sportdb-parser 0.7.1 → 0.8.0

data/lib/sportdb/parser.rb

@@ -1,112 +1,138 @@
-## 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/debuggable'    ## generic debug/logger helper
 
 
-require_relative 'parser/version'
 
 ##
-## 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_token'
+
+require_relative 'parser/lexer-prep_doc'
+require_relative 'parser/lexer-prep_line'
+
+require_relative 'parser/lexer-logger'   ## e.g. _trace, _warn, _info, etc.
+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
 ##    MUST uncomment  require 'racc/parser'  in generated parser!!!
-require_relative 'parser/parser_runtime'
+require_relative 'parser/parser-runtime'
 
 require_relative 'parser/parser'   ## auto-generated by racc (from parser.y)
-require_relative 'parser/racc_parser'
-require_relative 'parser/racc_tree'
+require_relative 'parser/parser-top'
+
+require_relative 'parser/parse_tree--core'    ## incl. core extensions eg. as_json on Object/Array/Hash
+require_relative 'parser/parse_tree'
+require_relative 'parser/parse_tree-match'
+require_relative 'parser/parse_tree-props'
+
+
+## deprecated - remove old RaccMatchParser name sooner than later!!
+##    use Fbtxt.parse for "public" api
+RaccMatchParser = SportDb::Parser
 
 
 
 ###
 #  make parser api (easily) available - why? why not?
+#
+#  use Fbtxt.parse( txt ) &&
+#      Fbtxt.lex( txt)            -- or Fbtxt.tokenize?
+#
+#   Fbtxt.lex( txt,
+#                 flatten: true|false,        ## keep nested [] for lines (no NEWLINE tokens)
+#                 transform: true|false )     ## builtin transforms eg. merge DATE & TIME => DATETIME, etc.
+#
+#   how to deal with errors?
+##    always return   [tokens, errors] or such?
+##           or return a Result struct - why? why not?
 
-=begin
-module SportDb
-   def self.parser() @@parser ||= Parser.new; end
-   def self.parse( ... )
-   end
-   def self.tokenize( ... )
-   end
-end  # module SportDb
-=end
+
+
+module Fbtxt
+
+def self.lex( txt, flatten: true )      ## returns SportDb::LexerResult (ok?/tokens/errors)
+  lexer = SportDb::Lexer.new( txt )
+  ## move "wrapping" into result obj inside lexer - why? why not?
+  tokens, errors = lexer.tokenize_with_errors( flatten: flatten )
+  SportDb::LexerResult.new( tokens, errors )
+end
+
+def self.parse( txt )   ## returns SportDb::ParserResult (ok?/tree/errors)
+  parser = SportDb::Parser.new( txt )
+  ## move "wrapping" into result obj inside parser - why? why not?
+  tree, errors = parser.parse_with_errors
+  SportDb::ParserResult.new( tree, errors )
+end
+end #  module Fbtxt
 
 
 
-module SportDb
-class Parser
-####################
-#  "default" lexer & parser  (wraps RaccMatchParser)
-
-  def tokenize_with_errors( lines, debug: false )
-     lexer = Lexer.new( lines )
-     tokens, errors = lexer.tokenize_with_errors
-     [tokens, errors]
-  end
-
-  ### convience helper - ignore errors by default
-  def tokenize( lines, debug: false )
-    tokens, _ = tokenize_with_errors( lines, debug: debug )
-    tokens
-  end
-
-
-  def parse_with_errors( lines, debug: false )
-    ## todo/check - if lines needs to chack for array of lines and such
-    ##                        or handled by tokenizer???
-    parser = RaccMatchParser.new( lines )
-    tree, errors = parser.parse_with_errors
-    [tree, errors]
-  end
-
-  ### convience helper - ignore errors by default
-  def parse( lines, debug: false )
-    tree, _ = parse_with_errors( lines, debug: debug )
-    tree
-  end
-end  # class Parser
-end  # module SportDb
 
 
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: sportdb-parser
 version: !ruby/object:Gem::Version
-  version: 0.7.1
+  version: 0.8.0
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-05-22 00:00:00.000000000 Z
+date: 2026-07-06 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: cocos
@@ -72,32 +72,50 @@ files:
 - README.md
 - Rakefile
 - lib/sportdb/parser.rb
-- lib/sportdb/parser/blocktxt.rb
+- lib/sportdb/parser/debuggable.rb
+- lib/sportdb/parser/lexer-logger.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/parse_tree--core.rb
+- lib/sportdb/parser/parse_tree-match.rb
+- lib/sportdb/parser/parse_tree-props.rb
+- lib/sportdb/parser/parse_tree.rb
+- lib/sportdb/parser/parser-runtime.rb
+- lib/sportdb/parser/parser-top.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/racc_parser.rb

@@ -1,88 +0,0 @@
-
-####
-#   RaccMatchParser support machinery (incl. node classes/abstract syntax tree)
-
-class RaccMatchParser
-
-
-def initialize( txt,  debug: false )
-    ## puts "==> txt:"
-    ## puts txt
- 
-    @tree   = [] 
-    @errors = []
-
-    ### todo:
-    ##  -  pass along debug flag
-    lexer = SportDb::Lexer.new( txt, debug: debug )
-    ##  note - use tokenize_with_errors and add/collect tokenize errors
-    @tokens, @errors = lexer.tokenize_with_errors
-    ## pp @tokens
-    
-    ## quick hack - convert to racc format single char literal tokens e.g. '@' etc.
-    @tokens = @tokens.map do |tok|
-                 if tok.size == 1
-                   [tok[0].to_s, tok[0].to_s]
-                 else 
-                   tok
-                 end
-               end
-  end
-
-
-  def debug( value ) @debug = value; end   ### fix: use setter-style e.g. debug=(value) !!!
-  def debug?()  @debug == true; end
-
-  ## debug - trace / print message
-  def trace( msg )
-     puts "  [parse] " + msg    if debug?
-  end
-
-
-
-
-  def next_token
-    tok = @tokens.shift
-    trace( "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_with_errors    
-     trace( "start parse:" )
-     do_parse
-     [@tree, @errors]
-  end
-
-  def parse  ## convenience shortcut (ignores errors)
-    tree, _ = parse_with_errors
-    tree 
-  end
-
-
-  attr_reader :errors
-  def errors?()   @errors.size > 0; end
-
-
-  def on_error(error_token_id, error_value, value_stack)
-    ## auto-add error_token (as string)
-    error_token = Racc_token_to_s_table[error_token_id] 
-    args = [error_token, error_token_id, error_value, value_stack]
-    puts
-    puts "!! on parse error:"
-    puts "args=#{args.pretty_inspect}"
-
-    @errors << "parse error on token: #{error_token} (#{error_token_id}) with value: #{error_value}, stack: #{value_stack.pretty_inspect}" 
-    ## exit 1  ##   exit for now  -  get and print more info about context etc.!!
-  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   # class RaccMatchParser