sportdb-parser 0.2.1 → 0.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 3657cedc5125ee2515efa8be4a1838d05b7290523dd893f7eba5b87024e71238
-  data.tar.gz: caf6d7e909e17fa0dcabf659ab8f5046ca1940d8f7c1c6f5312e485dc0089384
+  metadata.gz: 9ebb468318f2b87c33ca66afb6c46611ce5f420258e0c41b40a2cbfabcff7a49
+  data.tar.gz: 0cf1d511f3e936d73531442d1ca6bef94d90a50ae65346b5b57347d4d294dc77
 SHA512:
-  metadata.gz: 4063565aada304a1eb96009b6fe542392f41a55d4ad4d21b5de156004bd69a055c5f86b076bed1defbe50423c8c891dd538931ea6ca9b8ec41e237c23e699219
-  data.tar.gz: 91f6476810cb6617dfcd703ada57592cd38b87f3b4b9fc6fd4468a9457ff0e6ae6337a4e4f5c782e1b80f5f6b6015d5ce26ed6330915cd67a5fb6606f665017f
+  metadata.gz: e992ab97d7ae18c514de14078d30eb36adc40f5044242ce9ace089fb88b104c61b29ff86a2aa8101bb7257c3ff2ce32c6150439ff855e195bee1b26032bb0d9d
+  data.tar.gz: 25e66e45e7daf2783bc6507a3cb2c660d9153eab9530210ef51ef6e0d5d3fc531e5891897be3b0492b0ad7ea5fe3d406a0a3dd0559549b85518360d442ed4d8b

data/CHANGELOG.md

@@ -1,4 +1,4 @@
-### 0.2.1
+### 0.3.0
 
 ### 0.0.1 / 2024-07-12
 

data/Manifest.txt

@@ -2,11 +2,10 @@ CHANGELOG.md
 Manifest.txt
 README.md
 Rakefile
-bin/fbt
+bin/fbtok
 lib/sportdb/parser.rb
 lib/sportdb/parser/lang.rb
 lib/sportdb/parser/linter.rb
-lib/sportdb/parser/opts.rb
 lib/sportdb/parser/outline_reader.rb
 lib/sportdb/parser/parser.rb
 lib/sportdb/parser/token-date.rb

data/bin/{fbt → fbtok}

@@ -1,36 +1,25 @@
 #!/usr/bin/env ruby
 
 ## tip: to test run:
-##   ruby -I ./lib bin/fbt
+##   ruby -I ./lib bin/fbtok
 
-## our own code
 require 'sportdb/parser'
 
 
+require 'optparse'   ## check - already auto-required in cocos? keep? why? why not?
 
-require 'optparse'
 
-##
-## read textfile
-##   and dump tokens
-##
-##   fbt  ../openfootball/.../euro.txt
+args=ARGV
 
 
+opts = {
+    debug: true,
+    metal: false,
+}
 
-
- args = ARGV
- opts = { debug: false,
-          metal: false }
-
- parser = OptionParser.new do |parser|
+parser = OptionParser.new do |parser|
   parser.banner = "Usage: #{$PROGRAM_NAME} [options]"
 
-##
-## check if git has a offline option?? (use same)
-##             check for other tools - why? why not?
-
-
   parser.on( "--verbose", "--debug",
                "turn on verbose / debug output (default: #{opts[:debug]})" ) do |debug|
     opts[:debug] = debug
@@ -50,29 +39,12 @@ puts "ARGV:"
 p args
 
 
-
-
-
-paths =  if args.empty?
-            [
-              '../../../openfootball/euro/2021--europe/euro.txt',
-              '../../../openfootball/euro/2024--germany/euro.txt',
-            ]
-         else
-            ## check for directories
-            ##   and auto-expand
-
-            SportDb::Parser::Opts.expand_args( args )
-         end
-
-
-
 SportDb::Parser::Linter.debug = true    if opts[:debug]
 
 linter = SportDb::Parser::Linter.new
-
 errors = []
 
+paths = args
 paths.each_with_index do |path,i|
     puts "==> [#{i+1}/#{paths.size}] reading >#{path}<..."
     linter.read( path, parse: !opts[:metal] )
@@ -90,5 +62,6 @@ else
     puts "OK   no parse errors found in #{paths.size} datafile(s)"
 end
 
+
 puts "bye"
 

data/lib/sportdb/parser/linter.rb

@@ -5,11 +5,11 @@ class Parser
 ###
 ## note - Linter for now nested inside Parser - keep? why? why not?
 class Linter
-   
+
 def self.debug=(value) @@debug = value; end
 def self.debug?() @@debug ||= false; end  ## note: default is FALSE
-def debug?()  self.class.debug?; end   
-    
+def debug?()  self.class.debug?; end
+
 
 
 attr_reader :errors
@@ -35,7 +35,7 @@ def errors?() @errors.size > 0; end
   ##   Group B:   - remove colon
   ##    or lookup first
 
-  ATTRIB_RE = %r{^  
+  ATTRIB_RE = %r{^
                    [ ]*?     # slurp leading spaces
                 (?<key>[^:|\]\[()\/; -]
                        [^:|\]\[()\/;]{0,30}
@@ -50,12 +50,12 @@ def errors?() @errors.size > 0; end
 
 #########
 ## parse - false (default) - tokenize (only)
-##       - true            - tokenize & parse                
+##       - true            - tokenize & parse
 def read( path, parse: false )
   ## note: every (new) read call - resets errors list to empty
   @errors = []
 
-  nodes = OutlineReader.read( path ) 
+  nodes = OutlineReader.read( path )
 
   ##  process nodes
   h1 = nil
@@ -66,7 +66,7 @@ def read( path, parse: false )
 
   nodes.each do |node|
     type = node[0]
-    
+
     if type == :h1
         h1 = node[1]  ## get heading text
         puts
@@ -74,14 +74,14 @@ def read( path, parse: false )
     elsif type == :p
 
        if h1.nil?
-         orphans += 1    ## only warn once 
+         orphans += 1    ## only warn once
          puts "!! WARN - no heading for #{orphans} text paragraph(s); skipping parse"
          next
        end
 
        lines = node[1]
 
-       tree = [] 
+       tree = []
        lines.each_with_index do |line,i|
 
         if debug?
@@ -91,10 +91,10 @@ def read( path, parse: false )
 
 
         ## skip new (experimental attrib syntax)
-        if attrib_found == false && 
+        if attrib_found == false &&
             ATTRIB_RE.match?( line )
           ## note: check attrib regex AFTER group def e.g.:
-          ##         Group A: 
+          ##         Group A:
           ##         Group B:  etc.
           ##     todo/fix - change Group A: to Group A etc.
           ##                       Group B: to Group B
@@ -107,17 +107,17 @@ def read( path, parse: false )
           ## check if line ends with dot
           ##  if not slurp up lines to the next do!!!
           ## logger.debug "skipping key/value line - >#{line}<"
-          attrib_found = false   if line.end_with?( '.' ) 
+          attrib_found = false   if line.end_with?( '.' )
               # logger.debug "skipping key/value line (cont.) - >#{line}<"
               next
-        end    
-    
+        end
+
         t, error_messages  =  if parse
                                   @parser.parse_with_errors( line )
                               else
-                                  @parser.tokenize_with_errors( line )                   
+                                  @parser.tokenize_with_errors( line )
                               end
-         
+
 
          if error_messages.size > 0
             ## add to "global" error list
@@ -134,7 +134,7 @@ def read( path, parse: false )
 
          tree << t
        end
-       
+
        ## pp tree
     else
         pp node
@@ -146,4 +146,4 @@ end  # class Linter
 
 
 end   # class Parser
-end   # module SportDb
+end   # module SportDb

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

@@ -155,6 +155,35 @@ DATE_RE = Regexp.union(
 )
 
 
+##
+##  add a date parser helper
+def self.parse_date( str, start: )
+    if m=DATE_RE.match( str )
+
+      year    = m[:year].to_i(10)  if m[:year]
+      month   = MONTH_MAP[ m[:month_name].downcase ]   if m[:month_name]
+      day     = m[:day].to_i(10)   if m[:day]
+      wday    = DAY_MAP[ m[:day_name].downcase ]   if m[:day_name]
+
+      if year.nil?   ## try to calculate year
+        year =  if  month > start.month ||
+                   (month == start.month && day >= start.day)
+                  # assume same year as start_at event (e.g. 2013 for 2013/14 season)
+                  start.year
+                else
+                  # assume year+1 as start_at event (e.g. 2014 for 2013/14 season)
+                  start.year+1
+                end
+      end
+      Date.new( year,month,day )
+    else
+      puts "!! ERROR - unexpected date format; cannot parse >#{str}<"
+      exit 1
+    end
+end
+
+
+
 ###
 #  date duration
 #   use - or + as separator

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

@@ -1,12 +1,12 @@
-module SportDb 
+module SportDb
 class Parser
-   
-   
+
+
 ##  note - do NOT allow single alpha text for now
-##   add later??      A - B    C - D  - why?        
+##   add later??      A - B    C - D  - why?
 ## opt 1) one alpha
-## (?<text_i> [a-z])    # only allow single letter text (not numbers!!)  
-    
+## (?<text_i> [a-z])    # only allow single letter text (not numbers!!)
+
 ## opt 2) more than one alphanum
 
 
@@ -26,19 +26,19 @@ class Parser
 
 
 TEXT_RE = %r{
-    ## must start with alpha (allow unicode letters!!)      
-    (?<text>    
-           ## positive lookbehind 
+    ## must start with alpha (allow unicode letters!!)
+    (?<text>
+           ## positive lookbehind
            ##  (MUST be fixed number of chars - no quantifier e.g. +? etc.)
             (?<=[ ,;@|\[\]]
                  |^
             )
-            (?:  
+            (?:
                 # opt 1 - start with alpha
                  \p{L}+    ## all unicode letters (e.g. [a-z])
                    |
 
-                # opt 2 - start with num!! - allow special case (e.g. 1. FC) 
+                # opt 2 - start with num!! - allow special case (e.g. 1. FC)
                      \d+  # check for num lookahead (MUST be space or dot)
                       ## MUST be followed by (optional dot) and
                       ##                      required space !!!
@@ -46,69 +46,79 @@ TEXT_RE = %r{
                       \.?     ## optional dot
                       [ ]?   ## make space optional too  - why? why not?
                              ##  yes - eg. 1st, 2nd, 5th etc.
-                       \p{L}+               
+                       \p{L}+
                )
-        
+
               (?:(?:  (?:[ ]
                      (?!vs?\.?[ ])    ## note - exclude (v[ ]/vs[ ]/v.[ ]/vs.[ ])
-                       )  
+                       )
                       |     # only single spaces allowed inline!!!
-                     [-]                                              
+                     [-]
                   )?
                 (?:
                   \p{L} |
-                  [&/'] 
+                  [&/']
                     |
                  (?:
-                   \d+ 
-                   (?![0-9.:h'/+-])   
+                   \d+
+                   (?![0-9.:h'/+-])
                    ## negative lookahead for numbers
                    ##   note - include digits itself!!!
-                 )|  
-                 \.   
-               )  
+                 )|
+                 \.
+               )
               )*  ## must NOT end with space or dash(-)
               ##  todo/fix - possible in regex here
               ##     only end in alphanum a-z0-9 (not dot or & ???)
 
-  
+
             ## allow optional at the end
             ##  tag or year
-            ##   make it and in the future - why? why not?  
-            ##   
+            ##   make it and in the future - why? why not?
+            ##
+            ## change - fix
+            ##   do NOT use (A) for amateur
+            ##   use A or A. with NO ()!!!
             ## (A) -    allow with predined  alpha only for now
             ##          e.g. (A) - amateur a team or b?
+            ###  same for U21 or U9 etc
+            ##        use with NO ()!!! - why? why not?
             ##      or U21 U9 etc.   - why? why not?
             ##       or etc.
             ## (1879-1893) or allow years e.g. (1879-1893)
-            ###     
-            (?:
-               [ ]  
-                  \( (?: 
-                       A|B|
-                       U\d{1,2} 
-                     )
-                  \) 
-            )? 
+            ###
+            ##    add allow country code three to five letters for now
+            ##       change to generic 1 to 5 - why? why not?
+            ##     e.g. (A), (I),
+            ##          (AUT)
+            ##          (TRNC)   five? for UEFA code for northern cyprus
+            ##     change to 1 to 4 - why? why not?
+            ##   check - fix possible for upper case only here
+            ##                     inline for this group only?
             (?:
-               [ ] 
+               [ ]
                \(
                   \d{4}-\d{4}
                \)
-            )?                    
-
+            )?
+             (?:
+               [ ]+   ## allow more than once space - why? why not?
+                  \( (?:
+                       [A-Z]{1,5}
+                     )
+                  \)
+             )?
             ## add lookahead/lookbehind
-           ##    must be space!!! 
+           ##    must be space!!!
            ##   (or comma or  start/end of string)
            ##   kind of \b !!!
             ## positive lookahead
             (?=[ ,;@|\[\]]
                  |$
             )
-   )   
+   )
 }ix
 
 
 end # class Parser
-end # module SportDb 
-   
+end # module SportDb

data/lib/sportdb/parser/version.rb

@@ -3,8 +3,8 @@ module SportDb
   module Module
     module Parser
   MAJOR = 0    ## todo: namespace inside version or something - why? why not??
-  MINOR = 2
-  PATCH = 1
+  MINOR = 3
+  PATCH = 0
   VERSION = [MAJOR,MINOR,PATCH].join('.')
 
   def self.version

data/lib/sportdb/parser.rb

@@ -24,11 +24,11 @@ require_relative 'parser/lang'
 require_relative 'parser/parser'
 
 
-## more
+####
+##  todo/check - move outline reader upstream to cocos - why? why not?
+##       use  read_outline(), parse_outline()  - why? why not?
 require_relative 'parser/outline_reader'
 require_relative 'parser/linter'
-require_relative 'parser/opts'
-
 
 ###
 #  make parser api (easily) available - why? why not?

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: sportdb-parser
 version: !ruby/object:Gem::Version
-  version: 0.2.1
+  version: 0.3.0
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-08-24 00:00:00.000000000 Z
+date: 2024-09-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: cocos
@@ -75,7 +75,7 @@ dependencies:
 description: sportdb-parser - football.txt match parser (& tokenizer)
 email: gerald.bauer@gmail.com
 executables:
-- fbt
+- fbtok
 extensions: []
 extra_rdoc_files:
 - CHANGELOG.md
@@ -86,11 +86,10 @@ files:
 - Manifest.txt
 - README.md
 - Rakefile
-- bin/fbt
+- bin/fbtok
 - lib/sportdb/parser.rb
 - lib/sportdb/parser/lang.rb
 - lib/sportdb/parser/linter.rb
-- lib/sportdb/parser/opts.rb
 - lib/sportdb/parser/outline_reader.rb
 - lib/sportdb/parser/parser.rb
 - lib/sportdb/parser/token-date.rb

data/lib/sportdb/parser/opts.rb

@@ -1,70 +0,0 @@
-
-module SportDb
-class Parser
-
-###
-## note - Opts Helpers for now nested inside Parser - keep here? why? why not?
-class Opts
-
-    SEASON_RE = %r{ (?:
-                       \d{4}-\d{2}
-                     | \d{4}(--[a-z0-9_-]+)?
-                    )
-                  }x
-    SEASON = SEASON_RE.source    ## "inline" helper for embedding in other regexes - keep? why? why not?
-
-
-    ## note: if pattern includes directory add here
-    ##     (otherwise move to more "generic" datafile) - why? why not?
-    MATCH_RE = %r{ (?: ^|/ )      # beginning (^) or beginning of path (/)
-                       #{SEASON}
-                     /[a-z0-9_-]+\.txt$  ## txt e.g /1-premierleague.txt
-                }x
-
-
-def self.find( path )
-    datafiles = []
-
-    ## note: normalize path - use File.expand_path ??
-    ##    change all backslash to slash for now
-    ## path = path.gsub( "\\", '/' )
-    path = File.expand_path( path )
-    
-    ## check all txt files
-    ## note: incl. files starting with dot (.)) as candidates 
-    ##     (normally excluded with just *)
-    candidates = Dir.glob( "#{path}/**/{*,.*}.txt" )
-    ## pp candidates
-    candidates.each do |candidate|
-      datafiles << candidate    if MATCH_RE.match( candidate )
-    end
-
-    ## pp datafiles
-    datafiles
-end
-
-
-def self.expand_args( args )
-    paths = []
-
-    args.each do |arg|
-        ## check if directory
-        if Dir.exist?( arg )
-            datafiles = find( arg )
-            puts
-            puts "  found #{datafiles.size} match txt datafiles in #{arg}"
-            pp datafiles
-            paths += datafiles
-        else
-              ## assume it's a file
-            paths << arg
-        end
-    end
-
-    paths
-end
-end  # class Opts
-
-
-end   # class Parser
-end   # module SportDb