sportdb-parser 0.6.13 → 0.6.15

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 4a2e9867bfc8e1dec97214c0d3c335fe0d3cdb0cf5835184081af1447d509e4a
-  data.tar.gz: 1b038a316ad0de74bc559972e59832bfe3af5530b6c0547a9b2eaca0ecbfefbf
+  metadata.gz: d907a1b09e04c9c17884afe7881961b597b547ca9ab742ecb6e06f89f8bfe151
+  data.tar.gz: da287bc13de7217bf3c6a2ffcfdd8d7f3d376baff413b79a58dd544bb2bfcc3a
 SHA512:
-  metadata.gz: 11b6cb434a959e39b8dd8a935b9bb0d973e9e47e647c1f19dbc1c536a013b982dc948ca62c2d7a9a661a7aeb276444cfbfb81c4975df30ece993076124fce7bd
-  data.tar.gz: b0487df7f949c28c0d4df23d040aeff4eb3247c01b088fe45b65aa6a29a3891781d24bd56d85e99cd1e0d31d82efe953c13f6c689301a9623a01ffa2bdbb45f0
+  metadata.gz: 3be1466a2d4ef5a5d1129f6b0fa9f67c6258c5a1cc60aab831ac5c4c7121e691fa64a9f33e71508d91b9809bcdc1b77b152fc2f7c1580172937662865e8ee33e
+  data.tar.gz: fe14835fc3195b5e441bdd3f9763bfe1303aa471e041a908cafc027becfbfd5ffd2e9696948e2dd88647ef7de88ebabf39674e42ba6656144e478150822e699e

data/CHANGELOG.md

@@ -1,4 +1,4 @@
-### 0.6.13
+### 0.6.15
 ### 0.0.1 / 2024-07-12
 
 * Everything is new. First release.

data/config/rounds_es.txt

@@ -17,6 +17,9 @@ Zona B
 Final de Grupos
 Repechaje
 
+Final Absoluta
+
+
 Copa Libertadores
 Copa Sudamericana
 

data/lib/sportdb/parser/lang.rb

@@ -17,12 +17,17 @@ module Lang
 ## Group 1A or A1, B1  - used anywhere
 ##   yes - A1, A2, B1, C1, etc. used in UEFA Nations League for example!!
 ##
+##     exlcude
 ##  use "key" of group - why? why not?
+##
+##  note - will include group stage too
+##           make sure is_round gets called before is_group for now!!!
 
 GROUP_RE = %r{^
                 Group [ ]
-                   (?<key>[a-z0-9]+)
+                   (?<key> [a-z0-9]+ )
               $}ix
+
 def self.is_group?( text )
    ## use regex for match
    GROUP_RE.match?( text )

data/lib/sportdb/parser/lexer.rb

@@ -182,13 +182,16 @@ def tokenize_with_errors
         ## pass 1
         ##   replace all texts with keyword matches
         ##     (e.g. group, round, leg, etc.)
+        ##
+        ##   note - let is_round? get first (before is_group?)
+        ##            will match group stage  as round (NOT group)
         tokens = tokens.map do |t|        
                     if t[0] == :TEXT
                        text = t[1]
-                       t = if is_group?( text )
+                       t =  if is_round?( text ) || is_leg?( text ) || is_zone?( text )
+                               [:ROUND, text]   
+                            elsif is_group?( text )
                                [:GROUP, text]
-                             elsif is_round?( text ) || is_leg?( text ) || is_zone?( text )
-                               [:ROUND, text]
                              else
                                t  ## pass through as-is (1:1)
                              end
@@ -323,8 +326,7 @@ def _tokenize_line( line )
   
     ##  start with prop key (match will switch into prop mode!!!)
     ##   - fix - remove leading spaces in regex (upstream) - why? why not?
-    m = PROP_KEY_RE.match( line )
-    if m
+    if (m = PROP_KEY_RE.match( line ))
       ###  switch into new mode
       ##  switch context  to PROP_RE
         puts "  ENTER PROP_RE MODE"   if debug?
@@ -357,10 +359,15 @@ def _tokenize_line( line )
 
         offsets = [m.begin(0), m.end(0)]
         pos = offsets[1]    ## update pos
-    end
+    elsif (m = ROUND_OUTLINE_RE.match( line ))
+      puts "   ROUND_OUTLINE"  if debug?
+
+      tokens << [:ROUND_OUTLINE, m[:round_outline]]
 
-    m = PLAYER_WITH_SCORE_RE.match( line )
-    if m
+      ## note - eats-up line for now (change later to only eat-up marker e.g. »|>>)
+      offsets = [m.begin(0), m.end(0)]
+      pos = offsets[1]    ## update pos
+    elsif (m = PLAYER_WITH_SCORE_RE.match( line ))
       ##  switch context to GOAL_RE (goalline(s)
       ##   split token (automagically) into two!! - player AND minute!!!
       @re = GOAL_RE
@@ -380,10 +387,7 @@ def _tokenize_line( line )
   
       offsets = [m.begin(0), m.end(0)]
       pos = offsets[1]    ## update pos
-    end 
-
-    m = PLAYER_WITH_MINUTE_RE.match( line )
-    if m
+    elsif (m = PLAYER_WITH_MINUTE_RE.match( line ))
       ##  switch context to GOAL_RE (goalline(s)
       ##   split token (automagically) into two!! - player AND minute!!!
       @re = GOAL_RE
@@ -396,8 +400,6 @@ def _tokenize_line( line )
       ##    todo - find a better way? how possible?
       tokens << [:NONE, "<|NONE|>"]   if m[:none]
       
-
-
       ## auto-add player token first
       tokens << [:PLAYER, m[:name]]
       ## minute props
@@ -412,6 +414,7 @@ def _tokenize_line( line )
     end
   end
 
+  
 
   old_pos = -1   ## allows to backtrack to old pos (used in geo)
 
@@ -462,6 +465,8 @@ def _tokenize_line( line )
    
             case sym
             when ',' then [:',']
+            when '›' then [:',']  ## note - treat geo sep › (unicode) like comma for now!!!
+            when '>' then [:',']  ## note - treat geo sep > (ascii) like comma for now!!!
             when '[' then
                  ## get out-off geo mode and backtrack (w/ next)
                  puts "  LEAVE GEO_RE MODE, BACK TO TOP_LEVEL/RE"  if debug?