sportdb-parser 0.3.9 → 0.5.0

data/lib/sportdb/parser/token-text.rb

@@ -54,7 +54,7 @@ TEXT_RE = %r{
                )
 
               (?:(?:  (?:[ ]
-                     (?!vs?\.?[ ])    ## note - exclude (v[ ]/vs[ ]/v.[ ]/vs.[ ])
+                        (?!vs?[ ])    ## note - exclude (v[ ]/vs[ ])
                        )
                       |     # only single spaces allowed inline!!!
                      [-]

data/lib/sportdb/parser/token.rb

@@ -68,24 +68,41 @@ BASICS_RE = %r{
     (?<vs>
        (?<=[ ])	# Positive lookbehind for space
        (?:
-          vs\.?|   ## allow optional dot (eg. vs. v.)
-          v\.?|
-          -
-       )   # not bigger match first e.g. vs than v etc.
+          vs|v
+       )  
+           # not bigger match first e.g. vs than v etc.
+           # todo/fix - make vs|v case sensitive!!! only match v/vs - why? why not?
        (?=[ ])   # positive lookahead for space
     )
        |
+    (?<spaces> [ ]{2,}) |
+    (?<space>  [ ])
+        |
+    (?<sym>[;,@|\[\]-])
+}ix
+
+
+## removed from basics
+=begin
     (?<none>
        (?<=[ \[]|^)	 # Positive lookbehind for space or [
            -
         (?=[ ]*;)   # positive lookahead for space
     )
        |
-    (?<spaces> [ ]{2,}) |
-    (?<space>  [ ])
-        |
-    (?<sym>[;,@|\[\]])
-}ix
+   (?<vs>
+       (?<=[ ])	# Positive lookbehind for space
+       (?:
+          vs\.?|   ## allow optional dot (eg. vs. v.)
+          v\.?|
+          -
+       )   # not bigger match first e.g. vs than v etc.
+       (?=[ ])   # positive lookahead for space
+    )
+       |
+ 
+    make - into a simple symbol !!!
+=end
 
 
 MINUTE_RE = %r{
@@ -141,8 +158,7 @@ end
 
 
 
-def tokenize_with_errors( line, typed: false,
-                                debug: false )
+def tokenize_with_errors( line, debug: false )
   tokens = []
   errors = []   ## keep a list of errors - why? why not?
 
@@ -180,6 +196,10 @@ def tokenize_with_errors( line, typed: false,
 
     pp offsets   if debug
 
+    ##
+    ## note: racc requires pairs e.g. [:TOKEN, VAL]
+    ##         for VAL use "text" or ["text", { opts }]  array
+
     t = if m[:space]
            ## skip space
            nil
@@ -187,15 +207,17 @@ def tokenize_with_errors( line, typed: false,
            ## skip spaces
            nil
         elsif m[:text]
-          [:text, m[:text]]   ## keep pos - why? why not?
+          [:TEXT, m[:text]]   ## keep pos - why? why not?
         elsif m[:status]   ## (match) status e.g. cancelled, awarded, etc.
+          ## todo/check - add text (or status) 
+          #     to opts hash {} by default (for value)
           if m[:status_note]   ## includes note? e.g.  awarded; originally 2-0
-             [:status, m[:status], {note:m[:status_note]}]
+             [:STATUS, [m[:status], {status: m[:status], 
+                                     note:   m[:status_note]} ]]
           else
-             [:status, m[:status]]
+             [:STATUS, [m[:status], {status: m[:status] } ]]
           end
         elsif m[:time]
-          if typed
               ## unify to iso-format
               ###   12.40 => 12:40
               ##    12h40 => 12:40 etc.
@@ -208,15 +230,11 @@ def tokenize_with_errors( line, typed: false,
                  (minute >=0 && minute <= 59)
                ## note - for debugging keep (pass along) "literal" time
                ##   might use/add support for am/pm later
-               [:time, m[:time], {h:hour,m:minute}]
+               [:TIME, [m[:time], {h:hour,m:minute}]]
               else
                  raise ArgumentError, "parse error - time >#{m[:time]}< out-of-range"
               end
-          else
-            [:time, m[:time]]
-          end
         elsif m[:date]
-          if typed
             date = {}
 =begin
             ((?<day_name>#{DAY_NAMES})
@@ -237,14 +255,11 @@ def tokenize_with_errors( line, typed: false,
             date[:d]  = m[:day].to_i(10)   if m[:day]
             date[:wday] = DAY_MAP[ m[:day_name].downcase ]   if m[:day_name]
             ## note - for debugging keep (pass along) "literal" date
-            [:date, m[:date], date]
-          else
-            [:date, m[:date]]
-          end
+            [:DATE, [m[:date], date]]
         elsif m[:timezone]
-          [:timezone, m[:timezone]]
+          [:TIMEZONE, m[:timezone]]
         elsif m[:duration]
-          if typed
+            ## todo/check/fix - if end: works for kwargs!!!!!
             duration = { start: {}, end: {}}
             duration[:start][:y] = m[:year1].to_i(10)  if m[:year1]
             duration[:start][:m] = MONTH_MAP[ m[:month_name1].downcase ]   if m[:month_name1]
@@ -255,19 +270,11 @@ def tokenize_with_errors( line, typed: false,
             duration[:end][:d]  = m[:day2].to_i(10)   if m[:day2]
             duration[:end][:wday] = DAY_MAP[ m[:day_name2].downcase ]   if m[:day_name2]
             ## note - for debugging keep (pass along) "literal" duration
-            [:duration, m[:duration], duration]
-          else
-            [:duration, m[:duration]]
-          end
-        elsif m[:num]
-          if typed
+            [:DURATION, [m[:duration], duration]]
+        elsif m[:num]   ## fix - change to ord (for ordinal number!!!)
               ## note -  strip enclosing () and convert to integer
-             [:num, m[:value].to_i(10)]
-          else
-             [:num, m[:num]]
-          end
+             [:ORD, [m[:num], { value: m[:value].to_i(10) } ]]
         elsif m[:score]
-          if typed
               score = {}
               ## check for pen
               score[:p] = [m[:p1].to_i(10),
@@ -280,42 +287,39 @@ def tokenize_with_errors( line, typed: false,
                             m[:ht2].to_i(10)]  if m[:ht1] && m[:ht2]
 
             ## note - for debugging keep (pass along) "literal" score
-            [:score, m[:score], score]
-          else
-            [:score, m[:score]]
-          end
+            [:SCORE, [m[:score], score]]
         elsif m[:minute]
-          if typed
               minute = {}
               minute[:m]      = m[:value].to_i(10)
               minute[:offset] = m[:value2].to_i(10)   if m[:value2]
              ## note - for debugging keep (pass along) "literal" minute
-             [:minute, m[:minute], minute]
-          else
-             [:minute, m[:minute]]
-          end
+             [:MINUTE, [m[:minute], minute]]
         elsif m[:og]
-          typed  ?  [:og] : [:og, m[:og]]    ## for typed drop - string version/variants
+           [:OG, m[:og]]    ## for typed drop - string version/variants ??  why? why not?
         elsif m[:pen]
-          typed  ?  [:pen] : [:pen, m[:pen]]
+           [:PEN, m[:pen]]
         elsif m[:vs]
-          typed  ?  [:vs] : [:vs, m[:vs]]
-        elsif m[:none]
-          typed  ?  [:none] : [:none, m[:none]]
+           [:VS, m[:vs]]
         elsif m[:sym]
           sym = m[:sym]
           ## return symbols "inline" as is - why? why not?
+          ## (?<sym>[;,@|\[\]-])
+ 
           case sym
           when ',' then [:',']
           when ';' then [:';']
           when '@' then [:'@']
           when '|' then [:'|']
+          when '[' then [:'[']
+          when ']' then [:']']
+          when '-' then [:'-']
           else
             nil  ## ignore others (e.g. brackets [])
           end
         else
           ## report error
-          nil
+           puts "!!! TOKENIZE ERROR - no match found"
+           nil 
         end
 
     tokens << t    if t
@@ -342,10 +346,8 @@ end
 
 
 ### convience helper - ignore errors by default
-def tokenize(  line, typed: false,
-                     debug: false )
-   tokens, _ = tokenize_with_errors( line, typed: typed,
-                                           debug: debug )
+def tokenize(  line, debug: false )
+   tokens, _ = tokenize_with_errors( line, debug: debug )
    tokens
 end
 

data/lib/sportdb/parser/version.rb

@@ -3,8 +3,8 @@ module SportDb
   module Module
     module Parser
   MAJOR = 0    ## todo: namespace inside version or something - why? why not??
-  MINOR = 3
-  PATCH = 9
+  MINOR = 5
+  PATCH = 0
   VERSION = [MAJOR,MINOR,PATCH].join('.')
 
   def self.version

data/lib/sportdb/parser.rb

@@ -3,9 +3,6 @@ require 'cocos'
 require 'season/formats'  # e.g. Season() support machinery
 
 
-## more stdlibs
-require 'optparse'   ## check - already auto-required in cocos? keep? why? why not?
-
 
 
 ####
@@ -35,10 +32,6 @@ require_relative 'parser/parser'
 require_relative 'parser/outline_reader'
 
 
-require_relative 'parser/opts'
-require_relative 'parser/linter'
-require_relative 'parser/fbtok/main'
-
 
 ###
 #  make parser api (easily) available - why? why not?
@@ -54,5 +47,243 @@ end  # module SportDb
 =end
 
 
+
+module SportDb
+class Tokenizer  
+    
+   attr_reader :tokens
+ 
+   def initialize( txt )
+      parser = Parser.new
+ 
+      tree = []
+      
+      lines = txt.split( "\n" )
+      lines.each_with_index do |line,i|
+          next if line.strip.empty? || line.strip.start_with?( '#' )
+      
+         puts "line >#{line}<"
+         tokens = parser.tokenize( line )
+         pp tokens
+      
+         tree << tokens
+      end
+ 
+
+=begin   
+      ## quick hack
+      ##   turn all  text tokens followed by minute token
+      ##     into player tokens!!!
+      ##
+      ##   also auto-convert text tokens into team tokens - why? why not?
+      tree.each do |tokens|
+         tokens.each_with_index do |t0,idx|
+            t1 = tokens[idx+1]
+            if t1 && t1[0] == :minute && t0[0] == :text
+                 t0[0] = :player 
+            end
+         end
+      end
+=end
+
+=begin
+## auto-add/insert start tokens for known line patterns
+##    START_GOALS for  goals_line
+##    why? why not?
+=end
+
+      ## flatten
+      @tokens = []
+      tree.each do |tokens|
+         @tokens += tokens 
+         @tokens  << [:NEWLINE, "\n"]   ## auto-add newlines
+      end
+ 
+
+      ## convert to racc format
+      @tokens = @tokens.map do |tok|
+           if tok.size == 1
+             [tok[0].to_s, tok[0].to_s]
+           elsif tok.size == 2
+ #############
+ ## pass 1
+ ##   replace all texts with keyword matches (e.g. group, round, leg, etc.)
+               if tok[0] == :TEXT
+                  text = tok[1]
+                  tok = if parser.is_group?( text )
+                          [:GROUP, text]
+                        elsif parser.is_round?( text ) || parser.is_leg?( text )
+                          [:ROUND, text]
+                        else
+                          tok  ## pass through as-is (1:1)
+                        end
+               end
+ ## pass 2
+              tok
+       else
+              raise ArgumentError, "tokens of size 1|2 expected; got #{tok.pretty_inspect}"
+           end
+      end
+   end
+ 
+ 
+
+   def next_token
+      @tokens.shift
+   end
+ end  # class Tokenizer
+end # module SportDb
+
+
+
+####
+#   RaccMatchParser support machinery (incl. node classes/abstract syntax tree)
+
+class RaccMatchParser
+
+GroupDef   = Struct.new( :name, :teams ) do
+  def pretty_print( printer )
+    printer.text( "<GroupDef " )
+    printer.text( self.name )
+    printer.text( " teams=" + self.teams.pretty_inspect )
+    printer.text( ">" )
+  end
+end
+
+
+RoundDef   = Struct.new( :name, :date, :duration )  do
+  def pretty_print( printer )
+    printer.text( "<RoundDef " )
+    printer.text( self.name )
+    printer.text( " date=" + self.date.pretty_inspect ) if date
+    printer.text( " durattion=" + self.duration.pretty_inspect ) if duration
+    printer.text( ">" )
+  end
+end
+
+DateHeader = Struct.new( :date ) do
+  def pretty_print( printer )
+    printer.text( "<DateHeader " )
+    printer.text( "#{self.date.pretty_inspect}>" )
+  end
+end
+
+GroupHeader = Struct.new( :name ) do
+  def pretty_print( printer )
+    printer.text( "<GroupHeader " )
+    printer.text( "#{self.name}>" )
+  end
+end
+
+RoundHeader = Struct.new( :names ) do
+  def pretty_print( printer )
+    printer.text( "<RoundHeader " )
+    printer.text( "#{self.names.join(', ')}>" )
+  end
+end
+
+MatchLine   = Struct.new( :ord, :date, :time,
+                          :team1, :team2, :score, 
+                          :geo )  do   ## change to geos - why? why not?
+
+  def pretty_print( printer )
+    printer.text( "<MatchLine " )
+    printer.text( "#{self.team1} v #{self.team2}")
+    printer.breakable
+
+    members.zip(values) do |name, value|
+      next if [:team1, :team2].include?( name )
+      next if value.nil?
+      
+      printer.text( "#{name}=#{value.pretty_inspect}" )
+    end    
+
+    printer.text( ">" )
+  end  
+
+end
+
+GoalLine    = Struct.new( :goals1, :goals2 ) do
+  def pretty_print( printer )
+    printer.text( "<GoalLine " )
+    printer.text( "goals1=" + self.goals1.pretty_inspect + "," )
+    printer.breakable
+    printer.text( "goals2=" + self.goals2.pretty_inspect + ">" )
+  end  
+end
+
+Goal        = Struct.new( :player, :minutes ) do
+  def to_s
+    buf = String.new
+    buf << "#{self.player}"
+    buf << " "
+    buf << minutes.map { |min| min.to_s }.join(' ')
+    buf
+  end
+
+  def pretty_print( printer )
+    printer.text( to_s )
+  end  
+
+end
+
+Minute      = Struct.new( :m, :offset, :og, :pen )  do
+    def to_s
+      buf = String.new
+      buf << "#{self.m}"
+      buf << "+#{self.offset}"  if self.offset 
+      buf << "'"
+      buf << "(og)"   if self.og
+      buf << "(pen)"  if self.pen
+      buf
+    end
+ 
+    def pretty_print( printer ) 
+       printer.text( to_s ) 
+    end  
+end
+
+
+
+
+def initialize(input)
+    puts "==> input:"
+    puts input
+    @tokenizer = SportDb::Tokenizer.new(input)
+  end
+  
+
+  def next_token
+    tok = @tokenizer.next_token
+    puts "next_token => #{tok.pretty_inspect}"
+    tok
+  end
+  
+#  on_error do |error_token_id, error_value, value_stack|
+#      puts "Parse error on token: #{error_token_id}, value: #{error_value}"
+#  end  
+
+  def parse    
+     puts "parse:"
+     @tree = [] 
+     do_parse
+     @tree
+  end
+
+
+  def on_error(*args)
+    puts "!! on error:"
+    puts "args=#{args.pretty_inspect}"
+  end
+
+=begin
+on_error do |error_token_id, error_value, value_stack|
+    puts "Parse error on token: #{error_token_id}, value: #{error_value}"
+end
+=end
+
+end 
+
+
 puts SportDb::Module::Parser.banner    # say hello
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: sportdb-parser
 version: !ruby/object:Gem::Version
-  version: 0.3.9
+  version: 0.5.0
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-10-13 00:00:00.000000000 Z
+date: 2025-01-15 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: cocos
@@ -38,6 +38,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: racc
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rdoc
   requirement: !ruby/object:Gem::Requirement
@@ -64,18 +78,17 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.1'
+        version: '4.2'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.1'
+        version: '4.2'
 description: sportdb-parser - football.txt match parser (& tokenizer)
 email: gerald.bauer@gmail.com
-executables:
-- fbtok
+executables: []
 extensions: []
 extra_rdoc_files:
 - CHANGELOG.md
@@ -91,17 +104,13 @@ files:
 - Manifest.txt
 - README.md
 - Rakefile
-- bin/fbtok
 - config/rounds_de.txt
 - config/rounds_en.txt
 - config/rounds_es.txt
 - config/rounds_misc.txt
 - config/rounds_pt.txt
 - lib/sportdb/parser.rb
-- lib/sportdb/parser/fbtok/main.rb
 - lib/sportdb/parser/lang.rb
-- lib/sportdb/parser/linter.rb
-- lib/sportdb/parser/opts.rb
 - lib/sportdb/parser/outline_reader.rb
 - lib/sportdb/parser/parser.rb
 - lib/sportdb/parser/token-date.rb
@@ -131,7 +140,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.4.10
+rubygems_version: 3.5.22
 signing_key:
 specification_version: 4
 summary: sportdb-parser - football.txt match parser (& tokenizer)

data/bin/fbtok

@@ -1,13 +0,0 @@
-#!/usr/bin/env ruby
-
-## tip: to test run:
-##   ruby -I ./lib bin/fbtok
-
-require 'sportdb/parser'
-
-
-Fbtok.main( ARGV )
-
-
-puts "bye"
-