sportdb-parser 0.6.0 → 0.6.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 9af6317b144478400067502e60de2f8d6232ebf6e036b0f99b78f9c29922dba2
-  data.tar.gz: 7066483378693f6376f9c30ec71e5d4172c817c11025dd4e669da6d581b6ad54
+  metadata.gz: 85526406c8cd97a5b4e8580e64597b60f2046f4667a97080434238e067be2788
+  data.tar.gz: dcd5e6aaa854654974644c026fb99545c31ef2d5929d0518d8418630b5d6ea76
 SHA512:
-  metadata.gz: '039fdc82039d05ae8f51847a79dd77c0a657e316e8b0705a28bdf2f8e594f37531ea07a230c9e1a9133f96293975190dba070f50515d58bc9926e4ef3e8e152f'
-  data.tar.gz: e0f6483cd26ba7ef4800ecf76efd6f05e01e3a2458dbc6b65fe6582654c8d28627facbfc8228655e361df84c43418c9583826908cdcf3d61bf06d186288c56fa
+  metadata.gz: fc086846a66d2657d3debae5562fe20fbd2327741c8af1485972dfd0b8f46b3c649c0345ba173f2d3f40622bd4bddecc7ea0072d4d129bc5dc542554c539ebab
+  data.tar.gz: c0c4653cb40cb89e8086b6dc00ed853c62abf75fc18972cf98230a779cd8a7f73d797098ec53c2e942e6271c00c9d11a1da8f6975704141cbfbba599ec741098

data/CHANGELOG.md

@@ -1,4 +1,4 @@
-### 0.6.0
+### 0.6.2
 ### 0.0.1 / 2024-07-12
 
 * Everything is new. First release.

data/lib/sportdb/parser/lexer.rb

@@ -147,12 +147,15 @@ def initialize( lines, debug: false )
     ##   strip lines with comments and empty lines striped / removed
     ##      keep empty lines? why? why not?
     ##      keep leading spaces (indent) - why?
+    ##
+    ##  note - KEEP empty lines (get turned into BLANK token!!!!)
+
     @txt = String.new
     txt_pre.each_line do |line|    ## preprocess
        line = line.strip
-       next if line.empty? || line.start_with?('#')   ###  skip empty lines and comments
+       next if line.start_with?('#')   ###  skip comments
        
-       line = line.sub( /#.*/, '' ).strip             ###  cut-off end-of line comments too
+       line = line.sub( /#.*/, '' ).strip   ###  cut-off end-of line comments too
        
        @txt << line
        @txt << "\n"
@@ -193,6 +196,18 @@ def tokenize_with_errors
                    t
                  end
 
+        ### check for "section" starters e.g. Teams or such
+        t = tokens[0] 
+        if t[0] == :TEXT
+            text = t[1]
+            if text =~ /^teams$/i
+               t[0] = :TEAMS
+            elsif text =~  /^blank$/i   ### todo/fix -- remove!!! add real blanks!!
+               t[0] = :BLANK
+            else
+            end
+        end
+
         #################
         ## pass 2                  
         ##    transform tokens (using simple patterns) 
@@ -227,13 +242,22 @@ def tokenize_with_errors
           end
 
 
-          if buf.match?( :TEXT, [:SCORE, :VS, :'-'], :TEXT )
+          if buf.match?( :TEXT, [:SCORE, :SCORE_MORE, :VS, :'-'], :TEXT )
              nodes << [:TEAM, buf.next[1]]
              nodes << buf.next
              nodes << [:TEAM, buf.next[1]]
           elsif buf.match?( :TEXT, :MINUTE )
              nodes << [:PLAYER, buf.next[1]]
              nodes << buf.next
+          elsif buf.match?( :DATE, :TIME )   ## merge DATE TIME into DATETIME
+               date = buf.next[1]
+               time = buf.next[1]
+               ## puts "DATETIME:"
+               ## pp date, time
+               val =  [date[0] + ' ' + time[0],  ## concat string of two tokens
+                        { date: date[1], time: time[1] }
+                      ]
+               nodes << [:DATETIME, val]
           else
              ## pass through
              nodes << buf.next
@@ -253,7 +277,8 @@ def tokenize_with_errors
          end
 
          tokens  += tok 
-         tokens  << [:NEWLINE, "\n"]   ## auto-add newlines
+         ## auto-add newlines  (unless BLANK!!)
+         tokens  << [:NEWLINE, "\n"]   unless tok[0][0] == :BLANK
     end
 
     [tokens,errors]
@@ -267,6 +292,17 @@ def _tokenize_line( line )
 
   puts "line: >#{line}<"    if debug?
 
+
+   ### special case for empty line (aka BLANK)
+   if line.empty?
+       ## note - blank always resets parser mode to std/top-level!!!
+       @re = RE
+
+       tokens << [:BLANK, '<|BLANK|>']
+       return [tokens, errors]
+   end
+
+
   pos = 0
   ## track last offsets - to report error on no match
   ##   or no match in end of string
@@ -353,7 +389,9 @@ def _tokenize_line( line )
              puts "!!! TOKENIZE ERROR (PROP_RE) - no match found"
              nil 
          end
-      else  ## assume TOP_LEVEL (a.k.a. RE) machinery
+      ###################################################
+      ## assume TOP_LEVEL (a.k.a. RE) machinery
+      else  
         if m[:space] || m[:spaces]
            nil   ## skip space(s)
         elsif m[:prop_key]
@@ -372,6 +410,11 @@ def _tokenize_line( line )
           else
              [:STATUS, [m[:status], {status: m[:status] } ]]
           end
+        elsif m[:note]
+            ###  todo/check:
+            ##      use value hash - why? why not? or simplify to:
+            ##  [:NOTE, m[:note]] 
+             [:NOTE, [m[:note], {note: m[:note] } ]]
         elsif m[:time]
               ## unify to iso-format
               ###   12.40 => 12:40
@@ -420,7 +463,7 @@ def _tokenize_line( line )
         elsif m[:num]   ## fix - change to ord (for ordinal number!!!)
               ## note -  strip enclosing () and convert to integer
              [:ORD, [m[:num], { value: m[:value].to_i(10) } ]]
-        elsif m[:score]
+        elsif m[:score_more]
               score = {}
               ## check for pen
               score[:p] = [m[:p1].to_i(10),
@@ -433,8 +476,15 @@ def _tokenize_line( line )
                             m[:ht2].to_i(10)]  if m[:ht1] && m[:ht2]
 
             ## note - for debugging keep (pass along) "literal" score
-            [:SCORE, [m[:score], score]]
-        elsif m[:minute]
+            [:SCORE_MORE, [m[:score_more], score]]
+        elsif m[:score]
+            score = {}
+            ## must always have ft for now e.g. 1-1 or such
+            score[:ft] = [m[:ft1].to_i(10),
+                          m[:ft2].to_i(10)]  
+          ## note - for debugging keep (pass along) "literal" score
+          [:SCORE, [m[:score], score]]
+      elsif m[:minute]
               minute = {}
               minute[:m]      = m[:value].to_i(10)
               minute[:offset] = m[:value2].to_i(10)   if m[:value2]
@@ -459,7 +509,10 @@ def _tokenize_line( line )
           when '|' then [:'|']
           when '[' then [:'[']
           when ']' then [:']']
-          when '-' then [:'-']
+          when '-' then [:'-']        # level 1 OR (classic) dash
+          when '--'   then [:'--']    # level 2
+          when '---'  then [:'---']   # level 3
+          when '----' then [:'----']  # level 4
           else
             nil  ## ignore others (e.g. brackets [])
           end
@@ -504,7 +557,7 @@ def _tokenize_line( line )
         puts "  LEAVE PROP_RE MODE, BACK TO TOP_LEVEL/RE"  if debug?
         @re = RE 
         ## note - auto-add PROP_END (<PROP_END>)
-        tokens << [:PROP_END, "<PROP_END>"]    
+        tokens << [:PROP_END, "<|PROP_END|>"]    
      end
    end