sportdb-parser 0.5.9 → 0.6.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 43f5fed1d5709a2bd2642046f1e3d367212c5ebcc71084f1a17b00738e5036de
-  data.tar.gz: 3082d8cf6879adb3735e2dc28a9213dbc5f0781438d0e4248244ac10d02085b5
+  metadata.gz: b299ddece5e64b86bb7ee6b55578099b0624b11d8e5f10721363f45d6ef5d8d8
+  data.tar.gz: 5712c99b200e6116c9f07fba1215a4bf2560e5bd848c3c8cc48959aa17997b85
 SHA512:
-  metadata.gz: 50176108fb5d9c81ce7234a7abbdbc4e29aaf25af37005d442d0b351ea699c93dee02293d975a340d2e9f03a13c76836aa30d0278e6a6a82ba28fbbba313f50c
-  data.tar.gz: 90b3f15722d7540f70ebb10718a643473e19de09a15e014f70e65b815ff4ed014c4c953266918dbca2baf34da94bd5681e403d27cc2b97f643781bf6f714e23f
+  metadata.gz: 5d2fce54482e12542c35abd46a292d7f5e6b4db894bba3a7f911269f116d9fa530653d36ace4295e2f819bb974093b5567a5494a2d50b54ee3f250b314d40a73
+  data.tar.gz: 7b6ef8aaafa2d20c0356fcdc048211f24a04cc4f95819ad8d225b2c9a4a29e44d8f415190acfbe3e31b2f9cc457a12f8e75c460394e984d5b9b1f476f0f8e30f

data/CHANGELOG.md

@@ -1,4 +1,4 @@
-### 0.5.9
+### 0.6.1
 ### 0.0.1 / 2024-07-12
 
 * Everything is new. First release.

data/Manifest.txt

@@ -14,6 +14,8 @@ lib/sportdb/parser/parser.rb
 lib/sportdb/parser/racc_parser.rb
 lib/sportdb/parser/racc_tree.rb
 lib/sportdb/parser/token-date.rb
+lib/sportdb/parser/token-minute.rb
+lib/sportdb/parser/token-prop.rb
 lib/sportdb/parser/token-score.rb
 lib/sportdb/parser/token-status.rb
 lib/sportdb/parser/token-text.rb

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
@@ -247,8 +271,14 @@ def tokenize_with_errors
     ## flatten tokens
     tokens = []
     tokens_by_line.each do |tok|
+
+         if debug?
+           pp tok
+         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]
@@ -260,7 +290,18 @@ def _tokenize_line( line )
   tokens = []
   errors = []   ## keep a list of errors - why? why not?
 
-  puts ">#{line}<"    if debug?
+  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
@@ -275,10 +316,10 @@ def _tokenize_line( line )
 
 
   while m = @re.match( line, pos )
-    if debug?
-      pp m
-      puts "pos: #{pos}"
-    end
+#    if debug?
+#      pp m
+#      puts "pos: #{pos}"
+#    end
     offsets = [m.begin(0), m.end(0)]
 
     if offsets[0] != pos
@@ -298,7 +339,7 @@ def _tokenize_line( line )
 
     pos = offsets[1]
 
-    pp offsets   if debug?
+#    pp offsets   if debug?
 
     ##
     ## note: racc requires pairs e.g. [:TOKEN, VAL]
@@ -306,12 +347,8 @@ def _tokenize_line( line )
 
 
   t = if @re == PROP_RE
-         if m[:space]
-              ## skip space
-              nil
-         elsif m[:spaces]
-              ## skip spaces
-              nil
+         if m[:space] || m[:spaces]
+              nil    ## skip space(s)
          elsif m[:prop_name]
                if m[:name] == 'Y'
                  [:YELLOW_CARD, m[:name]]
@@ -339,11 +376,11 @@ def _tokenize_line( line )
             when '(' then [:'(']
             when ')' then [:')']
             when '-' then [:'-']
-            when '.' then 
-                ## switch back to top-level mode!!
-                puts "  LEAVE PROP_RE MODE, BACK TO TOP_LEVEL/RE"  if debug?
-                @re = RE 
-                [:'.']
+           # when '.' then 
+           #     ## switch back to top-level mode!!
+           #     puts "  LEAVE PROP_RE MODE, BACK TO TOP_LEVEL/RE"  if debug?
+           #     @re = RE 
+           #     [:'.']
             else
               nil  ## ignore others (e.g. brackets [])
             end
@@ -353,12 +390,8 @@ def _tokenize_line( line )
              nil 
          end
       else  ## assume TOP_LEVEL (a.k.a. RE) machinery
-        if m[:space]
-           ## skip space
-           nil
-        elsif m[:spaces]
-           ## skip spaces
-           nil
+        if m[:space] || m[:spaces]
+           nil   ## skip space(s)
         elsif m[:prop_key]
            ##  switch context  to PROP_RE
            @re = PROP_RE
@@ -397,6 +430,7 @@ def _tokenize_line( line )
  ## 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]
+            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]
@@ -417,10 +451,12 @@ def _tokenize_line( line )
             duration[:end][:wday] = DAY_MAP[ m[:day_name2].downcase ]   if m[:day_name2]
             ## note - for debugging keep (pass along) "literal" duration
             [:DURATION, [m[:duration], duration]]
+        elsif m[:wday]    ## standalone weekday e.g. Mo/Tu/We/etc.
+             [:WDAY, [m[:wday], { wday: DAY_MAP[ m[:day_name].downcase ] } ]]
         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 +469,15 @@ 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]
@@ -454,11 +497,15 @@ def _tokenize_line( line )
           case sym
           when ',' then [:',']
           when ';' then [:';']
+          when '/' then [:'/']
           when '@' then [:'@']
           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
@@ -472,11 +519,11 @@ def _tokenize_line( line )
 
     tokens << t    if t
 
-    if debug?
-      print ">"
-      print "*" * pos
-      puts "#{line[pos..-1]}<"
-    end
+#    if debug?
+#      print ">"
+#      print "*" * pos
+#      puts "#{line[pos..-1]}<"
+#    end
   end
 
   ## check if no match in end of string
@@ -489,6 +536,24 @@ def _tokenize_line( line )
   end
 
 
+   ##
+   ## if in prop mode continue if   last token is [,-]
+   ##        otherwise change back to "standard" mode
+   if @re == PROP_RE
+     if [:',', :'-'].include?( tokens[-1][0] )
+        ## continue/stay in PROP_RE mode
+        ##  todo/check - auto-add PROP_CONT token or such
+        ##                to help parser with possible NEWLINE
+        ##                  conflicts  - why? why not?
+     else
+        ## switch back to top-level mode!!
+        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|>"]    
+     end
+   end
+  
   [tokens,errors]
 end