sportdb-formats 1.0.3 → 1.1.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 4a27c364d2003ece8da886c892d34d595ffcffcc
-  data.tar.gz: 81530dfec5decf5d9476a3f2be90bcbeb0195824
+  metadata.gz: 4cdd2bc410771494ed506a24d384ca3c8b1c9684
+  data.tar.gz: 066f5288da503a00efe280369f57d6cd65bf4bf7
 SHA512:
-  metadata.gz: 4a2aabb65968939d55f2000c64e4ac5df4cbedfe6b3786fd12ff86d26bd5e6af12c32ea129b59a7ddf30736afa6f093e0e4f2e97f2c6f0b23bffd6901a83ca91
-  data.tar.gz: 63f4ca42ebb537173334a931a02331b8cbdc27d9e953052c8126d9e683e54a3103eb496103fea400283c6b0b08fa7c6e738466f215e26f9567a3b822a80c172d
+  metadata.gz: f61edee9495047fc49dfb5720c7afc1e0e316e5fab024d8fc0b1bd5fcdad70524f6e22d24d8fad0aa679380565e0a0c7e36fae79c170655f8a7a496dee170aca
+  data.tar.gz: b378202d2c8152ac46386618d3c69c03d802b1439e74175f8d84fad304ce111edb0bb1cd848fc969848ce55e0cf8de15a46bec322351d5d199652ed16d41b164

data/Manifest.txt

@@ -63,6 +63,7 @@ test/test_match.rb
 test/test_match_auto.rb
 test/test_match_auto_champs.rb
 test/test_match_auto_euro.rb
+test/test_match_auto_relegation.rb
 test/test_match_auto_worldcup.rb
 test/test_match_champs.rb
 test/test_match_eng.rb

data/lib/sportdb/formats/country/country_index.rb

@@ -30,11 +30,11 @@ class CountryIndex
     ## auto-fill countries
     ## pp recs
     recs.each do |rec|
-      ## rec e.g. { key:'af', fifa:'AFG', name:'Afghanistan'}
+      ## rec e.g. { key:'af', code:'AFG', name:'Afghanistan'}
 
       @countries << rec
 
-      ## add codes lookups - key, fifa, ...
+      ## add codes lookups - key, code, ...
       if @countries_by_code[ rec.key ]
         puts "** !! ERROR !! country code (key) >#{rec.key}< already exits!!"
         exit 1
@@ -42,13 +42,13 @@ class CountryIndex
         @countries_by_code[ rec.key ] = rec
       end
 
-      ## add fifa code (only) if different from key
-      if rec.key != rec.fifa.downcase
-        if @countries_by_code[ rec.fifa.downcase ]
-          puts "** !! ERROR !! country code (fifa) >#{rec.fifa.downcase}< already exits!!"
+      ## add  code (only) if different from key
+      if rec.key != rec.code.downcase
+        if @countries_by_code[ rec.code.downcase ]
+          puts "** !! ERROR !! country code  >#{rec.code.downcase}< already exits!!"
           exit 1
         else
-          @countries_by_code[ rec.fifa.downcase ] = rec
+          @countries_by_code[ rec.code.downcase ] = rec
         end
       end
 

data/lib/sportdb/formats/country/country_reader.rb

@@ -43,11 +43,31 @@ def parse
         #   e.g. East Germany        (-1989)  => East Germany (-1989)
         values = values.map { |value| value.strip.gsub( /[ \t]+/, ' ' ) }
         last_country.alt_names += values
+      elsif line =~ /^-[ ]*(\d{4})
+                        [ ]+
+                       (.+)$
+                    /x     ## check for historic lines e.g. -1989
+         year   = $1.to_i
+         parts  = $2.split( /=>|⇒/ )
+         values = parts[0].split( ',' )
+         values = values.map { |value| value.strip.gsub( /[ \t]+/, ' ' ) }
+
+         name = values[0]
+         code = values[1]
+
+         last_country = country = Country.new( name: "#{name} (-#{year})",
+                                               code: code )
+         ## country.alt_names << name    ## note: for now do NOT add name without year to alt_names - gets auto-add by index!!!
+
+         countries << country
+         ## todo/fix: add reference to country today (in parts[1] !!!!)
       else
         ## assume "regular" line
         ##  check if starts with id  (todo/check: use a more "strict"/better regex capture pattern!!!)
         ##   note: allow country codes upto 4 (!!) e.g. Northern Cyprus
-        if line =~ /^([a-z]{2,4})[ ]+(.+)$/
+        if line =~ /^([a-z]{2,4})
+                        [ ]+
+                       (.+)$/x
           key    = $1
           values = $2.split( ',' )
           ## strip and squish (white)spaces
@@ -60,14 +80,14 @@ def parse
           geos = split_geo( values[0] )
           name = geos[0]    ## note: ignore all other geos for now
 
-          ##   note: allow fifa country codes upto 4 (!!) e.g. Northern Cyprus
-          fifa = if values[1] && values[1] =~ /^[A-Z]{3,4}$/   ## note: also check format
+          ##   note: allow country codes up to 4 (!!) e.g. Northern Cyprus
+          code = if values[1] && values[1] =~ /^[A-Z]{3,4}$/   ## note: also check format
                    values[1]
                  else
                    if values[1]
-                     puts "** !!! ERROR !!! wrong fifa code format >#{values[1]}<; expected three (or four)-letter all up-case"
+                     puts "** !!! ERROR !!! wrong code format >#{values[1]}<; expected three (or four)-letter all up-case"
                    else
-                     puts "** !!! ERROR !!! missing fifa code for (canonical) country name"
+                     puts "** !!! ERROR !!! missing code for (canonical) country name"
                    end
                    exit 1
                  end
@@ -80,7 +100,7 @@ def parse
 
           last_country = country = Country.new( key: key,
                                                 name: name,
-                                                fifa: fifa,
+                                                code: code,
                                                 tags: tags )
           countries << country
         else

data/lib/sportdb/formats/league/league_outline_reader.rb

@@ -121,14 +121,31 @@ class LeagueOutlineReader   ## todo/check - rename to LeaguePageReader / LeagueP
     values
   end
 
-  def check_stage( name )
-    known_stages = ['regular season',
-                    'championship round',
-                    'relegation round',
-                    'play-offs'
-                   ]
 
-    if known_stages.include?( name.downcase )
+  # note: normalize names e.g. downcase and remove all non a-z chars (e.g. space, dash, etc.)
+  KNOWN_STAGES = [
+    'Regular Season',
+    'Regular Stage',
+    'Championship Round',
+    'Championship Playoff',
+    'Relegation Round',
+    'Relegation Playoff',
+    'Play-offs',
+    'Playoff Stage',
+    'Grunddurchgang',
+    'Finaldurchgang - Qualifikationsgruppe',
+    'Finaldurchgang - Qualifikation',
+    'Finaldurchgang - Meistergruppe',
+    'Finaldurchgang - Meister',
+    'EL Play-off',
+    'Europa League Play-off',
+    'Europa-League-Play-offs',
+  ].map {|name| name.downcase.gsub( /[^a-z]/, '' ) }
+
+
+  def check_stage( name )
+    # note: normalize names e.g. downcase and remove all non a-z chars (e.g. space, dash, etc.)
+    if KNOWN_STAGES.include?( name.downcase.gsub( /[^a-z]/, '' ) )
        ## everything ok
     else
       puts "** !!! ERROR - no (league) stage match found for >#{name}<, add to (builtin) stages table; sorry"

data/lib/sportdb/formats/league/league_reader.rb

@@ -113,9 +113,9 @@ def parse
             ##             add a list of (auto-)excluded country codes with conflicts? why? why not?
             ##                 cl - a) Chile  b) Champions League
             alt_names_auto << "#{country.key.upcase}"   if league_key == '1'   ## add shortcut for top level 1 (just country key)
-            if country.key.upcase != country.fifa
-              alt_names_auto << "#{country.fifa} #{league_key.upcase.gsub('.', ' ')}"
-              alt_names_auto << "#{country.fifa}"    if league_key == '1'   ## add shortcut for top level 1 (just country key)
+            if country.key.upcase != country.code
+              alt_names_auto << "#{country.code} #{league_key.upcase.gsub('.', ' ')}"
+              alt_names_auto << "#{country.code}"    if league_key == '1'   ## add shortcut for top level 1 (just country key)
             end
             alt_names_auto << "#{country.name} #{league_key}"  if league_key =~ /^[0-9]+$/   ## if all numeric e.g. add Austria 1 etc.
           else   ## assume int'l (no country) e.g. champions league, etc.

data/lib/sportdb/formats/match/mapper.rb

@@ -7,21 +7,21 @@ module SportDb
 ##   see https://github.com/textkit/textutils/blob/master/textutils/lib/textutils/title_mapper2.rb
 
 
-class MapperV2      ## todo/check: rename to NameMapper/TitleMapper ? why? why not??
+class MapperV2      ## todo/check: rename to NameMapper ? why? why not??
 
   include Logging
 
-  attr_reader :known_titles   ## rename to mapping or mappings or just titles - why? why not?
+  attr_reader :known_names   ## rename to mapping or mappings or just names - why? why not?
 
   ########
   ##  key:      e.g. augsburg
-  ##  title:    e.g. FC Augsburg
-  ##  length (of title(!!) - not regex pattern):   e.g. 11   -- do not count dots (e.g. U.S.A. => 3 or 6) why? why not?
-  MappingStruct =  Struct.new( :key, :title, :length, :pattern)     ## todo/check: use (rename to) TitleStruct - why? why not??
+  ##  name:     e.g. FC Augsburg
+  ##  length (of name(!!) - not regex pattern):   e.g. 11   -- do not count dots (e.g. U.S.A. => 3 or 6) why? why not?
+  MappingStruct =  Struct.new( :key, :name, :length, :pattern)     ## todo/check: use (rename to) NameStruct - why? why not??
 
   ######
   ## convenience helper - (auto)build ActiveRecord-like team records/structs
-  Record = Struct.new( :key, :title, :synonyms )
+  Record = Struct.new( :key, :name, :alt_names )
   def build_records( txt_or_lines )
     recs = []
 
@@ -44,12 +44,12 @@ class MapperV2      ## todo/check: rename to NameMapper/TitleMapper ? why? why n
       values = line.split( '|' )
       values = values.map { |value| value.strip }
 
-      title    = values[0]
+      name      = values[0]
       ## note: quick hack - auto-generate key, that is, remove all non-ascii chars and downcase
-      key      = title.downcase.gsub( /[^a-z]/, '' )
-      synonyms = values.size > 1 ? values[1..-1].join( '|' ) : nil
+      key       = name.downcase.gsub( /[^a-z]/, '' )
+      alt_names = values.size > 1 ? values[1..-1].join( '|' ) : nil
 
-      recs << Record.new( key, title, synonyms )
+      recs << Record.new( key, name, alt_names )
     end
     recs
   end
@@ -63,10 +63,10 @@ class MapperV2      ## todo/check: rename to NameMapper/TitleMapper ? why? why n
                                                                   (records_or_mapping.is_a?( Array ) && records_or_mapping[0].is_a?( String ))
 
     ## build mapping lookup table
-    @known_titles =  if records_or_mapping.is_a?( Hash )  ## assume "custom" mapping hash table (title/name=>record)
-                        build_title_table_for_mapping( records_or_mapping )
+    @known_names =  if records_or_mapping.is_a?( Hash )  ## assume "custom" mapping hash table (name=>record)
+                        build_name_table_for_mapping( records_or_mapping )
                      else  ## assume array of records
-                        build_title_table_for_records( records_or_mapping )
+                        build_name_table_for_records( records_or_mapping )
                      end
 
     ## build lookup hash by record (e.g. team/club/etc.) key
@@ -85,9 +85,9 @@ class MapperV2      ## todo/check: rename to NameMapper/TitleMapper ? why? why n
 
 
 
-  def map_titles!( line )   ## rename to just map! - why?? why not???
+  def map_names!( line )   ## rename to just map! - why?? why not???
     begin
-      found = map_title_for!( @tag, line, @known_titles )
+      found = map_name_for!( @tag, line, @known_names )
     end while found
   end
 
@@ -110,27 +110,27 @@ class MapperV2      ## todo/check: rename to NameMapper/TitleMapper ? why? why n
 
 
 private
-  def build_title_table_for_mapping( mapping )
-    known_titles = []
+  def build_name_table_for_mapping( mapping )
+    known_names = []
 
-    mapping.each do |title, rec|
+    mapping.each do |name, rec|
       m = MappingStruct.new
       m.key     = rec.key
-      m.title   = title
-      m.length  = title.length
-      m.pattern = Regexp.escape( title )   ## note: just use "standard" regex escape (e.g. no extras for umlauts,accents,etc.)
+      m.name    = name
+      m.length  = name.length
+      m.pattern = Regexp.escape( name )   ## note: just use "standard" regex escape (e.g. no extras for umlauts,accents,etc.)
 
-      known_titles << m
+      known_names << m
     end
 
     ## note: sort here by length (largest goes first - best match)
-    known_titles = known_titles.sort { |l,r| r.length <=> l.length }
-    known_titles
+    known_names = known_names.sort { |l,r| r.length <=> l.length }
+    known_names
   end
 
-  def build_title_table_for_records( records )
+  def build_name_table_for_records( records )
 
-    ## build known tracks table w/ synonyms e.g.
+    ## build known tracks table w/ alt names e.g.
     #
     # [[ 'wolfsbrug', 'VfL Wolfsburg'],
     #  [ 'augsburg',  'FC Augsburg'],
@@ -138,65 +138,65 @@ private
     #  [ 'augsburg',  'Augi3' ],
     #  [ 'stuttgart', 'VfB Stuttgart']]
 
-    known_titles = []
+    known_names = []
 
     records.each_with_index do |rec,index|
 
-      title_candidates = []
-      title_candidates << rec.title
+      name_candidates = []
+      name_candidates << rec.name
 
-      title_candidates += rec.synonyms.split('|') if rec.synonyms && !rec.synonyms.empty?
+      name_candidates += rec.alt_names.split('|') if rec.alt_names && !rec.alt_names.empty?
 
 
-      ## check if title includes subtitle e.g. Grand Prix Japan (Suzuka Circuit)
-      #  make subtitle optional by adding title w/o subtitle e.g. Grand Prix Japan
+      ## check if name includes subname e.g. Grand Prix Japan (Suzuka Circuit)
+      #  make subname optional by adding name w/o subname e.g. Grand Prix Japan
 
-      titles = []
-      title_candidates.each do |t|
-        titles << t
+      names = []
+      name_candidates.each do |t|
+        names << t
         if t =~ /\(.+\)/
-          extra_title = t.gsub( /\(.+\)/, '' ) # remove/delete subtitles
+          extra_name = t.gsub( /\(.+\)/, '' ) # remove/delete subnames
           # note: strip leading n trailing withspaces too!
           #  -- todo: add squish or something if () is inline e.g. leaves two spaces?
-          extra_title.strip!
-          titles << extra_title
+          extra_name.strip!
+          names << extra_name
         end
       end
 
-      titles.each do |t|
+      names.each do |name|
         m = MappingStruct.new
         m.key     = rec.key
-        m.title   = t
-        m.length  = t.length
+        m.name    = name
+        m.length  = name.length
         ## note: escape for regex plus allow subs for special chars/accents
-        m.pattern = title_esc_regex( t )
+        m.pattern = name_esc_regex( name )
 
-        known_titles << m
+        known_names << m
       end
 
-      logger.debug "  #{rec.class.name}[#{index+1}] #{rec.key} >#{titles.join('|')}<"
+      logger.debug "  #{rec.class.name}[#{index+1}] #{rec.key} >#{names.join('|')}<"
 
       ## note: only include code field - if defined
       if rec.respond_to?(:code) && rec.code && !rec.code.empty?
         m = MappingStruct.new
         m.key     = rec.key
-        m.title   = rec.code
+        m.name    = rec.code
         m.length  = rec.code.length
         m.pattern = rec.code   ## note: use code for now as is (no variants allowed fow now)
 
-        known_titles << m
+        known_names << m
       end
     end
 
     ## note: sort here by length (largest goes first - best match)
       #  exclude code and key (key should always go last)
-    known_titles = known_titles.sort { |l,r| r.length <=> l.length }
-    known_titles
+    known_names = known_names.sort { |l,r| r.length <=> l.length }
+    known_names
   end
 
 
 
-  def map_title_for!( tag, line, mappings )
+  def map_name_for!( tag, line, mappings )
     mappings.each do |mapping|
       key     = mapping.key
       pattern = mapping.pattern
@@ -234,9 +234,9 @@ private
 
 
 ####
-# title helper cut-n-paste copy from TextUtils
+# name helper cut-n-paste copy from TextUtils
 ##  see https://github.com/textkit/textutils/blob/master/textutils/lib/textutils/helper/title_helper.rb
-def title_esc_regex( title_unescaped )
+def name_esc_regex( name_unescaped )
 
       ##  escape regex special chars e.g.
       #    . to \. and
@@ -257,16 +257,16 @@ def title_esc_regex( title_unescaped )
       # e.g. Club Atlético Colón (Santa Fe)
       # e.g. Bauer Anton (????)
 
-      ## NB: cannot use Regexp.escape! will escape space '' to '\ '
-      ## title = Regexp.escape( title_unescaped )
-      title = title_unescaped.gsub( '.', '\.' )
-      title = title.gsub( '(', '\(' )
-      title = title.gsub( ')', '\)' )
-      title = title.gsub( '?', '\?' )
-      title = title.gsub( '*', '\*' )
-      title = title.gsub( '+', '\+' )
-      title = title.gsub( '$', '\$' )
-      title = title.gsub( '^', '\^' )
+      ## note: cannot use Regexp.escape! will escape space '' to '\ '
+      ## name = Regexp.escape( name_unescaped )
+      name = name_unescaped.gsub( '.', '\.' )
+      name = name.gsub( '(', '\(' )
+      name = name.gsub( ')', '\)' )
+      name = name.gsub( '?', '\?' )
+      name = name.gsub( '*', '\*' )
+      name = name.gsub( '+', '\+' )
+      name = name.gsub( '$', '\$' )
+      name = name.gsub( '^', '\^' )
 
       ##  match accented char with or without accents
       ##  add (ü|ue) etc.
@@ -309,10 +309,10 @@ def title_esc_regex( title_unescaped )
       ##   collect some more (real-world) examples first!!!!!
 
       alternatives.each do |alt|
-        title = title.gsub( alt[0], alt[1] )
+        name = name.gsub( alt[0], alt[1] )
       end
 
-      title
+      name
   end
 
 end # class MapperV2

data/lib/sportdb/formats/match/mapper_teams.rb

@@ -16,7 +16,7 @@ class TeamMapper
   end
 
   def map_teams!( line )
-    @mapper.map_titles!( line )
+    @mapper.map_names!( line )
   end
 end # class TeamMapper
 

data/lib/sportdb/formats/match/match_parser.rb

@@ -82,15 +82,14 @@ class MatchParser   ## simple match parser for team match schedules
     #    team1 team2 - match  (will get new auto-matchday! not last round)
     @last_round     = nil
 
-    title, pos = find_group_title_and_pos!( line )
+    name = find_group_name!( line )
 
-    logger.debug "    title: >#{title}<"
-    logger.debug "    pos: >#{pos}<"
+    logger.debug "    name: >#{name}<"
     logger.debug "  line: >#{line}<"
 
-    group = @groups[ title ]
+    group = @groups[ name ]
     if group.nil?
-      puts "!! ERROR - no group def found for >#{title}<"
+      puts "!! ERROR - no group def found for >#{name}<"
       exit 1
     end
 
@@ -104,19 +103,19 @@ class MatchParser   ## simple match parser for team match schedules
     @mapper_teams.map_teams!( line )
     teams = @mapper_teams.find_teams!( line )
 
-    title, pos = find_group_title_and_pos!( line )
+    name = find_group_name!( line )
 
     logger.debug "  line: >#{line}<"
 
-    group = Import::Group.new( pos: pos,
-                               title: title,
-                               teams: teams.map {|team| team.title } )
+    ## todo/check/fix: add back group key - why? why not?
+    group = Import::Group.new( name:  name,
+                               teams: teams.map {|team| team.name } )
 
-    @groups[ title ] = group
+    @groups[ name ] = group
   end
 
 
-  def find_group_title_and_pos!( line )
+  def find_group_name!( line )
     ## group pos - for now support single digit e.g 1,2,3 or letter e.g. A,B,C or HEX
     ## nb:  (?:)  = is for non-capturing group(ing)
 
@@ -125,37 +124,25 @@ class MatchParser   ## simple match parser for team match schedules
 
     ## todo:
     ##   check if Group A:  or [Group A]  works e.g. : or ] get matched by \b ???
-    regex = /(?:Group|Gruppe|Grupo)\s+((?:\d{1}|[A-Z]{1,3}))\b/
+    regex = /\b
+              (?:
+                (Group | Gruppe | Grupo)
+                   [ ]+
+                (\d+ | [A-Z]+)
+              )
+            \b/x
 
     m = regex.match( line )
 
-    return [nil,nil] if m.nil?
-
-    pos = case m[1]
-          when 'A' then 1
-          when 'B' then 2
-          when 'C' then 3
-          when 'D' then 4
-          when 'E' then 5
-          when 'F' then 6
-          when 'G' then 7
-          when 'H' then 8
-          when 'I' then 9
-          when 'J' then 10
-          when 'K' then 11
-          when 'L' then 12
-          when 'HEX' then 666    # HEX for Hexagonal - todo/check: map to something else ??
-          else  m[1].to_i
-          end
-
-    title = m[0]
-
-    logger.debug "   title: >#{title}<"
-    logger.debug "   pos: >#{pos}<"
-
-    line.sub!( regex, '[GROUP.TITLE+POS]' )
-
-    [title,pos]
+    return nil    if m.nil?
+
+    name = m[0]
+
+    logger.debug "   name: >#{name}<"
+
+    line.sub!( name, '[GROUP.NAME]' )
+
+    name
   end
 
 
@@ -180,198 +167,130 @@ class MatchParser   ## simple match parser for team match schedules
     end_date   = end_date.to_date
 
 
-    pos   = find_round_pos!( line )
-    title = find_round_def_title!( line )
-    # NB: use extracted round title for knockout check
-    knockout_flag = is_knockout_round?( title )
+    name  = find_round_def_name!( line )
+    # NB: use extracted round name for knockout check
+    knockout_flag = is_knockout_round?( name )
 
 
     logger.debug "    start_date: #{start_date}"
     logger.debug "    end_date:   #{end_date}"
-    logger.debug "    pos:      #{pos}"
-    logger.debug "    title:    >#{title}<"
+    logger.debug "    name:    >#{name}<"
     logger.debug "    knockout_flag:   #{knockout_flag}"
 
     logger.debug "  line: >#{line}<"
 
-    #######################################
-    # todo/fix: add auto flag is false !!!! - why? why not?
-    round = Import::Round.new( pos:        pos,
-                               title:      title,
+    round = Import::Round.new( name:       name,
                                start_date: start_date,
                                end_date:   end_date,
                                knockout:   knockout_flag,
                                auto:       false )
 
-    @rounds[ title ] = round
+    @rounds[ name ] = round
   end
 
 
 
-  def find_round_pos!( line )
-    # pass #1) extract optional round pos from line
-    # e.g.  (1)   - must start line
-    regex_pos = /^[ \t]*\((\d{1,3})\)[ \t]+/
-
-    # pass #2) find free standing number  e.g. Matchday 3 or Round 5 or 3. Spieltag etc.
-    # note: /\b(\d{1,3})\b/
-    #   will match -12
-    #  thus, use space required - will NOT match  -2 e.g. Group-2 Play-off
-    #  note:  allow  1. Runde  n
-    #                1^ Giornata
-    regex_num = /(?:^|\s)(\d{1,3})(?:[.\^\s]|$)/
-
-    if line =~ regex_pos
-      logger.debug "   pos: >#{$1}<"
-
-      line.sub!( regex_pos, '[ROUND.POS] ' )  ## NB: add back trailing space that got swallowed w/ regex -> [ \t]+
-      return $1.to_i
-    elsif line =~ regex_num
-      ## assume number in title is pos (e.g. Jornada 3, 3 Runde etc.)
-      ## NB: do NOT remove pos from string (will get removed by round title)
-
-      num = $1.to_i  # note: clone capture; keep a copy (another regex follows; will redefine $1)
-
-      #### fix:
-      #  use/make keywords required
-      #  e.g. Round of 16  -> should NOT match 16!
-      #    Spiel um Platz 3  (or 5) etc -> should NOT match 3!
-      #  Round 16 - ok
-      #  thus, check for required keywords
+  def find_round_def_name!( line )
+    # assume everything before pipe (\) is the round name
+    #  strip [ROUND.POS],  todo:?? [ROUND.NAME2]
 
-      ## quick hack for round of 16
-      # todo: mask match e.g. Round of xxx ... and try again - might include something
-      #  reuse pattern for Group XX Replays for example
-      if line =~ /^\s*Round of \d{1,3}\b/
-         return nil
-      end
-
-      logger.debug "   pos: >#{num}<"
-      return num
-    else
-      ## fix: add logger.warn no round pos found in line
-      return nil
-    end
-  end # method find_round_pos!
-
-  def find_round_def_title!( line )
-    # assume everything before pipe (\) is the round title
-    #  strip [ROUND.POS],  todo:?? [ROUND.TITLE2]
-
-    # todo/fix: add title2 w/  // or /  why? why not?
+    # todo/fix: add name2 w/  // or /  why? why not?
     #  -- strip / or / chars
 
     buf = line.dup
-    logger.debug "  find_round_def_title! line-before: >>#{buf}<<"
+    logger.debug "  find_round_def_name! line-before: >>#{buf}<<"
 
     ## cut-off everything after (including) pipe (|)
     buf = buf[ 0...buf.index('|') ]
-
-    # e.g. remove [ROUND.POS], [ROUND.TITLE2], [GROUP.TITLE+POS] etc.
-    buf.gsub!( /\[[^\]]+\]/, '' )    ## fix: use helper for (re)use e.g. remove_match_placeholder/marker or similar?
-    # remove leading and trailing whitespace
     buf.strip!
 
-    logger.debug "  find_round_def_title! line-after: >>#{buf}<<"
+    logger.debug "  find_round_def_name! line-after: >>#{buf}<<"
 
-    logger.debug "   title: >>#{buf}<<"
-    line.sub!( buf, '[ROUND.TITLE]' )
+    logger.debug "   name: >>#{buf}<<"
+    line.sub!( buf, '[ROUND.NAME]' )
 
     buf
   end
 
-  def find_round_header_title!( line )
-    # assume everything left is the round title
-    #  extract all other items first (round title2, round pos, group title n pos, etc.)
 
-    ## todo/fix:
-    ##  cleanup method
-    ##   use  buf.index( '//' ) to split string (see found_round_def)
-    ##     why? simpler why not?
-    ##  - do we currently allow groups if title2 present? add example if it works?
+    ##  split by or || or |||
+    ##           or ++ or +++
+    ##           or -- or ---
+    ##           or // or ///
+    ##  note: allow Final | First Leg  as ONE name same as
+    ##              Final - First Leg or
+    ##              Final, First Leg
+    ##   for cut-off always MUST be more than two chars
+    ##
+    ##  todo/check: find a better name than HEADER_SEP(ARATOR) - why? why not?
+    ##   todo/fix: move to parser utils and add a method split_name or such?
+    HEADER_SEP_RE = /  [ ]*      ## allow (strip) leading spaces
+                      (?:\|{2,} |
+                          \+{2,} |
+                           -{2,} |
+                          \/{2,}
+                      )
+                      [ ]*       ## allow (strip) trailing spaces
+                  /x
+
+  def find_round_header_name!( line )
+    # assume everything left is the round name
+    #  extract all other items first (round name2, round pos, group name n pos, etc.)
 
     buf = line.dup
-    logger.debug "  find_round_header_title! line-before: >>#{buf}<<"
+    logger.debug "  find_round_header_name! line-before: >>#{buf}<<"
+
+
+    parts = buf.split( HEADER_SEP_RE )
+    buf = parts[0]
 
-    buf.gsub!( /\[[^\]]+\]/, '' )   # e.g. remove [ROUND.POS], [ROUND.TITLE2], [GROUP.TITLE+POS] etc.
     buf.strip!    # remove leading and trailing whitespace
 
-    logger.debug "  find_round_title! line-after: >>#{buf}<<"
+    logger.debug "  find_round_name! line-after: >>#{buf}<<"
 
-    ### bingo - assume what's left is the round title
+    ### bingo - assume what's left is the round name
 
-    logger.debug "   title: >>#{buf}<<"
-    line.sub!( buf, '[ROUND.TITLE]' )
+    logger.debug "   name: >>#{buf}<<"
+    line.sub!( buf, '[ROUND.NAME]' )
 
     buf
   end
 
+    ## quick hack- collect all "fillwords" by language!!!!
+    ##    change later  and add to sportdb-langs!!!!
+    ##
+    ##    strip all "fillwords" e.g.:
+    ##      Nachtrag/Postponed/Addition/Supplemento names
+    ##
+    ##  todo/change: find a better name for ROUND_EXTRA_WORDS - why? why not?
+    ROUND_EXTRA_WORDS_RE = /\b(?:
+                               Nachtrag |     ## de
+                               Postponed |    ## en
+                               Addition  |    ## en
+                               Supplemento    ## es
+                              )
+                             \b/ix
 
   def parse_round_header( line )
     logger.debug "parsing round header line: >#{line}<"
 
-    ## todo/check/fix:
-    #   make sure  Round of 16  will not return pos 16 -- how? possible?
-    #   add unit test too to verify
-    pos = find_round_pos!( line )
-
-    title = find_round_header_title!( line )
+    name = find_round_header_name!( line )
 
     logger.debug "  line: >#{line}<"
 
+    name = name.sub( ROUND_EXTRA_WORDS_RE, '' )
+    name = name.strip
 
-    round = @rounds[ title ]
+    round = @rounds[ name ]
     if round.nil?    ## auto-add / create if missing
-      round = Import::Round.new( pos:   pos,
-                                 title: title )
-      @rounds[ title ] = round
+      ## todo/check: add num (was pos) if present - why? why not?
+      round = Import::Round.new( name: name )
+      @rounds[ name ] = round
     end
 
     ## todo/check: if pos match (MUST always match for now)
     @last_round = round
     @last_group = nil   # note: reset group to no group - why? why not?
-
-
-    ## NB: dummy/placeholder start_at, end_at date
-    ##  replace/patch after adding all games for round
-
-=begin
-    round_attribs = {
-      title:  title,
-      title2: title2,
-      knockout: knockout_flag
-    }
-
-    if pos > 999000
-      # no pos (e.g. will get autonumbered later) - try match by title for now
-      #  e.g. lets us use title 'Group Replays', for example, multiple times
-      @round = Round.find_by_event_id_and_title( @event.id, title )
-    else
-      @round = Round.find_by_event_id_and_pos( @event.id, pos )
-    end
-
-    if @round.present?
-      logger.debug "update round #{@round.id}:"
-    else
-      logger.debug "create round:"
-      @round = Round.new
-
-      round_attribs = round_attribs.merge( {
-        event_id: @event.id,
-        pos:   pos,
-        start_at: Date.parse('1911-11-11'),
-        end_at:   Date.parse('1911-11-11')
-      })
-    end
-
-    logger.debug round_attribs.to_json
-
-    @round.update_attributes!( round_attribs )
-
-    @patch_round_ids_pos   << @round.id    if pos > 999000
-    ### store list of round ids for patching start_at/end_at at the end
-    @patch_round_ids_dates << @round.id   # todo/fix/check: check if round has definition (do NOT patch if definition (not auto-added) present)
-=end
   end
 
 
@@ -457,11 +376,11 @@ class MatchParser   ## simple match parser for team match schedules
     ## todo/check: pass along round and group refs or just string (canonical names) - why? why not?
 
     @matches << Import::Match.new( date:    date,
-                                   team1:   team1,  ## note: for now always use mapping value e.g. rec (NOT string e.g. team1.title)
-                                   team2:   team2,  ## note: for now always use mapping value e.g. rec (NOT string e.g. team2.title)
+                                   team1:   team1,  ## note: for now always use mapping value e.g. rec (NOT string e.g. team1.name)
+                                   team2:   team2,  ## note: for now always use mapping value e.g. rec (NOT string e.g. team2.name)
                                    score:   score,
-                                   round:   round       ? round.title       : nil,   ## note: for now always use string (assume unique canonical name for event)
-                                   group:   @last_group ? @last_group.title : nil )  ## note: for now always use string (assume unique canonical name for event)
+                                   round:   round       ? round.name       : nil,   ## note: for now always use string (assume unique canonical name for event)
+                                   group:   @last_group ? @last_group.name : nil )  ## note: for now always use string (assume unique canonical name for event)
 
     ### todo: cache team lookups in hash?
 
@@ -517,7 +436,7 @@ class MatchParser   ## simple match parser for team match schedules
 
         round_attribs = {
           event_id: @event.id,
-          title: "Matchday #{date.to_date}",
+          name: "Matchday #{date.to_date}",
           pos: 999001+@patch_round_ids_pos.length,   # e.g. 999<count> - 999001,999002,etc.
           start_at:  date.to_date,
           end_at:    date.to_date
@@ -541,7 +460,7 @@ class MatchParser   ## simple match parser for team match schedules
       end
 
       ## note: will crash (round.pos) if round is nil
-      logger.debug( "  using round #{round.pos} >#{round.title}< start_at: #{round.start_at}, end_at: #{round.end_at}" )
+      logger.debug( "  using round #{round.pos} >#{round.name}< start_at: #{round.start_at}, end_at: #{round.end_at}" )
     else
       ## use round from last round header
       round = @round