sportdb-search 0.0.1 → 0.1.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: a675e2ad48a651c6b31ccc0f68f838907384a93078269dff90c87e5424472a56
-  data.tar.gz: 8abf21926e4e3924ef71df5da72330c834fe0a54679efffdb6657e90361e4bac
+  metadata.gz: a25ed40d388f7a3ce09acc8a67a5a16aec359696141c3260bda097b5d221c130
+  data.tar.gz: ec681b64b842e733d874ab9c264e65797356129a0beccf11d9d40d55d1ea34fa
 SHA512:
-  metadata.gz: db536b42aabda47e6a6eb7b2da08a7454f91c017801948a86ee7207762a5ffc9ebc5eb7a32e65378aeea4ad8a93af7c0db390458e667dd6cc350c2d3e9dfa24e
-  data.tar.gz: 5b69bf93a1dd374d2dcf50907c26aeaf7781af34ba127d91830f09d8e9406dfcf1345b806222258ba6da7d7bf2df8590790d8f3763298c9a1909ca536a02097a
+  metadata.gz: 78d1cbcea722b03bbb6d6986fda9837b3d15e1aaf9dd8b5085d382435e56a67bf62c228f0b33ff4ff01845e5a78e4ea86163ba1a150e8e9166ac66e0d5a81867
+  data.tar.gz: 7e17104eb185b9c4f82e12eff407e9df4c352c6884194ae04036dd4bae690d0671c10ff723387bb12da54212a6033a3490a1763dfdbb55fd91cb916025d7199d

data/CHANGELOG.md

@@ -1,3 +1,5 @@
+### 0.1.1
+
 ### 0.0.1 / 2024-08-25
 
 * Everything is new. First release.

data/Manifest.txt

@@ -5,5 +5,5 @@ Rakefile
 lib/sportdb/search.rb
 lib/sportdb/search/sport.rb
 lib/sportdb/search/structs.rb
+lib/sportdb/search/structs_world.rb
 lib/sportdb/search/version.rb
-lib/sportdb/search/world.rb

data/README.md

@@ -7,10 +7,71 @@
 
 
 
+
 ## Usage
 
+Let's use the [/clubs datasets](https://github.com/openfootball/clubs)
+(3000+ football clubs from around the world)
+to match name "variants" e.g. `Arsenal`  to canonical global unique
+names e.g. `Arsenal FC, London, England`:
+
+``` ruby
+require 'sportdb/search'
+
+
+Club = Sports::Club
+
+m = Club.match_by( name: 'Arsenal' )
+m.size     # 3 club matches found
+#=> 3
+m[0].name; m[0].city; m[0].country
+#=> "Arsenal FC", "London", "England"
+m[1].name; m[1].city; m[1].country
+#=> "Arsenal Tula", "Tula", "Russia"
+m[2].name; m[2].city; m[2].country
+#=> "Arsenal de Sarandí", "Sarandí", "Argentina"
+
+
+m = Club.match_by( name: 'Arsenal', country: 'eng' )
+# -or- try alternative names (and auto-generated spelling variants)
+m = Club.match_by( name: 'Arsenal FC', country: 'eng' )
+m = Club.match_by( name: 'Arsenal F.C.', country: 'eng' )
+m = Club.match_by( name: '...A.r.s.e.n.a.l... F.C...', country: 'eng' )
+m.size     # 1 club match found
+#=> 1
+m[0].name; m[0].city; m[0].country
+#=> "Arsenal FC", "London", "England"
+
+m = Club.match_by( name: 'Arsenal', country: 'ar' )
+# -or- try alternative names (and auto-generated spelling variants)
+m = Club.match_by( name: 'Arsenal Sarandí', country: 'ar' )
+m = Club.match_by( name: 'Arsenal Sarandi', country: 'ar' )
+m.size     # 1 club match found
+#=> 1
+m[0].name; m[0].city; m[0].country
+#=> "Arsenal de Sarandí", "Sarandí", "Argentina"
+
+
+# try some more
+m = Club.match_by( name: 'AZ' )
+m[0].name; m[0].city; m[0].country
+#=> "AZ Alkmaar", "Alkmaar", "Netherlands"
+
+m = Club.match_by( name: 'Bayern' )
+# -or- try alternative names (and auto-generated spelling variants)
+m = Club.match_by( name: 'Bayern München' )
+m = Club.match_by( name: 'Bayern Munchen' )
+m = Club.match_by( name: 'Bayern Muenchen' )
+m[0].name; m[0].city; m[0].country
+#=> "Bayern München", "München", "Germany"
+
+# and so on
+# ...
+```
+
+
+That's it.
 
-to be done
 
 
 ## License

data/lib/sportdb/search/sport.rb

@@ -302,7 +302,68 @@ class TeamSearch
     end
 
 
+
+  CLUB_NAME_RE =  %r{^
+          (?<name>[^()]+?)     ## non-greedy
+          (?:
+             \s+
+             \(
+               (?<code>[A-Z][A-Za-z]{2,3})    ## optional (country) code; support single code e.g. (A) - why? why not?
+             \)
+          )?
+        $}x   ## note - allow (URU) and (Uru) - why? why not
+
+
+    ###
+    #  note:  missing teams will get
+    ##            auto-created if possible
+    ##         only ambigious results (too many matches) raise expection!!!
     def _find_by!( name:, league:, mods: nil )
+      if mods && mods[ league.key ] && mods[ league.key ][ name ]
+        mods[ league.key ][ name ]
+      else
+        if league.clubs?
+          if league.intl?    ## todo/fix: add intl? to ActiveRecord league!!!
+                 ###
+                 ##  get country code from name
+                 ##    e.g. Liverpool FC (ENG) or
+                 ##         Liverpool FC (URU) etc.
+
+                 ## check for country code
+                 if m=CLUB_NAME_RE.match( name )
+                   if m[:code]
+                     @clubs.find_by!( name: m[:name],
+                                            country: m[:code] )
+                   else
+                      @clubs.find!( name )
+                   end
+                 else
+                   puts "!! PARSE ERROR - invalid club name; cannot match with CLUB_NAME_RE >#{team}<"
+                   exit 1
+                 end
+          else  ## assume clubs in domestic/national league tournament
+             ## note - search by league countries (may incl. more than one country
+             ##             e.g. us incl. ca, fr incl. mc, ch incl. li, etc.
+            rec = @clubs.find_by( name: name, league: league )
+            if rec.nil?
+                puts "auto-create (missing) club #{name}"
+                ##  todo/fix: add auto flag!!!!
+                ###              like in rounds!!!
+                ##   to track auto-created clubs
+                rec = SportDb::Import::Club.new( name: name )
+                rec.country = league.country  ## fix: country kwarg not yet supported!!
+                pp rec
+            end
+            rec
+          end
+        else   ## assume national teams (not clubs)
+          @national_teams.find!( name )
+        end
+      end
+    end # method _find_by!
+
+
+    def _find_by_v0!( name:, league:, mods: nil )
       if mods && mods[ league.key ] && mods[ league.key ][ name ]
         mods[ league.key ][ name ]
       else

data/lib/sportdb/search/structs.rb

@@ -10,47 +10,6 @@
 
 
 module Sports
-
-class Country
-    def self._search #### use service/api or such - why? why not?
-        SportDb::Import.world.countries
-    end
-   def self.find_by( code: nil, name: nil )
-        _search.find_by( code: code, name: name )
-   end
-
-   def self.find( q )   ## find by code (first) or name (second)
-       _search.find( q )
-   end
-
-   def self.parse_heading( line )
-      ## fix - move parse code here from search - why? why not?
-      _search.parse( line )
-   end
-
-   ## add alternate names/aliases
-   class << self
-    alias_method :[],      :find    ### keep shortcut - why? why not?
-    alias_method :heading, :parse_heading
-   end
-
-
-# open question - what name to use build or  parse_line or ?
-#                              or   parse_recs for CountryReader?
-#          remove CountryReader helper methods - why? why not?
-#   use parse_heading/heading for now !!!
-#
-#   def self.parse( line )  or build( line ) ??
-#      SportDb::Import.world.countries.parse( line )
-#   end
-#
-# !!!! note - conflict with
-#     def self.read( path )  CountryReader.read( path ); end
-#     def self.parse( txt )  CountryReader.parse( txt ); end
-#
-end # class Country
-
-
 ###
 ## todo/fix - add find_by( code: ), find_by( name: )
 ##                   split - why? why not?
@@ -113,11 +72,38 @@ class Club
     end
 
    def self.build_mods( mods )
-       _search_build_mods( mods )
+       _search.build_mods( mods )
     end
   end # class Club
 
 
+  class Team
+    def self._search
+        SportDb::Import.catalog.teams
+    end
+
+    ## todo/check: rename to/use map_by! for array version - why? why not?
+    def self.find_by!( name:, league:, mods: nil )
+        _search.find_by!( name: name,
+                          league: league,
+                          mods: mods )
+    end
+  end # class Team
+
+
+  class EventInfo
+    def self._search
+        SportDb::Import.catalog.events
+    end
+
+    def self.find_by( league:, season: )
+        _search.find_by( league: league,
+                         season: season )
+    end
+  end # class EventInfo
+
+
+
   class Ground
     def self._search
         SportDb::Import.catalog.grounds

data/lib/sportdb/search/structs_world.rb

@@ -0,0 +1,127 @@
+module Sports
+
+class City
+    def self._search #### use service/api or such - why? why not?
+        SportDb::Import.world.cities
+    end
+
+    def self.match_by( name: )
+       _search.match_by( name: name )
+    end
+end   # class City
+
+
+class Country
+    def self._search #### use service/api or such - why? why not?
+        SportDb::Import.world.countries
+    end
+
+    def self.find_by( code: nil, name: nil )
+        ## todo/fix upstream - change to find_by( code:, name:, ) too - why? why not?
+        if code && name.nil?
+          _search.find_by_code( code )
+        elsif name && code.nil?
+          _search.find_by_name( name )
+        else
+          raise ArgumentError, "CountrySearch#find_by - one (and only one arg) required - code: or name:"
+        end
+    end
+
+    def self.find( q )   ## find by code (first) or name (second)
+        _search.find_by_name_or_code( q )
+    end
+
+ ###
+ ##   split/parse country line
+ ##
+ ##  split on bullet e.g.
+ ##   split into name and code with regex - make code optional
+ ##
+ ##  Examples:
+ ##    Österreich • Austria (at)
+ ##    Österreich • Austria
+ ##    Austria
+ ##    Deutschland (de) • Germany
+ ##
+ ##   todo/check: support more formats - why? why not?
+ ##       e.g.  Austria, AUT  (e.g. with comma - why? why not?)
+ def self.parse_heading( line )
+    values = line.split( '•' )   ## use/support multi-lingual separator
+    country = nil
+    values.each do |value|
+       value = value.strip
+       ## check for trailing country code e.g. (at), (eng), etc
+       ##   allow code 1 to 5 for now - northern cyprus(fifa) with 5 letters?.
+       ##     add/allow  gb-eng, gb-wal (official iso2!!), in the future too - why? why not?
+       if value =~ /[ ]+\((?<code>[A-Za-z]{1,5})\)$/  ## e.g. Austria (at)
+         code =  $~[:code]
+         name = value[0...(value.size-code.size-2)].strip  ## note: add -2 for brackets
+         candidates = [ find_by( code: code ), find_by( name: name ) ]
+         if candidates[0].nil?
+           puts "** !!! ERROR Country.parse_heading - unknown code >#{code}< in line: #{line}"
+           pp line
+           exit 1
+         end
+         if candidates[1].nil?
+           puts "** !!! ERROR Country.parse_heading - unknown name >#{code}< in line: #{line}"
+           pp line
+           exit 1
+         end
+         if candidates[0] != candidates[1]
+           puts "** !!! ERROR Country.parse_heading - name and code do NOT match the same country:"
+           pp line
+           pp candidates
+           exit 1
+         end
+         if country && country != candidates[0]
+           puts "** !!! ERROR Country.parse_heading - names do NOT match the same country:"
+           pp line
+           pp country
+           pp candidates
+           exit 1
+         end
+         country = candidates[0]
+       else
+         ## just assume value is name or code
+         candidate = find( value )
+         if candidate.nil?
+           puts "** !!! ERROR Country.parse_heading - unknown name or code >#{value}< in line: #{line}"
+           pp line
+           exit 1
+         end
+         if country && country != candidate
+           puts "** !!! ERROR Country.parse_heading - names do NOT match the same country:"
+           pp line
+           pp country
+           pp candidate
+           exit 1
+         end
+         country = candidate
+       end
+    end
+    country
+  end # method parse_heading
+
+
+   ## add alternate names/aliases
+   class << self
+    alias_method :[],      :find    ### keep shortcut - why? why not?
+    alias_method :heading, :parse_heading
+   end
+
+
+# open question - what name to use build or  parse_line or ?
+#                              or   parse_recs for CountryReader?
+#          remove CountryReader helper methods - why? why not?
+#   use parse_heading/heading for now !!!
+#
+#   def self.parse( line )  or build( line ) ??
+#      SportDb::Import.world.countries.parse( line )
+#   end
+#
+# !!!! note - conflict with
+#     def self.read( path )  CountryReader.read( path ); end
+#     def self.parse( txt )  CountryReader.parse( txt ); end
+#
+end # class Country
+end  # module Sports

data/lib/sportdb/search/version.rb

@@ -1,9 +1,8 @@
 module SportDb
 module Module
 module Search
-
   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/search.rb

@@ -7,22 +7,17 @@ require 'sportdb/catalogs'
 ###  shared basics for search
 class SportSearch
   class Search    ## base search service - use/keep - why? why not?
+    attr_reader :service
     def initialize( service ) @service = service; end
   end  # class Search
 end
 
-class WorldSearch
-  class Search    ## base search service - use/keep - why? why not?
-    def initialize( service ) @service = service; end
-  end  # class Search
-end
 
 
 
 ## our own code
 require_relative 'search/version'
 require_relative 'search/sport'
-require_relative 'search/world'
 
 
 ########
@@ -76,8 +71,9 @@ class WorldSearch
     def initialize( countries:, cities: )
         ## change service to country_service or such - why? why not?
         ##  add city_service and such later
-        @countries = CountrySearch.new( countries )
-        @cities    = CitySearch.new( cities )
+
+        @countries =  countries
+        @cities    =  cities
     end
 
     ####
@@ -168,35 +164,12 @@ end   # module Sports
 ###
 ##   add/augment core classes with search services
 require_relative 'search/structs'
+require_relative 'search/structs_world'
 
 
 
 module SportDb
   module Import
-    Season = ::Season   ## add a convenience alias for top-level Season class
-
-    ## add "old" convenience aliases for structs - why? why not?
-    ##   todo/check: just use include Sports !!!!
-    Country      = ::Sports::Country
-    League       = ::Sports::League
-    Group        = ::Sports::Group
-    Round        = ::Sports::Round
-    Match        = ::Sports::Match
-    Matchlist    = ::Sports::Matchlist
-    Goal         = ::Sports::Goal
-    Team         = ::Sports::Team
-    NationalTeam = ::Sports::NationalTeam
-    Club         = ::Sports::Club
-    Standings    = ::Sports::Standings
-    TeamUsage    = ::Sports::TeamUsage
-
-    Ground       = ::Sports::Ground
-
-    Player       = ::Sports::Player
-
-    EventInfo    = ::Sports::EventInfo
-
-
     class Team
       ## add convenience lookup helper / method for name by season for now
       ##   use clubs history - for now kept separate from struct - why? why not?
@@ -205,7 +178,6 @@ module SportDb
         SportDb::Import.catalog.clubs_history.find_name_by( name: name, season: season ) || name
       end
     end  # class Team
-
   end   # module Import
 end     # module SportDb
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: sportdb-search
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.1.1
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-08-25 00:00:00.000000000 Z
+date: 2024-09-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: sportdb-catalogs
@@ -74,8 +74,8 @@ files:
 - lib/sportdb/search.rb
 - lib/sportdb/search/sport.rb
 - lib/sportdb/search/structs.rb
+- lib/sportdb/search/structs_world.rb
 - lib/sportdb/search/version.rb
-- lib/sportdb/search/world.rb
 homepage: https://github.com/sportdb/sport.db
 licenses:
 - Public Domain

data/lib/sportdb/search/world.rb

@@ -1,127 +0,0 @@
-###
-#  world search service api for countries and more
-#
-# core api is:
-# -  world.countries.find_by_code
-# -                 .find_by_name
-
-
-class WorldSearch
-
-class CitySearch < Search
-  ###################
-  ## core required delegates  - use delegate generator - why? why not?
-  def match_by( name: )
-    @service.match_by( name: name )
-  end
-end  # class CitySearch
-
-
-class CountrySearch < Search
-    ###################
-    ## core required delegates  - use delegate generator - why? why not?
-    def find_by_code( code )
-      puts "!! DEPRECATED - use CountrySearch#find_by( code: )"
-      @service.find_by_code( code )
-    end
-    def find_by_name( name )
-      puts "!! DEPRECATED - use CountrySearch#find_by( name: )"
-      @service.find_by_name( name )
-    end
-
-    def find_by( code: nil, name: nil )
-      ## todo/fix upstream - change to find_by( code:, name:, ) too
-      if code && name.nil?
-        @service.find_by_code( code )
-      elsif name && code.nil?
-        @service.find_by_name( name )
-      else
-        raise ArgumentError, "CountrySearch#find_by - one (and only one arg) required - code: or name:"
-      end
-    end
-
-    def find( q )
-      @service.find_by_name_or_code( q )
-    end
-    alias_method :[], :find    ### keep shortcut - why? why not?
-
-
-    ###############
-    ### more deriv support functions / helpers
-
- ###
- ##   split/parse country line
- ##
- ##  split on bullet e.g.
- ##   split into name and code with regex - make code optional
- ##
- ##  Examples:
- ##    Österreich • Austria (at)
- ##    Österreich • Austria
- ##    Austria
- ##    Deutschland (de) • Germany
- ##
- ##   todo/check: support more formats - why? why not?
- ##       e.g.  Austria, AUT  (e.g. with comma - why? why not?)
- def parse( line )
-   values = line.split( '•' )   ## use/support multi-lingual separator
-   country = nil
-   values.each do |value|
-      value = value.strip
-      ## check for trailing country code e.g. (at), (eng), etc
-      ##   allow code 1 to 5 for now - northern cyprus(fifa) with 5 letters?.
-      ##     add/allow  gb-eng, gb-wal (official iso2!!), in the future too - why? why not?
-      if value =~ /[ ]+\((?<code>[A-Za-z]{1,5})\)$/  ## e.g. Austria (at)
-        code =  $~[:code]
-        name = value[0...(value.size-code.size-2)].strip  ## note: add -2 for brackets
-        candidates = [ find_by( code: code ), find_by( name: name ) ]
-        if candidates[0].nil?
-          puts "** !!! ERROR Country.parse_heading - unknown code >#{code}< in line: #{line}"
-          pp line
-          exit 1
-        end
-        if candidates[1].nil?
-          puts "** !!! ERROR Country.parse_heading - unknown name >#{code}< in line: #{line}"
-          pp line
-          exit 1
-        end
-        if candidates[0] != candidates[1]
-          puts "** !!! ERROR Country.parse_heading - name and code do NOT match the same country:"
-          pp line
-          pp candidates
-          exit 1
-        end
-        if country && country != candidates[0]
-          puts "** !!! ERROR Country.parse_heading - names do NOT match the same country:"
-          pp line
-          pp country
-          pp candidates
-          exit 1
-        end
-        country = candidates[0]
-      else
-        ## just assume value is name or code
-        candidate = find( value )
-        if candidate.nil?
-          puts "** !!! ERROR Country.parse_heading - unknown name or code >#{value}< in line: #{line}"
-          pp line
-          exit 1
-        end
-        if country && country != candidate
-          puts "** !!! ERROR Country.parse_heading - names do NOT match the same country:"
-          pp line
-          pp country
-          pp candidate
-          exit 1
-        end
-        country = candidate
-      end
-   end
-   country
- end # method parse
-end  # class CountrySearch
-end  # class WorldSearch
-
-
-
-