sportdb-parser 0.6.7 → 0.6.9

data/lib/sportdb/parser/racc_tree.rb

@@ -13,6 +13,39 @@ RefereeLine = Struct.new( :name, :country ) do
   end
 end
 
+AttendanceLine = Struct.new( :att ) do 
+  def pretty_print( printer )
+    printer.text( "<AttendanceLine #{self.att}>" )
+  end
+end
+
+
+
+PenaltiesLine = Struct.new( :penalties ) do   
+  def pretty_print( printer )
+    printer.text( "<PenaltiesLine " )
+    printer.text( self.penalties.pretty_inspect )
+    printer.text( ">" )
+  end
+end
+
+Penalty = Struct.new( :name, :score, :note ) do
+  def to_s
+    buf = String.new
+    buf << "#{self.score} "   if self.score
+    buf <<  self.name
+    buf << " (#{self.note})"    if self.note 
+    buf
+  end
+
+  def pretty_print( printer )
+    printer.text( to_s )
+  end  
+end
+
+
+
+
 ##  find a better name for player (use bookings?) - note - red/yellow card for trainer possible
 CardsLine = Struct.new( :type, :bookings ) do   
   def pretty_print( printer )

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

@@ -30,30 +30,30 @@ PROP_NAME_RE = %r{
                               (?:
                                 (?<![ ])  ## use negative lookbehind                             
                                   [ ] 
-                                (?=\p{L}|')      ## use lookahead        
+                                (?=\p{L}|['"])      ## use lookahead        
                               )
-                                  |                         
+                              ## support (inline) quoted name e.g. "Rodri" or such
+                                  |
+                                  (?:
+                                     (?<=[ ])  ## use positive lookbehind                             
+                                     " \p{L}+ " 
+                                      ## require space here too - why? why not?
+                                   )                      
+                                  |   
                              (?:
                                 (?<=\p{L})   ## use lookbehind
-                                 ['-]   ## must be surrounded by letters
+                                 [-]   ## must be surrounded by letters
                                        ## e.g. One/Two NOT
                                        ##      One/ Two or One / Two or One /Two etc.
                                 (?=\p{L})      ## use lookahead        
                               )
                                  |   
-                              (?:
-                                (?<=[ ])   ## use lookbehind  -- add letter (plus dot) or such - why? why not?
-                                 [']   ## must be surrounded by leading space and
-                                       ## traling letters  (e.g. UDI 'Beter Bed)
-                                (?=\p{L})      ## use lookahead        
-                              )   
-                                 |
-                              (?:
-                                (?<=\p{L})   ## use lookbehind
-                                 [']   ## must be surrounded by leading letter and
-                                       ## trailing space PLUS letter  (e.g. UDI' Beter Bed)
-                                (?=[ ]\p{L})      ## use lookahead (space WITH letter         
-                              )   
+                              (?:  ## flex rule for quote - allow any
+                                    ##  only check for double quotes e.g. cannot follow other ' for now - why? why not?
+                                    ##        allows  rodrigez 'rodri' for example
+                                (?<!')  ## use negative lookbehind                             
+                                   '         
+                              )      
                                  |   ## standard case with letter(s) and optinal dot
                               (?: \p{L}+
                                     \.?  ## optional dot
@@ -148,6 +148,39 @@ PROP_NAME_RE = %r{
                    )
                  }ix
 
+
+PROP_NUM_RE = %r{
+             \b
+              (?<num>
+                    ## note allow underscore inline or space e.g.
+                    ##  5_000
+                    ##  allow space inline (e.g. 5 000) - why? why not?
+                  (?<value> [1-9]
+                            (?: _? 
+                                [0-9]+
+                             )* 
+                  )
+              )
+             \b
+            }ix
+
+### todo/fix - allow more chars in enclosed name  - why? why not?
+##                     e.g.  (') - Cote D'Ivore etc.
+##  change to PAREN_NAME or PARENTHESIS or such - why? why not?
+ENCLOSED_NAME_RE = %r{ 
+                 (?<enclosed_name>  
+                    \( 
+                   (?<name>   
+                       \p{L}+
+                       (?:
+                          [ ] 
+                            \p{L}+ 
+                       )*
+                   )
+                     \)
+                 )
+         }ix
+
                  
 
 PROP_BASICS_RE = %r{
@@ -160,22 +193,47 @@ PROP_BASICS_RE = %r{
 }ix
 
 PROP_RE = Regexp.union(
-   PROP_BASICS_RE, 
    MINUTE_RE,
    PROP_KEY_INLINE_RE,
    PROP_NAME_RE,
+   PROP_BASICS_RE, 
    ## todo/fix - add ANY_RE here too!!!
 )
 
 ## note - no inline keys possible
 ##         todo/fix - use custom (limited) prop basics too
 PROP_CARDS_RE =  Regexp.union(
-   PROP_BASICS_RE, 
    MINUTE_RE,
    PROP_NAME_RE,
+   PROP_BASICS_RE, 
    ## todo/fix - add ANY_RE here too!!!
 ) 
 
+
+PROP_PENALTIES_RE = Regexp.union(
+   SCORE_RE,               # e.g. 1-1 etc.
+   ENCLOSED_NAME_RE,       # e.g. (save), (post), etc.
+   PROP_NAME_RE,
+   PROP_BASICS_RE, 
+   ## todo/fix - add ANY_RE here too!!!
+) 
+
+
+PROP_REFEREE_RE = Regexp.union(
+   ENCLOSED_NAME_RE,       # e.g. (sold out) etc.  why? why not?
+   PROP_NUM_RE,                 # e.g. 28 000 or 28_000  (NOT 28,000 is not valid!!!)
+   PROP_KEY_INLINE_RE,
+   PROP_NAME_RE,
+   PROP_BASICS_RE, 
+   ## todo/fix - add ANY_RE here too!!!
+)  
+
+PROP_ATTENDANCE_RE  = Regexp.union(
+   ENCLOSED_NAME_RE,       # e.g. (sold out) etc.  why? why not?
+   PROP_NUM_RE,                 # e.g. 28 000 or 28_000  (NOT 28,000 is not valid!!!)
+   PROP_BASICS_RE, 
+   ## todo/fix - add ANY_RE here too!!!
+)  
     
 end  # class Lexer
 end  # module SportDb

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

@@ -56,8 +56,12 @@ TEXT_RE = %r{
                        \p{L}+
                   |
                 ## opt 3 - add weirdo case
+                ##   e.g.  1/8 Finals  1/4 1/2 ...
+                    1/ \d{1,2} [ ] \p{L}+
+                  |
+                ## opt 4 - add another weirdo case
                 ##   e.g. 5.-8. Platz Playoffs  - keep - why? why not?
-                    \d+\.-\d+\.  [ ]? \p{L}+
+                    \d+\.-\d+\.  [ ]? \p{L}+                 
                )
 
               (?:(?:  (?:[ ] 

data/lib/sportdb/parser/version.rb

@@ -4,7 +4,7 @@ module SportDb
     module Parser
   MAJOR = 0    ## todo: namespace inside version or something - why? why not??
   MINOR = 6
-  PATCH = 7
+  PATCH = 9
   VERSION = [MAJOR,MINOR,PATCH].join('.')
 
   def self.version

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: sportdb-parser
 version: !ruby/object:Gem::Version
-  version: 0.6.7
+  version: 0.6.9
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2025-03-01 00:00:00.000000000 Z
+date: 2025-03-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: cocos