sportdb-parser 0.6.15 → 0.6.17

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: d907a1b09e04c9c17884afe7881961b597b547ca9ab742ecb6e06f89f8bfe151
-  data.tar.gz: da287bc13de7217bf3c6a2ffcfdd8d7f3d376baff413b79a58dd544bb2bfcc3a
+  metadata.gz: 1ba31cc4284de6a4ea05615ad37a30546375c5e3218847a8f69c3c359074fb9c
+  data.tar.gz: 5b8c73196c6bd08a399c7687cd65f4e05f4c3a66580eaeecd6dbaad33837b822
 SHA512:
-  metadata.gz: 3be1466a2d4ef5a5d1129f6b0fa9f67c6258c5a1cc60aab831ac5c4c7121e691fa64a9f33e71508d91b9809bcdc1b77b152fc2f7c1580172937662865e8ee33e
-  data.tar.gz: fe14835fc3195b5e441bdd3f9763bfe1303aa471e041a908cafc027becfbfd5ffd2e9696948e2dd88647ef7de88ebabf39674e42ba6656144e478150822e699e
+  metadata.gz: cf4bc5b5a112effc59895c405e3484224d4d94aadf4771fecff90458b01aa63b43703632fdada70df9ee906aa8e517b1736dbdb2de795ba53b3b8e05d6c1ba4c
+  data.tar.gz: 46b6a97cb0af77debec3bd58ee24bd12b1236680dc53ff76a50b725621961bb9aabdfc151d9379d01672ebbef2423c3ce4f4d948adc506789c846491a7f30f15

data/CHANGELOG.md

@@ -1,4 +1,4 @@
-### 0.6.15
+### 0.6.17
 ### 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?
@@ -387,7 +401,12 @@ def _tokenize_line( line )
   
       offsets = [m.begin(0), m.end(0)]
       pos = offsets[1]    ## update pos
-    elsif (m = PLAYER_WITH_MINUTE_RE.match( line ))
+
+    ####  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
@@ -414,7 +433,7 @@ def _tokenize_line( line )
     end
   end
 
-  
+
 
   old_pos = -1   ## allows to backtrack to old pos (used in geo)
 

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

@@ -69,10 +69,12 @@ MINUTE_RE = %r{
 #                          or others with first matching position 
 #                          or if chars get eaten-up? 
 #                        let us know if \G is required here or not
+#
+##  note - use \A (instead of ^) - \A strictly matches the start of the string.
 
 
 PLAYER_WITH_MINUTE_RE = %r{
-           ^    ### note - MUST start line; leading spaces optional (eat-up)
+           \A    ### note - MUST start line; leading spaces optional (eat-up)
            [ ]*
              (?:      # optional open bracket ([) -- remove later
                 (?<open_bracket> \[ )
@@ -143,8 +145,11 @@ PLAYER_WITH_MINUTE_RE = %r{
 }ix
 
 
+
+##  note - use \A (instead of ^) - \A strictly matches the start of the string.
+
 PLAYER_WITH_SCORE_RE = %r{
-           ^    ### note - MUST start line; leading spaces optional (eat-up)
+           \A    ### note - MUST start line; leading spaces optional (eat-up)
            [ ]*
    (?<player_with_score>
                    (?<score>

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/token-text.rb

@@ -60,6 +60,10 @@ TEXT_RE = %r{
                     1/ \d{1,2} [ ] \p{L}+
                   |
                 ## opt 4 - add another weirdo case
+                ##   e.g.   's Gravenwezel-Schilde
+                    '[s]
+                  |
+                ## opt 5 - add another weirdo case
                 ##   e.g. 5.-8. Platz Playoffs  - keep - why? why not?
                     \d+\.-\d+\.  [ ]? \p{L}+                 
                )
@@ -70,18 +74,27 @@ TEXT_RE = %r{
                                ##    AND switch to case-sensitive (via -i!!!)
                       )
                       |     # only single spaces allowed inline!!!
-                     [-/]
+                     [_/]
                   )?
                 (?:
-                  \p{L} |
+                  \p{L} 
+                     |
                   [&'°]
-                    |
+                     |
+                   (?:  (?<! [ ])   ## todo - check regex - make sure lookbehind is always first/before!!
+                      [-]  ### allow e.g. Sport-  if lookbehind is unicode letter or dot (.)
+                          ###       or    U.N.A.M.-Pumas
+                          ##         (?<= [\p{L}.] )
+                            ## try more flexible (use negative lookbehind - no space)   
+                    )
+                     |
                  (?:
                    \d+
                    (?!
-                     [0-9h'+-] |    ## protected break on 12h / 12' / 1-1
-                                    ##  check usege for 3+4 - possible? where ? why?
-                     (?:[.:]\d)     ## protected/exclude/break on 12.03 / 12:03
+                     [0-9h'+] |    ## protected break on 12h / 12' / 1-1
+                                    ##  check usege for 3+4 - possible? where ? why?     
+                     (?:[.:-]\d)     ## protected/exclude/break on 12.03 / 12:03 / 12-12
+                                     ##  BUT allow Park21-Arena for example e.g. 21-A :-)
                     )
                    ## negative lookahead for numbers
                    ##   note - include digits itself!!!

data/lib/sportdb/parser/token.rb

@@ -199,7 +199,8 @@ PROP_GOAL_RE =  Regexp.union(
 
 ####
 # 
-ROUND_OUTLINE_RE = %r{  ^
+##  note - use \A (instead of ^) - \A strictly matches the start of the string.
+ROUND_OUTLINE_RE = %r{   \A
                            [ ]*  ## ignore leading spaces (if any)
                          (?: »|>> ) 
                            [ ]+

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 = 15
+  PATCH = 17
   VERSION = [MAJOR,MINOR,PATCH].join('.')
 
   def self.version

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: sportdb-parser
 version: !ruby/object:Gem::Version
-  version: 0.6.15
+  version: 0.6.17
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2025-03-09 00:00:00.000000000 Z
+date: 2025-03-10 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: cocos