sportdb-parser 0.7.2 → 0.8.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 35a69fdc1a31058e1a0069e89106bb68f2c9bbff10465a4111d0d0bf0eb3a210
-  data.tar.gz: 98f2e0b3f1d21d71b783d96c6e5cac7f2357d436951cae0865558a2289dd2bca
+  metadata.gz: 4b9d6cc209285ddfacdf76565fd43541500cac28c516edeecf779c65c98fe556
+  data.tar.gz: 329531cc04507bd0a5bc3f8e3488a2e2539ddebe955a48e37aa6d07760135777
 SHA512:
-  metadata.gz: b7f43a3c0fd14130b2f5c278b0d0742b4a2d71a489383b80c8a8349c70ebb14280c54b53fc008782c9b539b4ffa4df1442c2f4e82e70ebea05a695343472f613
-  data.tar.gz: 7b636234a8a66671f35d5d7bdc0113b2df0131b9e065e86eb8f8c688b58ce2eb4d931e21a91a69bfb23dab5d56ed2b78d04400fd308c960fe9f1eb3653f5ab53
+  metadata.gz: 57949562e6cd27b1d8058afa549e1d24a8cfd820590d14407d8ed5b34c2095b451ca53065057ef24b86db2dee5d69570cbb054830bd3796ae5b400008da343cf
+  data.tar.gz: 2e1b841018e66468d65727947290f30e143340789b271de27c4edd2fb18e169b04446bcc203433329ba343174138d42de112d9801eb608de8c7c5b9144d5c920

data/CHANGELOG.md

@@ -1,4 +1,4 @@
-### 0.7.2
+### 0.8.0
 ### 0.0.1 / 2024-07-12
 
 * Everything is new. First release.

data/Manifest.txt

@@ -3,6 +3,8 @@ Manifest.txt
 README.md
 Rakefile
 lib/sportdb/parser.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
@@ -16,10 +18,13 @@ lib/sportdb/parser/lexer-tokenize.rb
 lib/sportdb/parser/lexer.rb
 lib/sportdb/parser/lexer_buffer.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

data/Rakefile

@@ -28,3 +28,26 @@ Hoe.spec 'sportdb-parser' do
     required_ruby_version: '>= 3.1.0'
   }
 end
+
+
+
+###
+## add our own g task to build parser.rb (via racc & erb)
+
+desc 'Generate parser.rb (via racc & erb)'
+task g: 'lib/sportdb/parser/parser.rb'
+
+
+file 'lib/sportdb/parser/parser.rb' => 'fbtxt/parser.y' do |t|
+  puts "  [racc] generate parser.rb from parser.y"
+  ## sh "racc -v -o #{t.name} #{t.source}"
+  sh "racc -o #{t.name} #{t.source}"
+end
+
+
+file 'fbtxt/parser.y' => FileList['fbtxt/grammar/*.y'] do |t|
+  puts "  [erb] generate parser.y; merge grammar/*"
+  Dir.chdir( 'fbtxt' ) do
+    sh "erb parser.y.erb > parser.y"
+  end
+end

data/lib/sportdb/parser/debuggable.rb

@@ -0,0 +1,53 @@
+###
+## debugg(able) mix-in helper
+##      used for lexer & parser
+
+module Debuggable
+  def self.included(base)
+    base.extend(ClassMethods)
+  end
+
+  module ClassMethods
+    def debug=(value)
+        raise ArgumentError, "true|false required for debug flag; got #{value}"  unless value.is_a?(TrueClass) || value.is_a?(FalseClass)
+        @debug=value;
+    end
+    def debug()  @debug || false; end    ## note - default to false
+
+    def debug?() @debug == true; end
+  end
+
+
+  ########
+  ##  InstanceMethods
+  def debug?() self.class.debug?; end
+
+  ##
+  ## todo/check - allow/support multiple msg args - why? why not?
+
+  ##  note - use trace for debug for now - why? why not?
+  def _trace( *args )  __log__( :DEBUG, *args ); end
+
+  def _info( *args )   __log__( :INFO, *args );  end
+  def _warn( *args)    __log__( :WARN, *args ); end
+  def _error( *args )  __log__( :ERROR, *args ); end
+
+
+
+private
+  def __log__( level, *args )   ## :DEBUG, :INFO, :WARN, :ERROR, :FATAL etc.
+
+    if level == :DEBUG && !debug?
+        ## print nothing
+    else
+      msg = args.map do |arg|
+                arg.to_str
+               end.join(' | ')
+
+      fout =  level == :WARN || level == :ERROR || level == :FATAL ? STDERR : STDOUT
+
+      ## todo/check/fix - add/use/check for "local" logger
+      fout.puts "#{level} [#{self.class.name}] -- #{msg}"
+    end
+  end
+end  # module Debuggable

data/lib/sportdb/parser/lexer-logger.rb

@@ -0,0 +1,20 @@
+
+module SportDb
+class Lexer
+
+
+def log( msg )
+   ## append msg to ./logs.txt
+   ##     use ./errors.txt - why? why not?
+   ##
+   ##  change to ./logs_lexer.txt or such - why? why not?
+   ##    auto-add/prepend  [Lexer] and timestamp!!!  to msg - why? why not?
+   File.open( './logs.txt', 'a:utf-8' ) do |f|
+     f.write( "[Lexer] " + msg )
+     f.write( "\n" )
+   end
+end
+
+
+end   # class Lexer
+end   # module SportDb

data/lib/sportdb/parser/lexer-on_goal.rb

@@ -6,7 +6,6 @@ class Lexer
 
 GOAL_RE = Regexp.union(
     SPACES_RE,
-    GOAL_NONE_RE,
     GOAL_MINUTE_RE,
     GOAL_MINUTE_NA_RE,
     GOAL_COUNT_RE,
@@ -16,15 +15,11 @@ GOAL_RE = Regexp.union(
     ## todo/fix - add ANY_RE !!!!
 )
 
+
 def _on_goal( m, ctx: )
 
          if m[:space] || m[:spaces]
               nil    ## skip space(s)
-         elsif m[:goals_none]    ## note - eats-up semicolon!! e.g. -; or - ;
-             # was:[:GOALS_NONE,"<|GOALS_NONE|>"]
-             ##   use literal text!!
-             Token.new( :GOALS_NONE, m[:goals_none],
-                            lineno: ctx.lineno, offset: m.offset(:goals_none))
          elsif m[:goal_sep_alt]
              # was: [:GOAL_SEP_ALT, "<|GOAL_SEP_ALT|>" ]   ## e.g. dash (-) WITH leading & trailing space required
              Token.new( :GOAL_SEP_ALT, m[:goal_sep_alt],

data/lib/sportdb/parser/lexer-on_prop_misc.rb

@@ -7,9 +7,12 @@ class Lexer
 ##         todo/fix - use custom (limited) prop basics too
 PROP_CARDS_RE =  Regexp.union(
    SPACES_RE,
+   CARDS_NONE_LEFT_RE,
+   CARDS_NONE_RIGHT_RE,
    MINUTE_RE,
    PROP_NAME_RE,
-   /  (?<sym>  [;,-]) /x
+   CARDS_SEP_ALT_RE,     ##  note - add dash (-) with (required) spaces
+   /  (?<sym>  [;,]) /x
    ## todo/fix - add ANY_RE here too!!!
 )
 
@@ -18,6 +21,16 @@ def _on_prop_cards( m, ctx: )      ## note - m is MatchData object
 
          if m[:space] || m[:spaces]
               nil    ## skip space(s)
+         elsif m[:cards_none_left]
+              Token.new(:CARDS_NONE_LEFT, m[0],
+                               lineno: ctx.lineno, offset: m.offset(0))
+         elsif m[:cards_none_right]
+              Token.new(:CARDS_NONE_RIGHT, m[0],
+                               lineno: ctx.lineno, offset: m.offset(0))
+         elsif m[:cards_sep_alt]
+             Token.new( :CARDS_SEP_ALT, m[0],
+                              lineno: ctx.lineno, offset: m.offset(0))
+
          elsif m[:prop_name]
               Token.new(:PROP_NAME, m[:name],
                                lineno: ctx.lineno, offset: m.offset(:prop_name))

data/lib/sportdb/parser/lexer-on_top.rb

@@ -46,6 +46,14 @@ def _on_top( m, ctx: )      ## note - m is MatchData object
                                                   lineno: ctx.lineno, offset: m.offset(:inline_awd))
         elsif m[:inline_canc] then Token.new(:INLINE_CANC, m[:inline_canc],
                                                   lineno: ctx.lineno, offset: m.offset(:inline_canc))
+
+        elsif m[:inline_round_short] then Token.new(:INLINE_ROUND_SHORT, m[0],
+                                                  lineno: ctx.lineno, offset: m.offset(0),
+                                                     value: m[:inline_round_text])
+        elsif m[:inline_round_big] then Token.new(:INLINE_ROUND_BIG, m[0],
+                                                  lineno: ctx.lineno, offset: m.offset(0),
+                                                     value: m[:inline_round_text])
+
         elsif m[:status]      then Token.new(:STATUS, m[:status],
                                                   lineno: ctx.lineno, offset: m.offset(:status),
                                                   value: _build_status( m ))

data/lib/sportdb/parser/lexer-tokenize.rb

@@ -138,7 +138,8 @@ def _tokenize_line( line, lineno )
         ### todo/fix - add prop yellow/red cards too - why? why not?
         ##  todo/fix - separate sent off and red card
         ##     sent-off - incl. red card, yellow/red card and the era before red cards!!
-        if ['sent off'].include?( key.downcase)
+        if ['sent-off',
+            'sent off'].include?( key.downcase)
           @re = PROP_CARDS_RE    ## use CARDS_RE ???
           tokens << Token.new(:PROP_SENTOFF, m[:key],
                                    lineno: lineno, offset: m.offset(:key))
@@ -150,6 +151,18 @@ def _tokenize_line( line, lineno )
           @re = PROP_CARDS_RE
           tokens << Token.new(:PROP_YELLOWCARDS, m[:key],
                                    lineno: lineno, offset: m.offset(:key))
+
+        ## todo: check allow yellow/red cards: or such as key!!!
+        ##   check for alternate spellings
+        ##   'yellow/red cards' - note - meaning closes to yellow or red !!
+        ##    thus use yellow-red cards as default/standard - why? why not?
+        elsif ['yellow-red cards',
+               'yellowred cards'].include?( key.downcase )
+          @re = PROP_CARDS_RE
+          tokens << Token.new(:PROP_YELLOWREDCARDS, m[:key],
+                                   lineno: lineno, offset: m.offset(:key))
+
+
         elsif ['ref', 'referee',
                'refs', 'referees'   ## note - allow/support assistant refs
               ].include?( key.downcase )
@@ -354,7 +367,9 @@ def _tokenize_line( line, lineno )
               when '>' then @geo_count = 0;
                             Token.literal( ',', lineno: lineno, offset: m.offset(:sym))
                                ## note - treat geo sep > (ascii) like comma for now!!!
-              when '[' then
+
+
+              when '[','▪'
                     ##
                     ## todo/fix
                     ##   add virtual geo_end token!!!
@@ -363,6 +378,7 @@ def _tokenize_line( line, lineno )
                  @re = RE
                  pos = old_pos
                  next   ## backtrack (resume new loop step)
+               ## fix-fix-fix  merge  '▪' with '['
               else
                  Token.literal( m[:sym], lineno: lineno, offset: m.offset(:sym))
               end
@@ -426,7 +442,10 @@ def _tokenize_line( line, lineno )
       @re == PROP_PENALTIES_RE  ||
       @re == PROP_ATTENDANCE_RE ||
       @re == PROP_REFEREE_RE
-     if [',', '-', ';'].include?( tokens[-1].type)
+
+     ## note - add :CARDS_SEP_ALT (aka  -)!!! too
+     ##   maybe later  CARDS_NONE_LEFT too - why? why not?
+     if [',', '-', ';', :CARDS_SEP_ALT].include?( tokens[-1].type)
         ## continue/stay in PROP_RE mode
         ##  todo/check - auto-add PROP_CONT token or such
         ##                to help parser with possible NEWLINE

data/lib/sportdb/parser/lexer.rb

@@ -1,57 +1,30 @@
 
 module SportDb
-class Lexer
-
-
 
-def log( msg )
-   ## append msg to ./logs.txt
-   ##     use ./errors.txt - why? why not?
-   ##
-   ##  change to ./logs_lexer.txt or such - why? why not?
-   ##    auto-add/prepend  [Lexer] and timestamp!!!  to msg - why? why not?
-   File.open( './logs.txt', 'a:utf-8' ) do |f|
-     f.write( msg )
-     f.write( "\n" )
+class LexerResult
+   attr_reader :tokens, :errors
+   def initialize( tokens, errors=[] )
+       @tokens, @errors = tokens, errors
    end
-end
-
-
-def _trace( *args )
-  if debug?
-    print "[DEBUG] Lexer -- "
-    args.each { |arg| puts args }
-  end
-end
-
-def _warn( *args )
-  print "!! [WARN] Lexer -- "
-  args.each { |arg| puts args }
-end
-
-def _info( *args )
-  print "[INFO] Lexer -- "
-  args.each { |arg| puts args }
-end
 
-
-def debug?()  @debug == true; end
+   def ok?()  @errors.size == 0; end
+   def nok?() !ok?; end
+end  # class LexerResult
 
 
 
+class Lexer
+  include Debuggable     ## auto-adds debug?, _trace, _info, etc.
 
 
-def initialize( txt, debug: false )
-   raise ArgumentError, "text as string expected for lexer; got #{txt.class.name}"  unless txt.is_a?(String)
+def initialize( txt )
+   raise ArgumentError, "text as string expected for lexer; got #{txt.class}"  unless txt.is_a?(String)
 
    @txt   = txt
-   @debug = debug
 end
 
 
-
-
-def tokenize_with_errors
+def tokenize_with_errors( flatten: true )
 
     tokens_by_line = []   ## note: add tokens line-by-line (flatten later)
     errors         = []   ## keep a list of errors - why? why not?
@@ -91,13 +64,12 @@ def tokenize_with_errors
         ##  strip (inline) end-of-line comments (from line)
         ##    check/discuss: make - inline comment require trailing space
         ##                      e.g.   #1 vs # 1   - why? why not?
-        line = line.sub( /   [ ]*      ## (eat-up) optional leading space(s)
+        line = line.sub( /   [ ]*      ## (eat-up) optional leading space(s) too - why? why not?
                               \#{1,}.*?
                              \z
                             /x, '' )
 
 
-        ####
         #  support __END__ marker to cut-off input
         break if line.match?( /\A [ ]*   ## optional leading space(s)
                                    __END__
@@ -138,6 +110,10 @@ def tokenize_with_errors
           tokens_by_line  << more_tokens
           errors          += more_errors
         end
+
+        ## output last line from tokens by line in debug mode
+        _trace( "  #{tokens_by_line[-1].size} token(s): " + tokens_by_line[-1].pretty_inspect )
+
     end # each line
 
 
@@ -226,10 +202,10 @@ def tokenize_with_errors
     ## puts "tokens_by_line:"
     ## pp tokens_by_line
 
-
-    ## flatten tokens
-    tokens = []
-    tokens_by_line.each do |tok_line|
+    if flatten
+      ## flatten tokens
+      tokens = []
+      tokens_by_line.each do |tok_line|
 
         ## if debug?
         ##   pp tok_line
@@ -243,12 +219,12 @@ def tokenize_with_errors
             ##                  use last - why? why not?
             tokens  << Token.newline( lineno: tok_line[0].lineno )
          end
-    end
-
-    [tokens,errors]
-
-end   # method tokenize_with_errors
-
+      end
+      [tokens,errors]
+   else
+      [tokens_by_line, errors]
+   end
+end   # method tokenize
 
 
 end  # class Lexer

data/lib/sportdb/parser/parse_tree--core.rb

@@ -0,0 +1,29 @@
+
+
+# module JsonSerializable
+#  def to_json(*args)
+#    as_json.to_json(*args)
+#  end
+# end
+
+###
+##  add as_json  to Object/Array/Hash
+class Object
+  def as_json(*)
+    self
+  end
+end
+
+class Array
+  def as_json(*)
+    map { |v| v.as_json }
+  end
+end
+
+class Hash
+  def as_json(*)
+    each_with_object({}) do |(k, v), h|
+      h[k.to_s] = v.as_json
+    end
+  end
+end