rsssf-parser 0.0.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: 1b6cfe7842f0f46d242c1c2fc1f52b4c032b5c25fce314939583c7f96a486c65
+  data.tar.gz: ba5244b284f65129dca3b35e87d10984e1bf8906e571b3e42e85a4615eecb733
+SHA512:
+  metadata.gz: 9a63d121c858e35f757b59c490fc05cfd1457ac5c6e3294a291db787da141061f046c5ce2342fdf275e64bbb647934ef43d8547c2aa53eef537d12405d746185
+  data.tar.gz: e7a8f41d2d53e63fb72e35f22b1a0a0da370e15ab1b4aedf2ae1f37a2a1bc277ff31deb8362e541bbc403f7d49ea3f1120b4b271b9d36982359141ca80617d72

data/CHANGELOG.md

@@ -0,0 +1,3 @@
+### 0.0.1 / 2024-07-17
+
+* Everything is new. First release.

data/Manifest.txt

@@ -0,0 +1,15 @@
+CHANGELOG.md
+Manifest.txt
+README.md
+Rakefile
+bin/rsssf
+lib/rsssf/parser.rb
+lib/rsssf/parser/linter.rb
+lib/rsssf/parser/parser.rb
+lib/rsssf/parser/token-date.rb
+lib/rsssf/parser/token-goals.rb
+lib/rsssf/parser/token-note.rb
+lib/rsssf/parser/token-round.rb
+lib/rsssf/parser/token-score.rb
+lib/rsssf/parser/token-text.rb
+lib/rsssf/parser/token.rb

data/README.md

@@ -0,0 +1,11 @@
+#  rsssf-parser - football match schedule & results parser (& tokenizer) for the rsssf format / conventions
+
+
+
+
+
+
+
+
+
+

data/Rakefile

@@ -0,0 +1,29 @@
+require 'hoe'
+
+
+Hoe.spec 'rsssf-parser' do
+
+  self.version = '0.0.1'
+
+  self.summary = "rsssf-parser - football match schedule & results parser (& tokenizer) for the rsssf format / conventions"
+  self.description = summary
+
+  self.urls = { home: 'https://github.com/sportdb/sport.db' }
+
+  self.author = 'Gerald Bauer'
+  self.email = 'gerald.bauer@gmail.com'
+
+  # switch extension to .markdown for gihub formatting
+  self.readme_file  = 'README.md'
+  self.history_file = 'CHANGELOG.md'
+
+  self.licenses = ['Public Domain']
+
+  self.extra_deps = [
+      ['sportdb-parser'],     ### (re(use standard football.txt parser machinery - why? why not?
+  ]
+
+  self.spec_extras = {
+    required_ruby_version: '>= 2.2.2'
+  }
+end

data/bin/rsssf

@@ -0,0 +1,80 @@
+#!/usr/bin/env ruby
+
+## tip: to test run:
+##   ruby -I ./lib -I ../parser/lib  bin/rsssf
+
+require 'rsssf/parser'
+
+
+require 'optparse'
+
+
+args = ARGV
+opts = { debug: false,
+          metal: false }
+
+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
+  end
+
+  parser.on( "--metal",
+                 "turn off typed parse tree; show to the metal tokens"+
+                   " (default: #{opts[:metal]})" ) do |metal|
+    opts[:metal] = metal
+  end
+end
+parser.parse!( args )
+
+puts "OPTS:"
+p opts
+puts "ARGV:"
+p args
+
+
+paths =  if args.empty?
+          [
+            '../../../rsssf/austria/2010-11/cup.txt',
+          ]
+         else
+            ## todo/fix - expand_args!!!
+            args
+         end
+
+
+pp paths
+
+
+
+
+Rsssf::Parser::Linter.debug = true    if opts[:debug]
+
+linter = Rsssf::Parser::Linter.new
+
+
+paths.each_with_index do |path,i|
+    puts "==> [#{i+1}/#{paths.size}] reading >#{path}<..."
+    linter.read( path, parse: !opts[:metal] )
+end
+
+if linter.errors?
+    puts
+    pp linter.errors
+    puts "!!   #{linter.errors.size} parse error(s) in #{paths.size} datafiles(s)"
+else
+    puts "OK   no parse errors found in #{paths.size} datafile(s)"
+end
+
+puts "bye"
+
+
+
+

data/lib/rsssf/parser/linter.rb

@@ -0,0 +1,84 @@
+
+module Rsssf
+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   
+    
+
+
+
+attr_reader :errors
+
+def initialize
+  @errors = []
+  @parser = Parser.new   ## use own parser instance (not shared) - why? why not?
+end
+
+
+def errors?() @errors.size > 0; end
+
+
+
+#########
+## parse - false (default) - tokenize (only)
+##       - true            - tokenize & parse                
+def read( path, parse: false )
+
+  ## fix - (re)use outline reader later!!!
+  ##   plus check for headings etc.
+
+  text = File.open( path, 'r:utf-8' ) { |f| f.read } 
+  lines = text.split( "\n" )
+
+
+  ##  process lines 
+  tree = [] 
+  lines.each do |line|
+
+    ## skip blank and comment lines
+    next if line.strip.empty? || line.strip.start_with?('#')
+
+    ## strip inline (end-of-line) comments
+    line = line.sub( /#.+$/, '' )
+
+
+    if debug?
+      puts
+      puts "line >#{line}<"
+    end
+
+    t, error_messages  =  if parse
+                            @parser.parse_with_errors( line )
+                          else
+                            @parser.tokenize_with_errors( line )                   
+                          end
+         
+
+    if error_messages.size > 0
+      ## add to "global" error list
+      ##   make a triplet tuple (file / msg / line text)
+            error_messages.each do |msg|
+                @errors << [ path,
+                             msg,
+                             line
+                           ]
+            end
+    end
+
+    pp t   if debug?
+
+    tree << t
+  end       
+  ## pp tree
+end  # read
+end  # class Linter
+
+
+end   # class Parser
+end   # module Rsssf

data/lib/rsssf/parser/parser.rb

@@ -0,0 +1,100 @@
+module Rsssf
+class Parser
+    
+
+## transforms
+##
+##  Netherlands  1-2 (1-1)   England
+##   =>  text => team  
+##       score|vs  
+##       text => team
+
+
+
+##
+##
+##  add !!!!
+##   collect_until e.g. collect_until( :text )
+
+
+def parse_with_errors( line, debug: false )
+    errors = []
+    tokens, token_errors = tokenize_with_errors( line )
+    errors += token_errors
+
+
+=begin
+#############
+## pass 1 
+##   replace all texts with keyword matches (e.g. group, round, leg, etc.)
+     tokens = tokens.map do |t|
+                      if t[0] == :text
+                          text = t[1]
+                          if is_group?( text )
+                             ### expects to be followed by num (or text ABC??)
+                             [:group, text]   
+                          elsif is_matchday?( text )
+                             ### expects to be followed by num
+                             ##  use different name e.g. :fix_round or such?
+                             [:matchday, text]   
+                          elsif is_leg?( text )
+                             [:leg, text]
+                          elsif is_round?( text )
+                             [:round, text]
+                          else
+                              t   ## pass through as-is (1:1)
+                          end
+                      else
+                         t
+                      end
+                end
+
+
+    ## puts "tokens:"
+    ## pp tokens
+=end
+
+## transform tokens into (parse tree/ast) nodes    
+    nodes = []
+    
+    ## note - (re)use token buffer from "standard" parser here !!!!
+    buf = SportDb::Parser::Tokens.new( tokens )
+    ## pp buf
+
+
+    loop do 
+          if buf.match?( :text, [:score, 
+                                 :score_awd,
+                                 :score_abd,
+                                 :score_ppd,
+                                 :score_np,
+                                 :score_wo,
+                                 :vs], :text )
+             nodes << [:team, buf.next[1]]
+             nodes << buf.next
+             nodes << [:team, buf.next[1]]
+          elsif buf.match?( :text, :minute )    ## assume player+minute
+             nodes << [:player, buf.next[1]]
+             nodes << buf.next
+          else
+             ## pass through
+             nodes << buf.next
+          end
+
+          break if buf.eos?
+    end
+
+    [nodes,errors]
+end
+
+
+### convience helper - ignore errors by default
+def parse( line, debug: false )
+  nodes, _ = parse_with_errors( line, debug: debug )
+  nodes
+end
+
+
+end #  class Parser
+end  # module Rsssf
+    

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

@@ -0,0 +1,161 @@
+module Rsssf
+class Parser
+   
+
+
+def self.parse_names( txt )
+  lines = [] # array of lines (with words)
+
+  txt.each_line do |line|
+    line = line.strip
+
+    next if line.empty?
+    next if line.start_with?( '#' )   ## skip comments too
+
+    ## strip inline (until end-of-line) comments too
+    ##   e.g. Janvier  Janv  Jan  ## check janv in use??
+    ##   =>   Janvier  Janv  Jan
+
+    line = line.sub( /#.*/, '' ).strip
+    ## pp line
+
+    values = line.split( /[ \t]+/ )
+    ## pp values
+
+    ## todo/fix -- add check for duplicates
+    lines << values
+  end
+  lines
+
+end # method parse
+
+
+def self.build_names( lines )
+  ## join all words together into a single string e.g.
+  ##   January|Jan|February|Feb|March|Mar|April|Apr|May|June|Jun|...
+  lines.map { |line| line.join('|') }.join('|')
+end
+
+
+
+## add normalize option (for downcase) - why? why not?
+def self.build_map( lines )
+    ## note: downcase name!!!
+   ## build a lookup map that maps the word to the index (line no) plus 1 e.g.
+   ##  {"january" => 1,  "jan" => 1,
+   ##   "february" => 2, "feb" => 2,
+   ##   "march" => 3,    "mar" => 3,
+   ##   "april" => 4,    "apr" => 4,
+   ##   "may" => 5,
+   ##   "june" => 6,     "jun" => 6, ...
+   lines.each_with_index.reduce( {} ) do |h,(line,i)|
+     line.each { |name| h[ name.downcase ] = i+1 }  ## note: start mapping with 1 (and NOT zero-based, that is, 0)
+     h
+   end
+end
+
+
+ ## note -  support only 5 letter max for now
+ ##    now January|February|August etc.
+MONTH_LINES = parse_names( <<TXT )
+Jan
+Feb
+March      Mar
+April      Apr
+May
+June       Jun
+July       Jul
+Aug
+Sept       Sep
+Oct
+Nov
+Dec
+TXT
+
+MONTH_NAMES = build_names( MONTH_LINES )
+# pp MONTH_NAMES
+MONTH_MAP   = build_map( MONTH_LINES )
+# pp MONTH_MAP
+
+
+### nnote - only support two or three letters
+##    no Tues | Thur | Thurs | Sunday etc.
+DAY_LINES = parse_names( <<TXT )
+Mon  Mo
+Tue  Tu
+Wed  We
+Thu  Th
+Fri  Fr
+Sat  Sa
+Sun  Su
+TXT
+
+
+DAY_NAMES = build_names( DAY_LINES )
+# pp DAY_NAMES
+DAY_MAP   = build_map( DAY_LINES )
+# pp DAY_MAP
+
+
+
+#=>
+# "Jan|Feb|March|Mar|April|Apr|May|June|Jun|
+#  July|Jul|Aug|Sept|Sep|Oct|Nov|Dec"
+#
+# "Mon|Mo|Tue|Tu|Wed|We|
+#  Thu|Th|Fri|Fr|Sat|Sa|Sun|Su"
+
+
+
+## todo - add more date variants !!!!
+
+# e.g.  Fri Aug 9
+DATE_RE = %r{
+ ## note - do not include [] in capture for now - why? why not
+    ## eat-up/consume optional [] - part i
+    (?: \[ | \b
+     )
+(?<date>
+
+     (?:  ######  
+          ## variant I/1/one
+          ###   Fri June 24 
+
+     ## optional day name
+     ((?<day_name>#{DAY_NAMES})
+          [ ]
+     )?    
+     ##  allow 1 or 2 spaces e.g. Jul  2 / Jun 27 to pretty print
+     (?<month_name>#{MONTH_NAMES})
+         [ ]{1,2}
+     (?<day>\d{1,2})
+     ## optional year
+     (  [ ]
+        (?<year>\d{4})
+     )?   
+     )
+    |
+     (?: #### 
+         ## variant II/2/two
+         ##   17- 3-22   - allow space befor mont
+         ##   17-3-22
+            \d{1,2}
+             -
+            [ ]*\d{1,2} 
+             -
+             (?:
+                \d{4} |   ## 2024
+                \d{2}     ## or 24 only
+             )
+     )
+     )  ## end date capture
+  ## eat-up/consume optional [] - part ii
+  (?: \] | \b
+  )        
+}ix
+
+
+
+end  #   class Parser
+end  # module Rsssf
+   

data/lib/rsssf/parser/token-goals.rb

@@ -0,0 +1,68 @@
+module Rsssf
+class Parser
+
+
+## cannot start with number
+## cannot have number inside
+## cannot end with number!!!
+##
+##  check if can end in dot - why? why not?
+##    e.g.  jr. or such?
+
+##
+##   allow  45+/90+  too
+##     or   90+pen or
+##          90+ pen/90+p/90+ og
+
+
+MINUTE_RE = %r{
+     (?<minute>
+         \b
+            \d{1,3}   
+            '?   ## optional minute quote (')
+            (?:       
+               # optional offset/extra e.g. 45+ / 90+ or 45+10 / 90+5
+                (?: \+
+                  (?: 
+                     (?! [0-9])   ## negative look ahead (not a number) required
+                     |    
+                    (?:
+                      \d{1,3} 
+                      '?   ## optional minute quote (')
+                      (?= (og|pen|p)? ([ ;,\]]|$))
+                    )
+                  )
+                )
+                |
+                (?= (og|pen|p)? ([ ;,\]]|$))  # note - break can be og|pen|p too
+         )
+      )}ix
+### note - word boundary (\b) will NOT work for quoet (')  
+##             because quote is NOT alphanum (like dot etc.)
+
+
+
+##   goal types
+GOAL_PEN_RE = %r{
+   (?<pen> 
+        (?<=\d|\+|[ ]|')	## must follow a number or plus (e.g. 45p / 45+p / 45 p / 45'p) or space
+            (?: pen|p )
+            \b 
+    )
+}ix
+
+
+GOAL_OG_RE = %r{
+   (?<og> 
+        (?<=\d|\+|[ ]|')	## must follow a number or plus (e.g. 45og / 45+og / 45 og) or space
+          og
+          \b
+   )
+}ix
+
+
+
+
+end # class Parser
+end # module Rsssf
+        

data/lib/rsssf/parser/token-note.rb

@@ -0,0 +1,113 @@
+module Rsssf
+class Parser
+    
+
+###
+##  move to token-note(s) file !!!!
+##
+
+NOTE_RE = %r{
+    \[
+   (?<note>
+     (?:  ##  starting with ___   PLUS requiring more text
+       (?:
+          nb:
+          ##  e.g. [NB: between top-8 of regular season]
+          #        [NB: América, Morelia and Tigres qualified on better record regular season]
+          #        [NB: Celaya qualified on away goals]
+          #        [NB: Alebrijes qualified on away goal]
+          #        [NB: Leones Negros qualified on away goals]
+          #
+          # todo/fix:
+          # add "top-level" NB: version
+          ##   with full (end-of) line note - why? why not?
+          |
+          (?: originally[ ])? scheduled
+          ## e.g. [originally scheduled to play in Mexico City] 
+          |
+          rescheduled
+          ## e.g.  [Rescheduled due to earthquake occurred in Mexico on September 19]
+          |
+          remaining
+          ## e.g. [remaining 79']   
+          ##      [remaining 84'] 
+          ##      [remaining 59']   
+          ##      [remaining 5']
+          |
+          played  
+          ## e.g. [played in Macaé-RJ]
+          ##      [played in Caxias do Sul-RS]
+          ##      [played in Sete Lagoas-MG]
+          ##      [played in Uberlândia-MG]
+          ##      [played in Brasília-DF]
+          ##      [played in Vöcklabruck]
+          ##      [played in Pasching]
+          |
+          declared
+          ## e.g.  [declared void]
+          |
+          inter-group
+          ## e.g. [inter-group A-B]
+          ##      [inter-group C-D]
+       )
+      [ ]
+      [^\]]+?    ## slurp all to next ] - (use non-greedy) 
+     )
+      |
+     (?:
+       ## starting with in  - do NOT allow digits
+       ##   name starting with in possible - why? why not?
+           in[ ]
+            [^0-9\]]+?
+       ## e.g. [In Estadio La Corregidora] 
+       ##      [in Unidad Deportiva Centenario]
+       ##      [in Estadio Olímpico Universitario]
+       ##      [in Estadio Victoria]
+       ##      [in UD José Brindis]
+       ##      [in Colomos Alfredo "Pistache" Torres stadium]
+     )
+      |
+      (?:
+          (?:
+             postponed    
+             ## e.g. [postponed due to problems with the screen of the stadium]
+             ##      [postponed by storm]
+             ##      [postponed due to tropical storm "Hanna"]
+             ##      [postponed from Sep 10-12 due to death Queen Elizabeth II]
+             ##     [postponed]  -- include why? why not?
+             |
+             awarded
+             ## e.g. [awarded match to Leones Negros by undue alignment; original result 1-2]
+             ##     [awarded 3-0 to Cafetaleros by undue alignment; originally ended 2-0]
+             ##     [awarded 3-0; originally 0-2, América used ineligible player (Federico Viñas)]
+             |
+             abandoned
+             ## e.g. [abandoned at 1-1 in 65' due to cardiac arrest Luton player Tom Lockyer]
+             ##      [abandoned at 0-0 in 6' due to waterlogged pitch]
+             ##     [abandoned at 5-0 in 80' due to attack on assistant referee by Cerro; result stood]
+             ##    [abandoned at 1-0 in 31']
+             ##    [abandoned at 0-1' in 85 due to crowd trouble]
+             |
+              suspended
+              ## e.g. [suspended at 0-0 in 12' due to storm]  
+              ##      [suspended at 84' by storm; result stood]
+              |
+              annulled
+              ## e.g.  [annulled]
+              |
+              replay
+              ## e.g.  [replay]
+          )
+        ([ ]    ## note - optional text
+          [^\]]+?
+         )?         ## slurp all to next ] - (use non-greedy) 
+      )
+    )    # note capture    
+     \] 
+}ix
+
+
+    
+end  #   class Parser    
+end  #   module Rsssf
+