sportdb-parser 0.7.1 → 0.7.2

data/lib/sportdb/parser.rb

@@ -1,52 +1,80 @@
-## pulls in
 require 'cocos'
 
 
-####
-# try a (simple) tokenizer/parser with regex
 
-## note - match line-by-line
-#            avoid massive backtracking by definition
-#             that is, making it impossible
 
-## sym(bols) -
-##  text - change text to name - why? why not?
+require_relative 'parser/version'
 
 
-require_relative 'parser/version'
+
+## core machinery
 
 ##
-## generic helper
-require_relative 'parser/blocktxt'
+## add shared/most basic regexes here
+## todo - use ANY_RE  to token_commons or such - for shared by many?
+module SportDb
+class Lexer
+
+## general catch-all  (RECOMMENDED (ALWAYS) use as last entry in union)
+##   to avoid advance of pos match!!!
+ANY_RE = %r{
+               (?<any> .)
+          }ix
+
+SPACES_RE = %r{
+                  (?<spaces> [ ]{2,})
+                | (?<space>  [ ])
+             }ix
+
+
+end # class Lexer
+end # module SportDb
 
-## core machinery
 
 require_relative 'parser/token-score'
+require_relative 'parser/token-score_full'
 require_relative 'parser/token-score_fuller'
 require_relative 'parser/token-score_legs'
+require_relative 'parser/token-score--helpers'
+
 require_relative 'parser/token-time'
+require_relative 'parser/token-date--names'
 require_relative 'parser/token-date'
 require_relative 'parser/token-date_duration'
+require_relative 'parser/token-date--helpers'
+
 require_relative 'parser/token-text'
+require_relative 'parser/token-prop'    ## team prop(erty) mode (note - must be before token)
 require_relative 'parser/token-prop_name'    ## a.k.a token-text_ii
 require_relative 'parser/token-status'
+require_relative 'parser/token-status_inline'
 require_relative 'parser/token-note'
 require_relative 'parser/token-goals'
-require_relative 'parser/token-prop'    ## team prop(erty) mode (note - must be before token)
+require_relative 'parser/token-goals--helpers'
 require_relative 'parser/token-geo'
 require_relative 'parser/token-group'
 require_relative 'parser/token-round'
-require_relative 'parser/token-table'
 require_relative 'parser/token'
 
 
-### add token ("private") parse helpers  e.g.  _parse_team() etc.
-require_relative 'parser/token_helpers'
+
 
 
 require_relative 'parser/lexer_buffer'   ## incl. Tokens (aka TokenBuffer)
+
+require_relative 'parser/lexer-prep_doc'
+require_relative 'parser/lexer-prep_line'
+
+require_relative 'parser/lexer_token'
+require_relative 'parser/lexer-tokenize'
+require_relative 'parser/lexer-on_round_def'
+require_relative 'parser/lexer-on_group_def'
+require_relative 'parser/lexer-on_prop_misc'
+require_relative 'parser/lexer-on_prop_lineup'
+require_relative 'parser/lexer-on_prop_penalties'
+require_relative 'parser/lexer-on_goal'
+require_relative 'parser/lexer-on_top'
 require_relative 'parser/lexer'
-require_relative 'parser/lexer_tty'      ## teletype (tty) mode
 
 
 ## note - use "embeded" racc parser runtime

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: sportdb-parser
 version: !ruby/object:Gem::Version
-  version: 0.7.1
+  version: 0.7.2
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-05-22 00:00:00.000000000 Z
+date: 2026-06-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: cocos
@@ -72,32 +72,45 @@ files:
 - README.md
 - Rakefile
 - lib/sportdb/parser.rb
-- lib/sportdb/parser/blocktxt.rb
+- lib/sportdb/parser/lexer-on_goal.rb
+- lib/sportdb/parser/lexer-on_group_def.rb
+- lib/sportdb/parser/lexer-on_prop_lineup.rb
+- lib/sportdb/parser/lexer-on_prop_misc.rb
+- lib/sportdb/parser/lexer-on_prop_penalties.rb
+- lib/sportdb/parser/lexer-on_round_def.rb
+- lib/sportdb/parser/lexer-on_top.rb
+- lib/sportdb/parser/lexer-prep_doc.rb
+- lib/sportdb/parser/lexer-prep_line.rb
+- lib/sportdb/parser/lexer-tokenize.rb
 - lib/sportdb/parser/lexer.rb
 - lib/sportdb/parser/lexer_buffer.rb
-- lib/sportdb/parser/lexer_tty.rb
+- lib/sportdb/parser/lexer_token.rb
 - lib/sportdb/parser/parser.rb
 - lib/sportdb/parser/parser_runtime.rb
 - lib/sportdb/parser/racc_parser.rb
 - lib/sportdb/parser/racc_tree.rb
+- lib/sportdb/parser/token-date--helpers.rb
+- lib/sportdb/parser/token-date--names.rb
 - lib/sportdb/parser/token-date.rb
 - lib/sportdb/parser/token-date_duration.rb
 - lib/sportdb/parser/token-geo.rb
+- lib/sportdb/parser/token-goals--helpers.rb
 - lib/sportdb/parser/token-goals.rb
 - lib/sportdb/parser/token-group.rb
 - lib/sportdb/parser/token-note.rb
 - lib/sportdb/parser/token-prop.rb
 - lib/sportdb/parser/token-prop_name.rb
 - lib/sportdb/parser/token-round.rb
+- lib/sportdb/parser/token-score--helpers.rb
 - lib/sportdb/parser/token-score.rb
+- lib/sportdb/parser/token-score_full.rb
 - lib/sportdb/parser/token-score_fuller.rb
 - lib/sportdb/parser/token-score_legs.rb
 - lib/sportdb/parser/token-status.rb
-- lib/sportdb/parser/token-table.rb
+- lib/sportdb/parser/token-status_inline.rb
 - lib/sportdb/parser/token-text.rb
 - lib/sportdb/parser/token-time.rb
 - lib/sportdb/parser/token.rb
-- lib/sportdb/parser/token_helpers.rb
 - lib/sportdb/parser/version.rb
 homepage: https://github.com/sportdb/sport.db
 licenses:

data/lib/sportdb/parser/blocktxt.rb

@@ -1,99 +0,0 @@
-###
-#  generic block text/txt helper
-
-## todo/chekc - find a better name SectTxt or ??
-
-class BlockTxt
-
-def self.parse( txt ) new( txt); end 
-def self.read( path ) parse( read_text( path )); end
-
-
-def initialize( txt )
-   @sects = _parse( txt )
-   self
-end
-
-def size()       @sects.size; end
-
-def each( &blk )            @sects.each( &blk ); end
-def each_with_index( &blk ) @sects.each_with_index( &blk ); end
-
-
-def text
-    ## only get all txt1 parts joined as single all-in-one string txt
-    @sects.map {|sect| sect[0] }.join( "\n" )
-end
-
-
-def dump   ## for debugging
-    puts "==> sects (#{@sects.size}):"
-    pp @sects
-    puts "    #{@sects.size} sect(s)"
-end
-
-
-##
-#  quick support for  __END__
-END_RE = %r{ ^
-                          [ ]* __END__ [ ]*
-                            .*?
-                         \z   ## note - until end-of-string/file !!!
-                      }mx
-
-
-##    SECT_RE (old) = %r{^
-##                                 [ ]* --- [ ]*
-##                              $}x
-##
-##  do NOT use --- (used in fbtxt and markdown and yaml etc.)
-
-
-## e.g.  §  or §§ or § § § or such
-##    maybe allow  :: § :: or such too   or --- § --- or such
-SECT_RE = %r{^
-                     [ ]* § 
-                         ([ ]*§)* 
-                     [ ]*
-                 $}x
-
-## split by " => or  =====> "
-## todo/check - subsect?? find a better name?  in/out or txt1/txt2 
-SUBSECT_RE = %r{^
-                        [ ]* 
-                            =+ > 
-                        [ ]*
-                  $}x
-
-
-
-def _parse( txt )
-    blocks = []   ## note - holds [txt,exp] pairs
-
-    txt = txt.sub( END_RE, '' )
-
-    ## split by §
-    sections = txt.split( SECT_RE  )
-
-    sections.each_with_index do |sect,i|
-       ## puts ">>> start #{i+1}"
-       ## pp sect
-       ## puts "<<< end #{i+1}"
-
-      txt1, txt2 = sect.split( SUBSECT_RE )
-      blocks << [txt1,txt2]
-    end
-
-    blocks
-end
-
-
-
-end # class BlockTxt
-
-
-###
-#  function-style helpers
-
-def read_blocktxt( path ); BlockTxt.read( path ); end
-def parse_blocktxt( txt ); BlockTxt.new( txt ); end

data/lib/sportdb/parser/lexer_tty.rb

@@ -1,111 +0,0 @@
-module SportDb
-class Lexer
-
-
-########
-##   experimental teletype mode 
-##    only space, A-Z and 0-9 allowed
-IS_TTY_LINE_RE = %r{  \A  
-                       ## note - use NEGATIVE lookahead to exclude blank lines
-                         (?! [ ]*\z)
-
-                          [A-Z0-9 ]+
-                      \z
-                  }x
-
-
-TTY_SPACES_RE = %r{ (?<spaces> [ ]{2,}) |
-                    (?<space>  [ ])
-                  }x
-TTY_NUM_RE    = %r{   \b  (?<num> \d+ ) \b 
-                  }x
-
-##
-##  note - TEXT for now allows    A, 1A, A1, A1A, A1 B1 C1, 
-##                                A1AA1 2B22 3C33
-##    - single space only for concat
-##       text segments MUST NOT be all numbers e.g. 1, 11, etc.                  
-TTY_TEXT_RE   = %r{   \b (?<text>                         
-                           (?:
-                              [A-Z]  ## MUST start with letter  
-                                |
-                               [0-9]+[A-Z]   ## or numbers followed by letter 
-                             )
-                             [0-9A-Z]*
-                             (?:
-                                 ### allow move segements separated
-                                 ##     by single space
-                                  [ ]
-                                 (?: 
-                                     [A-Z]  ## MUST start with letter  
-                                      |
-                                     [0-9]+[A-Z]   ## or numbers followed by letter 
-                                  )
-                                 [0-9A-Z]*
-                             )*
-                          )
-                          \b   
-                  }x 
-
-
-TTY_RE = Regexp.union(
-                TTY_SPACES_RE,
-                TTY_TEXT_RE,
-                TTY_NUM_RE,
-                ##  fix add ANY_RE,  
-)
-
-
-def _tokenize_tty_line( line )
-   line = line.strip
-
-   tokens = []
-   
-   ## track last offsets - to report error on no match
-   ##   or no match in end of string
-   offsets = [0,0]
-   pos = 0
-   m = nil   
- 
-
-  while m = TTY_RE.match( line, pos )
-    offsets = [m.begin(0), m.end(0)]
-
-    if offsets[0] != pos
-      ## match NOT starting at start/begin position!!!
-      ##  report parse error!!!
-      msg =  "!! WARN - tokenize (tty) error - skipping >#{line[pos..(offsets[0]-1)]}< @#{offsets[0]},#{offsets[1]} in line >#{line}<"
-      puts msg
-      log( msg )
-    end
-
-    pos = offsets[1]
-
-    t =  if m[:spaces] || m[:space] 
-               nil    ## skip spaces
-          elsif m[:text]
-            [:TTY_TEXT, m[:text]]
-          elsif m[:num]
-            [:TTY_NUM, m[:num].to_i(10)] 
-          else
-              ## report error/raise expection
-              puts "!!! TTY TOKENIZE ERROR - no match found"
-              nil 
-          end
-     
-    tokens << t    if t
-  end
-
-  ## check if no match in end of string
-  if offsets[1] != line.size
-      msg =  "!! WARN - tokenize (tty) error - skipping >#{line[offsets[1]..-1]}< @#{offsets[1]},#{line.size} in line >#{line}<"
-      puts msg
-      log( msg )
-  end
-
-  tokens
-end
-
-end  # class Lexer
-end # module SportDb
-

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

@@ -1,149 +0,0 @@
-module SportDb
-class Lexer
-
-
-###
-## check for 
-##   table (standing) lines
-##
-##  e.g.
-##
-##        Pld W D L GF-GA Pts   |  d d d d-d d
-##        Pld GF-GA Pts         |  d d-d d
-##        Pld Pts W D L GF-GA   |  d d d d d d-d
-##
-## Pld   = matches played
-## GF-GA = goal for, goal against
-
-
-##        Pld W D L GF-GA Pts   |  d d d d-d d
-##
-##  1.BRAZIL            3  2  1  0   7- 2  7
-##  2.MEXICO            3  2  1  0   4- 1  7
-##  3.Croatia           3  1  0  2   6- 6  3
-##  4.Cameroon          3  0  0  3   1- 9  0 
-
-##  add more headings?? e.g.
-##    Final Table:
-##
-
-
-TABLE_HEADING_I_RE = %r{
-       \A
-        [ ]*  ## ignore leading spaces (if any)
-       (?<table_heading>
-         \b
-          P(?:ld)?  [ ]+ 
-           W        [ ]+
-           D        [ ]+
-           L        [ ]+
-           Gls      [ ]+
-           Pts
-        \b
-         )
-        [ ]*  ## ignore trailing spaces (if any) 
-        \z
-   }xi
-
-
-##
-##  "solid"-style
-##     -----------------------------------------------------
-##  "dashed"-style ??
-##     - - - - - - - - - - - - - - - - - - - - - - - - - - -
-
-
-TABLE_DIVIDER_RE = %r{
-            \A
-        [ ]*  ## ignore leading spaces (if any)
-            (?<table_divider>
-                (?:  ---   ## note - require three dashes minimum (---)
-                      [-]* 
-                )  
-                  |
-                (?: - [ ]+ - [ ]+ -  ## note - require three dashes minimum (- - -)
-                      (?: [ ]+ -)*   ##   todo/check - restrict spaces to 2 or 3 or such - why? why not?
-                )  
-            )
-        [ ]*  ## ignore trailing spaces (if any) 
-            \z
-      }xi
-
-
-####
-##   1.SOLOMON I.    1  1  0  0  3- 1  3
-##   2.TAHITI        1  0  0  1  1- 3  0
-##   -.Cook Islands  withdrew after first match (annulled) due to Covid-19 outbreak in squad
-##   -.Vanuatu       withdrew before playing any matches due to Covid-19 outbreak in squad  -->
-##
-##  note - starting with -. is a table note!!!
-
-
-TABLE_NOTE_RE = %r{
-       \A
-        [ ]*  ## ignore leading spaces (if any)
-           -\.
-           [ ]*
-       (?<table_note>
-            .+?   ## note - use non-greedy       
-         )
-        [ ]*  ## ignore trailing spaces (if any) 
-        \z
-}xi
-
-TABLE_I_RE = %r{
-        (?<table>\b 
-             \d{1,2} [ ]+                        # Pld
-             \d{1,2} [ ]+                        # W
-             \d{1,2} [ ]+                        # D
-             \d{1,2} [ ]+                        # L
-             (?: \d{1,3} - [ ]* \d{1,3} [ ]+ )   # GF-GA
-             \d{1,3}                             # Pts   
-              \b 
-        )}xi
-
-##      Pld Pts W D L GF-GA   |  d d d d d d-d
-##
-## 1. ARG^         3  6  3 0 0    10-4
-## 2. CHI          3  4  2 0 1     5-3
-## 3. FRA          3  2  1 0 2     4-3
-## 4. MEX          3  0  0 0 3     4-13
-
-TABLE_II_RE = %r{
-        (?<table>\b 
-             \d{1,2} [ ]+                        # Pld
-             \d{1,3} [ ]+                        # Pts   
-             \d{1,2} [ ]+                        # W
-             \d{1,2} [ ]+                        # D
-             \d{1,2} [ ]+                        # L
-             (?: \d{1,3} - [ ]* \d{1,3})   # GF-GA
-              \b 
-        )}xi
-
-
-
-#############################################
-# map tables
-#  note: order matters; first come-first matched/served
-
-##  possible start lines for a table
-##    excludes NOTE
-##    and RULER (e.g. --- or) or such in the future
-TABLE_RE = Regexp.union(
-    TABLE_HEADING_I_RE,
-    TABLE_I_RE,
-    TABLE_II_RE,
-)
-
-## all possible continuation for a table
-##   excludes HEADING
-TABLE_MORE_RE = Regexp.union(
-    TABLE_NOTE_RE,
-    TABLE_DIVIDER_RE,
-    TABLE_I_RE,
-    TABLE_II_RE,
-)
-
-
-end  #   class Lexer
-end  # module SportDb

data/lib/sportdb/parser/token_helpers.rb

@@ -1,92 +0,0 @@
-module SportDb
-class Lexer
-
-
-=begin
-def self._mk_is( re )
-    ##   add  \A ... \z to regex
-    ##     for strict matching of beginning and end of string
-    ##     regex note -  \z will NOT allow trailing newline(s)!!!!
-    ##      note - must double espace \\A,\\z  in quoted string!!
-    Regexp.new( %Q<   \\A
-                    (?:#{re.source})
-                       \\z
-                  >, re.options )
-end
-
-
-  IS_TEAM_RE = _mk_is( TEXT_RE )   ## todo/fix - rename TEXT_RE to TEAM_RE!!!
-  IS_DATE_RE = _mk_is( DATE_IIII_RE )    ## DATE_RE )
-=end
-
-
-def self._parse_team( str )  
-    ## note - strip - leading/trailing spaces
-    m = TEXT_RE.match( str.strip )
-    if m && m.pre_match == '' && m.post_match == ''
-      m
-    elsif  m
-        ## note - match BUT not anchored to start and end-of-string!!!
-        ##  report, error somehow??
-      nil   
-    else
-      nil  ## no match - return nil
-    end
-end
-
-
-def self._parse_date( str )
-    ## note - strip - leading/trailing spaces
-    m = DATE_RE.match( str.strip )
-    
-    ####  todo/fix/check:
-    ###   wrapped with  \A \z NOT working with union  - check later - why?
-    ###   use hand-coded  with pre_match = "" and post_match = ""
-    
-    if m && m.pre_match == '' && m.post_match == ''
-      ## return hash table with captured components
-      date = {}
-      ## 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]
-      ## check - use y too for two-digit year or keep separate - why? why not?
-      date[:yy] = m[:yy].to_i(10)    if m[:yy]    ## two digit year (e.g. 25 or 78 etc.)
-      date[:m]  = m[:month].to_i(10)  if m[:month]
-      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]
-      date 
-    elsif  m
-        ## note - match BUT not anchored to start and end-of-string!!!
-        ##  report, error somehow??
-      nil   
-    else
-      nil  ## no match - return nil
-    end
-end   
-
-
-def self._parse_score_full( str )
-    ## note - strip - leading/trailing spaces
-    m=SCORE_FULL_RE.match( str )
-
-    if m && m.pre_match == '' && m.post_match == ''
-       score = {}
-       score[:p]  = [m[:p1].to_i,m[:p2].to_i]     if m[:p1] && m[:p2]
-       score[:et] = [m[:et1].to_i,m[:et2].to_i]   if m[:et1] && m[:et2]
-       score[:ft] = [m[:ft1].to_i,m[:ft2].to_i]   if m[:ft1] && m[:ft2]
-       score[:ht] = [m[:ht1].to_i,m[:ht2].to_i]   if m[:ht1] && m[:ht2]
-       ## score[:agg] = [m[:agg1].to_i,m[:agg2].to_i]   if m[:agg1] && m[:agg2]
-       score
-    elsif  m
-        ## note - match BUT not anchored to start and end-of-string!!!
-        ##  report, error somehow??
-      nil   
-    else
-      nil  ## no match - return nil
-    end
-end       
-       
-
-end  #   class Lexer
-end  # module SportDb