not?
+#   move class into Match e.g. Match::Status  - why? why not?
+    CANCELLED = 'CANCELLED'   # canceled (US spelling), cancelled (UK spelling) - what to use?
+    AWARDED   = 'AWARDED'
+    POSTPONED = 'POSTPONED'
+    ABANDONED = 'ABANDONED'
+    REPLAY    = 'REPLAY'
+  end # class Status
+
+
+
+  class StatusParser
+
+    def self.parse( str )
+      ## note: returns nil if no match found
+      ## note: english usage - cancelled (in UK), canceled (in US)
+      if str =~ /^(cancelled|
+                   canceled|
+                   can\.
+                  )/xi
+        Status::CANCELLED
+      elsif str =~ /^(awarded|
+                       awd\.
+                      )/xi
+        Status::AWARDED
+      elsif str =~ /^(postponed
+                      )/xi
+        Status::POSTPONED
+      elsif str =~ /^(abandoned|
+                       abd\.
+                      )/xi
+        Status::ABANDONED
+      elsif str =~ /^(replay
+                      )/xi
+        Status::REPLAY
+      else
+        # no match
+        nil
+      end
+    end
+
+
+    RUN_RE = /\[
+                 (?<text>[^\]]+)
+               \]
+             /x
+    def self.find!( line )
+      ## for now check all "protected" text run blocks e.g. []
+      ##  puts "line: >#{line}<"
+
+      status = nil
+
+      str = line
+      while m = str.match( RUN_RE )
+        str = m.post_match  ## keep on processing rest of line/str (a.k.a. post match string)
+
+        ## check for status match
+        match_str = m[0]  ## keep a copy of the match string (for later sub)
+        text = m[:text].strip
+        ## puts "  text: >#{text}<"
+
+        status = parse( text )
+
+        if status
+           line.sub!( match_str, "[STATUS.#{status}]" )
+           break
+        end
+      end  # while match
+
+      status
+    end # method find!
+  end # class StatusParser
+
+end # module SportDb
+

data/lib/sportdb/quick/quick_league_outline_reader.rb

@@ -41,6 +41,15 @@ class QuickLeagueOutlineReader
           pp heading
           exit 1
         end
+      elsif node[0] == :h2
+         ## todo/check - make sure parsed h1 first
+         heading = node[1]
+         ## reuse league, season from h1
+         secs << { league: secs[-1][:league],
+                   season: secs[-1][:season],
+                   stage:  heading,
+                   lines:  []
+                 }
       elsif node[0] == :p   ## paragraph with (text) lines
         lines = node[1]
         ## note: skip lines if no heading seen

data/lib/sportdb/quick/version.rb

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

data/lib/sportdb/quick.rb

@@ -17,8 +17,10 @@ end
 ## our own code
 require_relative 'quick/version'
 require_relative 'quick/opts'
-require_relative 'quick/linter'
-require_relative 'quick/outline_reader'
+
+# require_relative 'quick/linter'
+# require_relative 'quick/outline_reader'
+
 
 require_relative 'quick/match_parser'
 
@@ -26,6 +28,39 @@ require_relative 'quick/quick_league_outline_reader'
 require_relative 'quick/quick_match_reader'
 
 
+
+
+###
+#  csv (tabular dataset) support / machinery
+require_relative 'quick/csv/match_status_parser'
+require_relative 'quick/csv/goal'
+require_relative 'quick/csv/goal_parser_csv'
+require_relative 'quick/csv/match_parser_csv'
+
+
+### add convenience shortcut helpers
+module Sports
+  class Match
+    def self.read_csv( path, headers: nil, filters: nil, converters: nil, sep: nil )
+       SportDb::CsvMatchParser.read( path,
+                                       headers:    headers,
+                                       filters:    filters,
+                                       converters: converters,
+                                       sep:        sep )
+    end
+
+    def self.parse_csv( txt, headers: nil, filters: nil, converters: nil, sep: nil )
+       SportDb::CsvMatchParser.parse( txt,
+                                        headers:    headers,
+                                        filters:    filters,
+                                        converters: converters,
+                                        sep:        sep )
+    end
+  end # class Match
+end # module Sports
+
+
+
 puts SportDb::Module::Quick.banner    # say hello
 
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: sportdb-quick
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.1.1
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-08-27 00:00:00.000000000 Z
+date: 2024-09-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: sportdb-parser
@@ -86,7 +86,7 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '4.1'
-description: sportdb-quick - football.txt (quick) match parsers and more
+description: sportdb-quick - football.txt (quick) match readers and more
 email: gerald.bauer@gmail.com
 executables:
 - fbt
@@ -102,10 +102,12 @@ files:
 - Rakefile
 - bin/fbt
 - lib/sportdb/quick.rb
-- lib/sportdb/quick/linter.rb
+- lib/sportdb/quick/csv/goal.rb
+- lib/sportdb/quick/csv/goal_parser_csv.rb
+- lib/sportdb/quick/csv/match_parser_csv.rb
+- lib/sportdb/quick/csv/match_status_parser.rb
 - lib/sportdb/quick/match_parser.rb
 - lib/sportdb/quick/opts.rb
-- lib/sportdb/quick/outline_reader.rb
 - lib/sportdb/quick/quick_league_outline_reader.rb
 - lib/sportdb/quick/quick_match_reader.rb
 - lib/sportdb/quick/version.rb
@@ -133,5 +135,5 @@ requirements: []
 rubygems_version: 3.4.10
 signing_key:
 specification_version: 4
-summary: sportdb-quick - football.txt (quick) match parsers and more
+summary: sportdb-quick - football.txt (quick) match readers and more
 test_files: []

data/lib/sportdb/quick/linter.rb

@@ -1,149 +0,0 @@
-
-module SportDb
-module Quick
-
-###
-## note - Linter for now nested inside Parser - keep? why? why not?
-class Linter
-
-def self.debug=(value) @@debug = value; end
-def self.debug?() @@debug ||= false; end  ## note: default is FALSE
-def debug?()  self.class.debug?; end
-
-
-
-attr_reader :errors
-
-def initialize
-  @errors = []
-  @parser = Parser.new   ## use own parser instance (not shared) - why? why not?
-end
-
-
-def errors?() @errors.size > 0; end
-
-
-
-  ## note:  colon (:) MUST be followed by one (or more) spaces
-  ##      make sure mon feb 12 18:10 will not match
-  ##        allow 1. FC Köln etc.
-  ##               Mainz 05:
-  ##           limit to 30 chars max
-  ##          only allow  chars incl. intl buut (NOT ()[]/;)
-  ##
-  ##   Group A:
-  ##   Group B:   - remove colon
-  ##    or lookup first
-
-  ATTRIB_RE = %r{^
-                   [ ]*?     # slurp leading spaces
-                (?<key>[^:|\]\[()\/; -]
-                       [^:|\]\[()\/;]{0,30}
-                 )
-                   [ ]*?     # slurp trailing spaces
-                   :[ ]+
-                (?<value>.+)
-                    [ ]*?   # slurp trailing spaces
-                   $
-                }ix
-
-
-#########
-## parse - false (default) - tokenize (only)
-##       - true            - tokenize & parse
-def read( path, parse: false )
-  ## note: every (new) read call - resets errors list to empty
-  @errors = []
-
-  nodes = OutlineReader.read( path )
-
-  ##  process nodes
-  h1 = nil
-  orphans = 0    ## track paragraphs's with no heading
-
-  attrib_found = false
-
-
-  nodes.each do |node|
-    type = node[0]
-
-    if type == :h1
-        h1 = node[1]  ## get heading text
-        puts
-        puts "  = Heading 1 >#{node[1]}<"
-    elsif type == :p
-
-       if h1.nil?
-         orphans += 1    ## only warn once
-         puts "!! WARN - no heading for #{orphans} text paragraph(s); skipping parse"
-         next
-       end
-
-       lines = node[1]
-
-       tree = []
-       lines.each_with_index do |line,i|
-
-        if debug?
-         puts
-         puts "line >#{line}<"
-        end
-
-
-        ## skip new (experimental attrib syntax)
-        if attrib_found == false &&
-            ATTRIB_RE.match?( line )
-          ## note: check attrib regex AFTER group def e.g.:
-          ##         Group A:
-          ##         Group B:  etc.
-          ##     todo/fix - change Group A: to Group A etc.
-          ##                       Group B: to Group B
-           attrib_found = true
-           ## logger.debug "skipping key/value line - >#{line}<"
-           next
-        end
-
-        if attrib_found
-          ## check if line ends with dot
-          ##  if not slurp up lines to the next do!!!
-          ## logger.debug "skipping key/value line - >#{line}<"
-          attrib_found = false   if line.end_with?( '.' )
-              # logger.debug "skipping key/value line (cont.) - >#{line}<"
-              next
-        end
-
-        t, error_messages  =  if parse
-                                  @parser.parse_with_errors( line )
-                              else
-                                  @parser.tokenize_with_errors( line )
-                              end
-
-
-         if error_messages.size > 0
-            ## add to "global" error list
-            ##   make a triplet tuple (file / msg / line text)
-            error_messages.each do |msg|
-                @errors << [ path,
-                             msg,
-                             line
-                           ]
-            end
-         end
-
-         pp t   if debug?
-
-         tree << t
-       end
-
-       ## pp tree
-    else
-        pp node
-        raise ArgumentError, "unsupported (node) type >#{type}<"
-    end
-  end  # each node
-end  # read
-end  # class Linter
-
-
-end   # module Quick
-end   # module SportDb

data/lib/sportdb/quick/outline_reader.rb

@@ -1,97 +0,0 @@
-
-
-module SportDb
-
-class OutlineReader
-
-  def self.debug=(value) @@debug = value; end
-  def self.debug?() @@debug ||= false; end
-  def debug?()  self.class.debug?; end
-
-
-
-  def self.read( path )   ## use - rename to read_file or from_file etc. - why? why not?
-    txt = File.open( path, 'r:utf-8' ) {|f| f.read }
-    parse( txt )
-  end
-
-  def self.parse( txt )
-    new( txt ).parse
-  end
-
-  def initialize( txt )
-    @txt = txt
-  end
-
-  ## note: skip "decorative" only heading e.g. ========
-  ##  todo/check:  find a better name e.g. HEADING_EMPTY_RE or HEADING_LINE_RE or ???
-  HEADING_BLANK_RE = %r{\A
-                        ={1,}
-                        \z}x
-
-  ## note: like in wikimedia markup (and markdown) all optional trailing ==== too
-  HEADING_RE = %r{\A
-                  (?<marker>={1,})       ## 1. leading ======
-                    [ ]*
-                  (?<text>[^=]+)         ## 2. text   (note: for now no "inline" = allowed)
-                    [ ]*
-                    =*                   ## 3. (optional) trailing ====
-                  \z}x
-
-  def parse
-    outline=[]   ## outline structure
-    start_para = true      ## start new para(graph) on new text line?
-
-    @txt.each_line do |line|
-        line = line.strip      ## todo/fix: keep leading and trailing spaces - why? why not?
-
-        if line.empty?    ## todo/fix: keep blank line nodes?? and just remove comments and process headings?! - why? why not?
-          start_para = true
-          next
-        end
-
-        break if line == '__END__'
-
-        next if line.start_with?( '#' )   ## skip comments too
-        ## strip inline (until end-of-line) comments too
-        ##  e.g Eupen | KAS Eupen ## [de]
-        ##   => Eupen | KAS Eupen
-        ##  e.g bq   Bonaire,  BOE        # CONCACAF
-        ##   => bq   Bonaire,  BOE
-        line = line.sub( /#.*/, '' ).strip
-        pp line    if debug?
-
-        ## todo/check: also use heading blank as paragraph "breaker" or treat it like a comment ?? - why? why not?
-        next if HEADING_BLANK_RE.match( line )  # skip "decorative" only heading e.g. ========
-
-         ## note: like in wikimedia markup (and markdown) all optional trailing ==== too
-        if m=HEADING_RE.match( line )
-           start_para = true
-
-           heading_marker = m[:marker]
-           heading_level  = heading_marker.length   ## count number of = for heading level
-           heading        = m[:text].strip
-
-           puts "heading #{heading_level} >#{heading}<"   if debug?
-           outline << [:"h#{heading_level}", heading]
-        else    ## assume it's a (plain/regular) text line
-           if start_para
-             outline << [:p, [line]]
-             start_para = false
-           else
-             node = outline[-1]    ## get last entry
-             if node[0] == :p      ##  assert it's a p(aragraph) node!!!
-                node[1] << line    ## add line to p(aragraph)
-             else
-               puts "!! ERROR - invalid outline state / format - expected p(aragraph) node; got:"
-               pp node
-               exit 1
-             end
-           end
-        end
-    end
-    outline
-  end # method read
-end # class OutlineReader
-
-end # module SportDb