sportdb-parser 0.1.0 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: e5af84e3a141fc577287c8c788eb27a79bf1fc78ed0c08e80df6004383788b66
-  data.tar.gz: 0e23fca8e4566021eb220d20925f97694fcc5b8b7c165c6ce469b5f08feb9cc1
+  metadata.gz: c94dcd42fc13a7043f6b926ca1d947df3199877693b22e53e4f50b5aa522bf5d
+  data.tar.gz: 33eb689dcfb2bab0728c19b7d706da1556ddefafbfbcc6e424ac5bcbe3bccef6
 SHA512:
-  metadata.gz: b6434b5d4df17e72a83f9b63ceef117ddfe50157073cbe4657e6e47e8aa820e8aaf986030642fc86160fd9c551cc55c9e9a35187cf09de3e0c346a00d1f58f17
-  data.tar.gz: dc9b9fd5c782409c019aa2de0d4aea5bdeb90a7a4e01c83ed58b08d4315f1a87ae84268f269d20a10463567633471e1b05052c24bfa47bc49e988bb927e2f927
+  metadata.gz: 97ef8d76ffa26312d66359f364588af3d7c76a3b0cebd3644b1f1ae775463defa9cb9552b267f26677c2c6f4e9b7b9fe62479dd34a7211fd1a4a3c1b5e9af830
+  data.tar.gz: ca9b56c6c02c132f3924fb40c293e90379812b830a2899e2be02c1d6469a278456c6d68db7f73d5f5fd69b372c958953e3fefd829ac1120cf56b0944176a2b87

data/CHANGELOG.md

@@ -1,4 +1,4 @@
-### 0.1.0
+### 0.2.0
 
 ### 0.0.1 / 2024-07-12
 

data/Rakefile

@@ -21,7 +21,7 @@ Hoe.spec 'sportdb-parser' do
   self.licenses = ['Public Domain']
 
   self.extra_deps = [
-       ['cocos'],
+       ['cocos', '>= 0.4.0'],
        ['season-formats'],
   ]
 

data/bin/fbt

@@ -11,7 +11,7 @@ require 'sportdb/parser'
 require 'optparse'
 
 ##
-## read textfile 
+## read textfile
 ##   and dump tokens
 ##
 ##   fbt  ../openfootball/.../euro.txt
@@ -32,7 +32,7 @@ require 'optparse'
 
 
   parser.on( "--verbose", "--debug",
-               "turn on verbose / debug output (default: #{opts[:debug]} )" ) do |debug|
+               "turn on verbose / debug output (default: #{opts[:debug]})" ) do |debug|
     opts[:debug] = debug
   end
 
@@ -53,9 +53,6 @@ p args
 
 
 
-
-
-
 paths =  if args.empty?
             [
               '../../../openfootball/euro/2020--europe/euro.txt',
@@ -64,7 +61,7 @@ paths =  if args.empty?
          else
             ## check for directories
             ##   and auto-expand
-             
+
             SportDb::Parser::Opts.expand_args( args )
          end
 
@@ -86,8 +83,10 @@ end
 if errors.size > 0
     puts
     pp errors
+    puts
     puts "!!   #{errors.size} parse error(s) in #{paths.size} datafiles(s)"
 else
+    puts
     puts "OK   no parse errors found in #{paths.size} datafile(s)"
 end
 

data/lib/sportdb/parser/lang.rb

@@ -15,7 +15,7 @@ class Parser
 
 GROUP_RE = %r{^
                 Group [ ]
-                   (?<key>[a-z0-9]+)                   
+                   (?<key>[a-z0-9]+)
               $}ix
 def is_group?( text )
    ## use regex for match
@@ -28,42 +28,68 @@ end
 ROUND_RE = %r{^(
 
    # round  - note - requiers number e.g. round 1,2, etc.
+   #   note - use 1-9 regex (cannot start with 0) - why? why not?
+   #             make week 01 or round 01 or matchday 01 possible?
       (?: (?: Round |
               Matchday |
               Week
            )
-           [ ] [0-9]+
+           [ ] [1-9][0-9]*
       )
        |
+   ##  starting with qual(ification)
+   ## Qual. Round 1 / Qual. Round 2 / Qual. Round 3
+     (?:  Qual \. [ ]
+          Round
+           [ ] [1-9][0-9]*
+      )
+       |
+   ## 1. Round / 2. Round / 3. Round / etc.
+   ##  Play-off Round
+      (?:
+           (?: [1-9][0-9]* \.
+                |
+                Play-?off
+           )
+             [ ] Round
+       )
+       |
+  ## starting with preliminary
+     (?:  Preliminary  [ ]
+           (?:  Semi-?finals |
+                Final
+           )
+     )
+     |
    # more (kockout) rounds
    # playoffs  - playoff, play-off, play-offs
-        (?: Play-?offs? 
+        (?: Play-?offs?
            (?: [ ]for[ ]quarter-?finals )?
         )
-        |    
+        |
    # round32
-        (?: Round[ ]of[ ]32 | 
+        (?: Round[ ]of[ ]32 |
             Last[ ]32 )
           |
-   # round16   
+   # round16
         (?: Round[ ]of[ ]16 |
-            Last[ ]16 | 
+            Last[ ]16 |
             8th[ ]finals )
            |
    # fifthplace
          (?:
-             (?: (Fifth|5th)[ -]place 
+             (?: (Fifth|5th)[ -]place
                   (?: [ ] (?: match|play-?off|final ))?
               ) |
              (?: Match[ ]for[ ](?: fifth|5th )[ -]place )
          )
           |
    # thirdplace
-          (?: 
-              (?: (Third|3rd)[ -]place 
-                     (?: [ ] (?: match|play-?off|final ))? 
+          (?:
+              (?: (Third|3rd)[ -]place
+                     (?: [ ] (?: match|play-?off|final ))?
                ) |
-              (?: Match[ ]for[ ](?: third|3rd )[ -]place ) 
+              (?: Match[ ]for[ ](?: third|3rd )[ -]place )
            )
            |
    # quarterfinals
@@ -72,18 +98,24 @@ ROUND_RE = %r{^(
               Quarters |
               Last[ ]8
           )
-          |     
+          |
    # semifinals
-        (?:   
+        (?:
              Semi-?finals? |
              Semis |
              Last[ ]4
         )
         |
    # final
-         Finals? 
-       
-        )$}ix
+         Finals?
+         |
+    ## add replays
+    ##  Final Replay
+     (?:
+        Final
+        [ ] Replay
+      )
+)$}ix
 
 
 def is_round?( text )
@@ -95,9 +127,9 @@ end
 ##
 LEG_RE = %r{^
   # leg1
-     (?: 1st|First)[ ]leg 
+     (?: 1st|First)[ ]leg
      |
-  # leg2 
+  # leg2
      (?: 2nd|Second)[ ]leg
 $}ix
 

data/lib/sportdb/parser/outline_reader.rb

@@ -1,8 +1,4 @@
 
-###
-## todo/fix -  move to sportdb-parser - why? why not? !!!!!!
-##
-
 
 module SportDb
 
@@ -10,7 +6,7 @@ class OutlineReader
 
   def self.debug=(value) @@debug = value; end
   def self.debug?() @@debug ||= false; end
-  def debug?()  self.class.debug?; end   
+  def debug?()  self.class.debug?; end
 
 
 

data/lib/sportdb/parser/parser.rb

@@ -1,24 +1,24 @@
-module SportDb 
+module SportDb
 class Parser
-    
+
 
 ## transforms
 ##
 ##  Netherlands  1-2 (1-1)   England
-##   =>  text => team  
-##       score|vs  
+##   =>  text => team
+##       score|vs
 ##       text => team
 
 
 ## token iter/find better name
 ##  e.g. TokenBuffer/Scanner or such ??
-class Tokens  
+class Tokens
     def initialize( tokens )
         @tokens = tokens
         @pos = 0
     end
 
-    def pos()  @pos; end  
+    def pos()  @pos; end
     def eos?() @pos >= @tokens.size; end
 
 
@@ -47,17 +47,17 @@ class Tokens
     ## return token type  (e.g. :text, :num, etc.)
     def cur()           peek(0); end
     ## return content (assumed to be text)
-    def text(offset=0)  
+    def text(offset=0)
         ## raise error - why? why not?
         ##   return nil?
         if peek( offset ) != :text
             raise ArgumentError, "text(#{offset}) - token not a text type"
         end
-        @tokens[@pos+offset][1]  
+        @tokens[@pos+offset][1]
     end
 
 
-    def peek(offset=1)  
+    def peek(offset=1)
         ## return nil if eos
         if @pos+offset >= @tokens.size
             nil
@@ -66,7 +66,7 @@ class Tokens
         end
     end
 
-    ## note - returns complete token 
+    ## note - returns complete token
     def next
        # if @pos >= @tokens.size
        #     raise ArgumentError, "end of array - #{@pos} >= #{@tokens.size}"
@@ -81,7 +81,7 @@ class Tokens
     def collect( &blk )
         tokens = []
         loop do
-          break if eos?  
+          break if eos?
           tokens <<  if block_given?
                         blk.call( self.next )
                      else
@@ -106,7 +106,7 @@ def parse_with_errors( line, debug: false )
     errors += token_errors
 
 #############
-## pass 1 
+## pass 1
 ##   replace all texts with keyword matches (e.g. group, round, leg, etc.)
      tokens = tokens.map do |t|
                       if t[0] == :text
@@ -129,24 +129,40 @@ def parse_with_errors( line, debug: false )
     ## puts "tokens:"
     ## pp tokens
 
-## transform tokens into (parse tree/ast) nodes    
+## transform tokens into (parse tree/ast) nodes
     nodes = []
-    
+
     buf = Tokens.new( tokens )
     ## pp buf
 
 
-    loop do 
-          if buf.pos == 0   
-            ## check for 
-            ##    group def or round def 
-            if buf.match?( :round, :'|' )    ## assume round def (change round to round_def)
+    loop do
+          break if buf.eos?
+
+          ## simplify - remove separator for round + leg pair
+          ##     e.g.  Round of 16, 1st Leg
+          ##     allow Round of 16 - 1st Leg  too - why? why not?
+          if buf.match?( :round, [:',', :'|',
+                                    :'-',
+                                    :vs,   ### fix - change parser to issue :'-' only for (-) not :vs!!!
+                                    ], :leg )
+                    nodes << [:round, buf.next[1]]
+                    buf.next  ## swallow separator
+                    nodes << [:leg, buf.next[1]]
+                    next
+          end
+
+
+          if buf.pos == 0   ## MUST start line
+            ## check for
+            ##    group def or round def
+            if buf.match?( :round, :'|', [:date, :duration] )    ## assume round def (change round to round_def)
                       nodes << [:round_def, buf.next[1]]
                       buf.next ## swallow pipe
                       nodes += buf.collect
                       break
             end
-            if buf.match?( :group, :'|' )    ## assume group def (change group to group_def)
+            if buf.match?( :group, :'|', :text )    ## assume group def (change group to group_def)
                       nodes << [:group_def, buf.next[1]]
                       buf.next ## swallow pipe
                       ## change all text to team
@@ -154,11 +170,15 @@ def parse_with_errors( line, debug: false )
                                 t[0] == :text ? [:team, t[1]] : t
                                }
                       break
-            end  
+            end
           end
 
 
-          if buf.match?( :text, [:score, :vs], :text )
+          if buf.match?( :text, :'-', :text )  ## hacky? convert "generic" :- to :vs
+             nodes << [:team, buf.next[1]]     ##    keep this rule/option - why? why not?
+             nodes << [:vs]
+             nodes << [:team, buf.next[1]]
+          elsif buf.match?( :text, [:score, :vs], :text )
              nodes << [:team, buf.next[1]]
              nodes << buf.next
              nodes << [:team, buf.next[1]]
@@ -170,14 +190,12 @@ def parse_with_errors( line, debug: false )
                ##   only change text to geo
               nodes += buf.collect  { |t|
                            t[0] == :text ? [:geo, t[1]] : t
-                            } 
+                            }
               break
           else
              ## pass through
              nodes << buf.next
           end
-
-          break if buf.eos?
     end
 
     [nodes,errors]
@@ -192,5 +210,5 @@ end
 
 
 end #  class Parser
-end  # module SportDb 
-    
+end  # module SportDb
+

data/lib/sportdb/parser/token.rb

@@ -1,6 +1,6 @@
 
 
-module SportDb 
+module SportDb
 class Parser
 
 
@@ -15,7 +15,7 @@ TIME_RE = %r{
                  (?: :|\.|h )
               (?<minute>\d{2})
               \b
-    )     
+    )
 }ix
 
 
@@ -28,7 +28,7 @@ TIME_RE = %r{
 # (CEST/UTC+2)  - central european summer time  - daylight saving time (DST).
 # (EET/UTC+1)  - eastern european time
 # (EEST/UTC+2)  - eastern european summer time  - daylight saving time (DST).
-# 
+#
 # UTC+3
 # UTC+4
 # UTC+0
@@ -45,7 +45,7 @@ TIME_RE = %r{
 
 TIMEZONE_RE = %r{
    ## e.g. (UTC-2) or (CEST/UTC-2) etc.
-   (?<timezone>  
+   (?<timezone>
       \(
            ## optional "local" timezone name eg. BRT or CEST etc.
            (?:  [a-z]+
@@ -63,28 +63,28 @@ TIMEZONE_RE = %r{
 
 BASICS_RE = %r{
     ## e.g. (51) or (1) etc.  - limit digits of number???
-    (?<num> \(  (?<value>\d+) \) )    
+    (?<num> \(  (?<value>\d+) \) )
        |
-    (?<vs>   
-       (?<=[ ])	# Positive lookbehind for space	
-       (?: 
+    (?<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
-    ) 
-       | 
+    )
+       |
     (?<none>
-       (?<=[ \[]|^)	 # Positive lookbehind for space or [	
+       (?<=[ \[]|^)	 # Positive lookbehind for space or [
            -
         (?=[ ]*;)   # positive lookahead for space
     )
        |
     (?<spaces> [ ]{2,}) |
-    (?<space>  [ ]) 
+    (?<space>  [ ])
         |
-    (?<sym>[;,@|\[\]]) 
+    (?<sym>[;,@|\[\]])
 }ix
 
 
@@ -94,13 +94,13 @@ MINUTE_RE = %r{
            (?<value>\d{1,3})      ## constrain numbers to 0 to 999!!!
         (?: \+
             (?<value2>\d{1,3})
-        )? 	
+        )?
         '     ## must have minute marker!!!!
      )
 }ix
 
 
-##  (match) status 
+##  (match) status
 ##    note: english usage - cancelled (in UK), canceled (in US)
 ##
 ##  add more variants - why? why not?
@@ -115,30 +115,30 @@ STATUS_RE = %r{
                |
             postponed
                |
-            awarded|awd\. 
+            awarded|awd\.
                |
-            replay     
+            replay
          )
    (?=[ \]]|$)
      )}ix
 
 ## todo/check:  remove loakahead assertion here - why require space?
-## note: \b works only after non-alphanum  
-##          to make it work with awd. (dot) "custom" lookahead neeeded 
+## note: \b works only after non-alphanum
+##          to make it work with awd. (dot) "custom" lookahead neeeded
 
 
 ##   goal types
-# (pen.) or (pen) or (p.) or (p)  
+# (pen.) or (pen) or (p.) or (p)
 ## (o.g.) or (og)
 GOAL_PEN_RE = %r{
-   (?<pen> \(  
-           (?:pen|p)\.? 
+   (?<pen> \(
+           (?:pen|p)\.?
            \)
     )
 }ix
 GOAL_OG_RE = %r{
-   (?<og> \(  
-          (?:og|o\.g\.) 
+   (?<og> \(
+          (?:og|o\.g\.)
           \)
    )
 }ix
@@ -158,11 +158,11 @@ RE = Regexp.union(   STATUS_RE,
 
 
 def log( msg )
-   ## append msg to ./logs.txt  
+   ## append msg to ./logs.txt
    ##     use ./errors.txt - why? why not?
    File.open( './logs.txt', 'a:utf-8' ) do |f|
      f.write( msg )
-     f.write( "\n" ) 
+     f.write( "\n" )
    end
 end
 
@@ -176,7 +176,7 @@ def tokenize_with_errors( line, typed: false,
   puts ">#{line}<"    if debug
 
   pos = 0
-  ## track last offsets - to report error on no match 
+  ## track last offsets - to report error on no match
   ##   or no match in end of string
   offsets = [0,0]
   m = nil
@@ -184,7 +184,7 @@ def tokenize_with_errors( line, typed: false,
   while m = RE.match( line, pos )
     if debug
       pp m
-      puts "pos: #{pos}"  
+      puts "pos: #{pos}"
     end
     offsets = [m.begin(0), m.end(0)]
 
@@ -213,10 +213,10 @@ def tokenize_with_errors( line, typed: false,
         elsif m[:spaces]
            ## skip spaces
            nil
-        elsif m[:text] 
+        elsif m[:text]
           [:text, m[:text]]   ## keep pos - why? why not?
         elsif m[:status]   ## (match) status e.g. cancelled, awarded, etc.
-          [:status, m[:status]]   
+          [:status, m[:status]]
         elsif m[:time]
           if typed
               ## unify to iso-format
@@ -230,7 +230,7 @@ def tokenize_with_errors( line, typed: false,
               if (hour >= 0 && hour <= 24) &&
                  (minute >=0 && minute <= 59)
                ## note - for debugging keep (pass along) "literal" time
-               ##   might use/add support for am/pm later                
+               ##   might use/add support for am/pm later
                [:time, m[:time], {h:hour,m:minute}]
               else
                  raise ArgumentError, "parse error - time >#{m[:time]}< out-of-range"
@@ -241,54 +241,68 @@ def tokenize_with_errors( line, typed: false,
         elsif m[:date]
           if typed
             date = {}
-=begin            
+=begin
             ((?<day_name>#{DAY_NAMES})
             [ ]
-       )?    
+       )?
        (?<month_name>#{MONTH_NAMES})
            (?: \/|[ ] )
        (?<day>\d{1,2})
        ## optional year
        (  [ ]
           (?<year>\d{4})
-       )?   
+       )?
 =end
  ## map month names
  ## note - allow any/upcase JULY/JUL etc. thus ALWAYS downcase for lookup
-            date[:y] = m[:year].to_i(10)  if m[:year]       
+            date[:y] = m[:year].to_i(10)  if m[:year]
             date[:m] = MONTH_MAP[ m[:month_name].downcase ]   if m[:month_name]
             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]        
+            ## note - for debugging keep (pass along) "literal" date
+            [:date, m[:date], date]
           else
             [:date, m[:date]]
           end
         elsif m[:timezone]
           [:timezone, m[:timezone]]
         elsif m[:duration]
-          [:duration, m[:duration]]
+          if typed
+            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]
+            duration[:start][:d]  = m[:day1].to_i(10)   if m[:day1]
+            duration[:start][:wday] = DAY_MAP[ m[:day_name1].downcase ]   if m[:day_name1]
+            duration[:end][:y] = m[:year2].to_i(10)  if m[:year2]
+            duration[:end][:m] = MONTH_MAP[ m[:month_name2].downcase ]   if m[:month_name2]
+            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
               ## note -  strip enclosing () and convert to integer
              [:num, m[:value].to_i(10)]
-          else 
+          else
              [:num, m[:num]]
           end
         elsif m[:score]
           if typed
               score = {}
               ## check for pen
-              score[:p] = [m[:p1].to_i(10), 
+              score[:p] = [m[:p1].to_i(10),
                            m[:p2].to_i(10)]  if m[:p1] && m[:p2]
-              score[:et] = [m[:et1].to_i(10), 
+              score[:et] = [m[:et1].to_i(10),
                             m[:et2].to_i(10)]  if m[:et1] && m[:et2]
-              score[:ft] = [m[:ft1].to_i(10), 
+              score[:ft] = [m[:ft1].to_i(10),
                             m[:ft2].to_i(10)]  if m[:ft1] && m[:ft2]
-              score[:ht] = [m[:ht1].to_i(10), 
+              score[:ht] = [m[:ht1].to_i(10),
                             m[:ht2].to_i(10)]  if m[:ht1] && m[:ht2]
 
-            ## note - for debugging keep (pass along) "literal" score                
+            ## note - for debugging keep (pass along) "literal" score
             [:score, m[:score], score]
           else
             [:score, m[:score]]
@@ -298,7 +312,7 @@ def tokenize_with_errors( line, typed: false,
               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                
+             ## note - for debugging keep (pass along) "literal" minute
              [:minute, m[:minute], minute]
           else
              [:minute, m[:minute]]
@@ -318,16 +332,16 @@ def tokenize_with_errors( line, typed: false,
           when ',' then [:',']
           when ';' then [:';']
           when '@' then [:'@']
-          when '|' then [:'|']   
+          when '|' then [:'|']
           else
             nil  ## ignore others (e.g. brackets [])
           end
         else
-          ## report error 
+          ## report error
           nil
         end
 
-    tokens << t    if t    
+    tokens << t    if t
 
     if debug
       print ">"
@@ -346,7 +360,7 @@ def tokenize_with_errors( line, typed: false,
   end
 
 
-  [tokens,errors] 
+  [tokens,errors]
 end
 
 
@@ -360,5 +374,4 @@ end
 
 
 end  # class Parser
-end # module SportDb 
-  
+end # module SportDb

data/lib/sportdb/parser/version.rb

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

data/lib/sportdb/parser.rb

@@ -1,5 +1,5 @@
-## pulls in   
-require 'cocos' 
+## pulls in
+require 'cocos'
 require 'season/formats'  # e.g. Season() support machinery
 
 
@@ -36,7 +36,7 @@ require_relative 'parser/opts'
 =begin
 module SportDb
    def self.parser() @@parser ||= Parser.new; end
-   def self.parse( ... ) 
+   def self.parse( ... )
    end
    def self.tokenize( ... )
    end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: sportdb-parser
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.2.0
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-07-22 00:00:00.000000000 Z
+date: 2024-08-22 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: cocos
@@ -16,14 +16,14 @@ dependencies:
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: '0'
+        version: 0.4.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: '0'
+        version: 0.4.0
 - !ruby/object:Gem::Dependency
   name: season-formats
   requirement: !ruby/object:Gem::Requirement