sportdb-parser 0.6.14 → 0.6.16

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 51b568bfcb315049faf125718d09615346959cdd7301934a5131cb0fac1b6f9b
-  data.tar.gz: c8bd6486e70d28d4b121a3fb1ebec206b46880571a2404de64418fd76fe8b039
+  metadata.gz: 3da9280d27bf1e4662eb10f9451679e4aace18b9a0e1bfa29dd1e7b6bcbdc5e5
+  data.tar.gz: e6786f648848cd075ef3e0f6d8d7fda2d31743f989653c0fcf2312a33a223357
 SHA512:
-  metadata.gz: 1236c350c6b2cef3a7b9caff9b943ef09d69a02e5e6f2fe3ed55c5677d4a20ed17bba6423c77d035a4ea4c772bdaf598a81ee0738f25185ad2bbf5b63ab4e901
-  data.tar.gz: a16da2520f22b6392330bfbec90e2c8c2ca0f784bb31556ddb0140e25d53d7a293d6ca615b4a2a08d426e0069b0f6dcab8ed06ba35e688a03befa6429d9c732d
+  metadata.gz: 04250d17d120c12dc0b3980ff971b02fa178e617f35af70651f86011d9f5d4cad1d81df84a1f5af97ab73cb9023cc6cb190b13c420af71f3bcb2af7df6a526f1
+  data.tar.gz: 120486063a9a82891a63914654965b799aef774680695de8bda3bb52894399d0800b98efb852f87f672dbe303dc6c415b91a10a094989e94ddf3e319b3183cc9

data/CHANGELOG.md

@@ -1,4 +1,4 @@
-### 0.6.14
+### 0.6.16
 ### 0.0.1 / 2024-07-12
 
 * Everything is new. First release.

data/lib/sportdb/parser/lexer.rb

@@ -290,6 +290,20 @@ end   # method tokenize_with_errors
 
 
 
+### add a QUICK_PLAYER_WITH_MINUTE  check
+QUICK_PLAYER_WITH_MINUTE_RE = %r{
+      \b
+         \d{1,3}      ## constrain numbers to 0 to 999!!!
+        (?: (?: 
+                \+\d{1,3}   
+            )?
+            |
+            (?: \?{2} | _{2} )  ## add support for n/a (not/available)
+        )           
+        '   ## must have minute marker!!!!
+}ix 
+ 
+
 def _tokenize_line( line )
   tokens = []
   errors = []   ## keep a list of errors - why? why not?
@@ -326,8 +340,7 @@ 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?
@@ -360,10 +373,7 @@ def _tokenize_line( line )
 
         offsets = [m.begin(0), m.end(0)]
         pos = offsets[1]    ## update pos
-    end
-
-    m = ROUND_OUTLINE_RE.match( line )
-    if m
+    elsif (m = ROUND_OUTLINE_RE.match( line ))
       puts "   ROUND_OUTLINE"  if debug?
 
       tokens << [:ROUND_OUTLINE, m[:round_outline]]
@@ -371,10 +381,7 @@ def _tokenize_line( line )
       ## 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
-    end
-
-    m = PLAYER_WITH_SCORE_RE.match( line )
-    if m
+    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
@@ -394,10 +401,12 @@ 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
+    ####  FIX/FIX/TODO
+    ### looks to hang in player with minute 
+    ###  FIX - improve / rework PLAYER_WITH_MINUTE_RE  regex!!!!
+    elsif (_quick = QUICK_PLAYER_WITH_MINUTE_RE.match(line) &&
+                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
@@ -410,8 +419,6 @@ 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
@@ -427,6 +434,7 @@ def _tokenize_line( line )
   end
 
 
+
   old_pos = -1   ## allows to backtrack to old pos (used in geo)
 
   while m = @re.match( line, pos )

data/lib/sportdb/parser/token-score.rb

@@ -43,6 +43,28 @@ class Lexer
                 ## todo/check:  remove loakahead assertion here - why require space?
                 ## note: \b works only after non-alphanum e.g. )
 
+   ####
+   ## support short all-in-one e.g.
+   ##  e.g.      3-4 pen. 2-2 a.e.t. (1-1, 1-1) becomes
+   ##   3-4 pen. (2-2, 1-1, 1-1)            
+         
+   SCORE__P_ET_FT_HT_V2__RE = %r{
+          (?<score_more>
+               \b
+                (?<p1>\d{1,2}) - (?<p2>\d{1,2})
+                   [ ]* #{P_EN} [ ]+       
+                   \(
+               (?<et1>\d{1,2}) - (?<et2>\d{1,2})
+                   [ ]*, [ ]*
+               (?<ft1>\d{1,2}) - (?<ft2>\d{1,2})
+                   [ ]*, [ ]*
+               (?<ht1>\d{1,2}) - (?<ht2>\d{1,2})
+                   [ ]*
+                \)
+               (?=[ ,\]]|$)
+            )}ix       ## todo/check:  remove loakahead assertion here - why require space?
+                               ## note: \b works only after non-alphanum e.g. )
+
 
     ## e.g. 3-4 pen. 2-2 a.e.t. (1-1, 1-1)  or
     ##      3-4p 2-2aet (1-1, )     or
@@ -128,7 +150,8 @@ class Lexer
 ## check - find a better name for SCORE_MORE - SCORE_EX, SCORE_BIG, or ___ - why? why not?
 
 SCORE_MORE_RE = Regexp.union(
-  SCORE__P_ET_FT_HT__RE,  # e.g. 5-1 pen. 2-2 a.e.t. (1-1, 1-0)
+  SCORE__P_ET_FT_HT_V2__RE,  # e.g. 5-1 pen. (2-2, 1-1, 1-0)  
+  SCORE__P_ET_FT_HT__RE,    # e.g. 5-1 pen. 2-2 a.e.t. (1-1, 1-0)
   SCORE__P_FT_HT__RE,     # e.g. 5-1 pen. (1-1)
   SCORE__P_ET__RE,        # e.g. 2-2 a.e.t.  or  5-1 pen. 2-2 a.e.t.
   SCORE__P__RE,           # e.g. 5-1 pen.

data/lib/sportdb/parser/version.rb

@@ -4,7 +4,7 @@ module SportDb
     module Parser
   MAJOR = 0    ## todo: namespace inside version or something - why? why not??
   MINOR = 6
-  PATCH = 14
+  PATCH = 16
   VERSION = [MAJOR,MINOR,PATCH].join('.')
 
   def self.version

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: sportdb-parser
 version: !ruby/object:Gem::Version
-  version: 0.6.14
+  version: 0.6.16
 platform: ruby
 authors:
 - Gerald Bauer