sportdb-parser 0.2.0 → 0.2.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: c94dcd42fc13a7043f6b926ca1d947df3199877693b22e53e4f50b5aa522bf5d
-  data.tar.gz: 33eb689dcfb2bab0728c19b7d706da1556ddefafbfbcc6e424ac5bcbe3bccef6
+  metadata.gz: 0c9225b21f400b9f9cced2052c3062f41a091ed81d3d4239164c9652f53ebc6e
+  data.tar.gz: f7250eaa21324962df27e7cdd397857afa570c610f00c80c31e5105e40964002
 SHA512:
-  metadata.gz: 97ef8d76ffa26312d66359f364588af3d7c76a3b0cebd3644b1f1ae775463defa9cb9552b267f26677c2c6f4e9b7b9fe62479dd34a7211fd1a4a3c1b5e9af830
-  data.tar.gz: ca9b56c6c02c132f3924fb40c293e90379812b830a2899e2be02c1d6469a278456c6d68db7f73d5f5fd69b372c958953e3fefd829ac1120cf56b0944176a2b87
+  metadata.gz: 471c938c233d8f81d7a0fd5e4470a27a52486906764816b6c35ea3d88e19650c81302fd5ff9ee30b85d3a8e9f81ada8eef20b49bd3de924c7238acb106ba6082
+  data.tar.gz: 24d1cf3846404859ad7e751895325b256321d43e2881413fda6325c744ca0c31b52ef2032a9dfc8e56e67d7a06df54a6d2780a297982440b8e40b7055fe06c26

data/CHANGELOG.md

@@ -1,4 +1,4 @@
-### 0.2.0
+### 0.2.2
 
 ### 0.0.1 / 2024-07-12
 

data/Manifest.txt

@@ -2,12 +2,8 @@ CHANGELOG.md
 Manifest.txt
 README.md
 Rakefile
-bin/fbt
 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
 lib/sportdb/parser/token-score.rb

data/Rakefile

@@ -26,6 +26,6 @@ Hoe.spec 'sportdb-parser' do
   ]
 
   self.spec_extras = {
-    required_ruby_version: '>= 2.2.2'
+    required_ruby_version: '>= 3.1.0'
   }
 end

data/lib/sportdb/parser/lang.rb

@@ -27,6 +27,12 @@ end
 
 ROUND_RE = %r{^(
 
+   ## add special case for group play-off rounds!
+   ##  group 2 play-off   (e.g. worldcup 1954, 1958)
+     (?:   Group [ ] [a-z0-9]+ [ ]
+           Play-?offs?
+     )
+        |
    # round  - note - requiers number e.g. round 1,2, etc.
    #   note - use 1-9 regex (cannot start with 0) - why? why not?
    #             make week 01 or round 01 or matchday 01 possible?
@@ -46,17 +52,23 @@ ROUND_RE = %r{^(
        |
    ## 1. Round / 2. Round / 3. Round / etc.
    ##  Play-off Round
+   ##  First Round
+   ##  Final Round   (e.g. Worldcup 1950)
       (?:
-           (?: [1-9][0-9]* \.
-                |
-                Play-?off
+           (?: [1-9][0-9]* \.  |
+                Play-?off   |
+                1st | First   |
+                2nd | Second  |
+                Final
            )
              [ ] Round
        )
        |
   ## starting with preliminary
+  #   e.g.  Preliminary round
      (?:  Preliminary  [ ]
-           (?:  Semi-?finals |
+           (?:  Round |
+                Semi-?finals |
                 Final
            )
      )
@@ -110,10 +122,15 @@ ROUND_RE = %r{^(
          Finals?
          |
     ## add replays
-    ##  Final Replay
+    ##  e.g. Final Replay
+    ##       Quarter-finals replays
+    ##       First round replays
      (?:
-        Final
-        [ ] Replay
+        (?: First [ ] Round |
+            Quarter-?finals? |
+            Finals?
+         )
+        [ ] Replays?
       )
 )$}ix
 

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

@@ -1,6 +1,6 @@
-module SportDb 
+module SportDb
 class Parser
-   
+
 
 
 def self.parse_names( txt )
@@ -47,8 +47,8 @@ def self.build_map( lines, downcase: false )
   ##   "may" => 5,
   ##   "june" => 6,     "jun" => 6, ...
   lines.each_with_index.reduce( {} ) do |h,(line,i)|
-    line.each do |name| 
-       h[ downcase ? name.downcase : name ] = i+1 
+    line.each do |name|
+       h[ downcase ? name.downcase : name ] = i+1
     end  ## note: start mapping with 1 (and NOT zero-based, that is, 0)
     h
   end
@@ -109,28 +109,85 @@ DAY_MAP   = build_map( DAY_LINES, downcase: true )
 ## todo - add more date variants !!!!
 
 # e.g. Fri Aug/9  or Fri Aug 9
-DATE_RE = %r{
+DATE_I_RE = %r{
 (?<date>
   \b
      ## optional day name
      ((?<day_name>#{DAY_NAMES})
           [ ]
-     )?    
+     )?
      (?<month_name>#{MONTH_NAMES})
          (?: \/|[ ] )
      (?<day>\d{1,2})
      ## optional year
      (  [ ]
         (?<year>\d{4})
-     )?   
-  \b   
+     )?
+  \b
 )}ix
 
 
+# e.g. 3 June  or 10 June
+DATE_II_RE = %r{
+(?<date>
+  \b
+     ## optional day name
+     ((?<day_name>#{DAY_NAMES})
+          [ ]
+     )?
+     (?<day>\d{1,2})
+         [ ]
+     (?<month_name>#{MONTH_NAMES})
+     ## optional year
+     (  [ ]
+        (?<year>\d{4})
+     )?
+  \b
+)}ix
+
+
+#############################################
+# map tables
+#  note: order matters; first come-first matched/served
+DATE_RE = Regexp.union(
+   DATE_I_RE,
+   DATE_II_RE
+)
+
+
+##
+##  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  
+#  date duration
 #   use - or + as separator
-#    in theory plus( +) only if dates 
+#    in theory plus( +) only if dates
 #     are two days next to each other
 #
 #   otherwise  define new dates type in the future? why? why not?
@@ -147,7 +204,7 @@ DATE_RE = %r{
 #  Jun/25 .. 26        - why? why not???
 #  Jun/25 to 26        - why? why not???
 #  Jun/25 + 26        - add - why? why not???
-#  Sun-Wed Jun/23-26  -  add - why? why not??? 
+#  Sun-Wed Jun/23-26  -  add - why? why not???
 #  Wed+Thu Jun/26+27 2024  -  add - why? why not???
 #
 #  maybe use comman and plus for list of dates
@@ -157,39 +214,89 @@ DATE_RE = %r{
 #   add back optional comma (before) year - why? why not?
 
 
-DURATION_RE =  %r{
+##
+#   todo add plus later on - why? why not?
+
+DURATION_I_RE =  %r{
 (?<duration>
     \b
    ## optional day name
    ((?<day_name1>#{DAY_NAMES})
       [ ]
-   )?    
+   )?
    (?<month_name1>#{MONTH_NAMES})
       (?: \/|[ ] )
    (?<day1>\d{1,2})
    ## optional year
    ( [ ]
       (?<year1>\d{4})
-   )?   
+   )?
 
    ## support + and -  (add .. or such - why??)
-   [ ]*[+-][ ]*   
-  
+   [ ]*[-][ ]*
+
    ## optional day name
    ((?<day_name2>#{DAY_NAMES})
       [ ]
-   )?    
+   )?
    (?<month_name2>#{MONTH_NAMES})
       (?: \/|[ ] )
    (?<day2>\d{1,2})
    ## optional year
    ( [ ]
       (?<year2>\d{4})
-   )?   
-   \b   
+   )?
+   \b
+)}ix
+
+
+###
+#   variant ii
+# e.g. 26 July - 27 July
+
+DURATION_II_RE =  %r{
+(?<duration>
+    \b
+   ## optional day name
+   ((?<day_name1>#{DAY_NAMES})
+      [ ]
+   )?
+   (?<day1>\d{1,2})
+      [ ]
+   (?<month_name1>#{MONTH_NAMES})
+   ## optional year
+   ( [ ]
+      (?<year1>\d{4})
+   )?
+
+   ## support + and -  (add .. or such - why??)
+   [ ]*[-][ ]*
+
+   ## optional day name
+   ((?<day_name2>#{DAY_NAMES})
+      [ ]
+   )?
+   (?<day2>\d{1,2})
+      [ ]
+   (?<month_name2>#{MONTH_NAMES})
+   ## optional year
+   ( [ ]
+      (?<year2>\d{4})
+   )?
+   \b
 )}ix
 
 
+#############################################
+# map tables
+#  note: order matters; first come-first matched/served
+DURATION_RE = Regexp.union(
+   DURATION_I_RE,
+   DURATION_II_RE
+)
+
+
+
 end  #   class Parser
-end  # module SportDb 
-   
+end  # module SportDb
+

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 = 2
-  PATCH = 0
+  PATCH = 2
   VERSION = [MAJOR,MINOR,PATCH].join('.')
 
   def self.version

data/lib/sportdb/parser.rb

@@ -24,11 +24,6 @@ require_relative 'parser/lang'
 require_relative 'parser/parser'
 
 
-## more
-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: 0.2.0
+  version: 0.2.2
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-08-22 00:00:00.000000000 Z
+date: 2024-08-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: cocos
@@ -74,8 +74,7 @@ dependencies:
         version: '4.1'
 description: sportdb-parser - football.txt match parser (& tokenizer)
 email: gerald.bauer@gmail.com
-executables:
-- fbt
+executables: []
 extensions: []
 extra_rdoc_files:
 - CHANGELOG.md
@@ -86,12 +85,8 @@ files:
 - Manifest.txt
 - README.md
 - Rakefile
-- bin/fbt
 - 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
 - lib/sportdb/parser/token-score.rb
@@ -112,7 +107,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 2.2.2
+      version: 3.1.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="

data/bin/fbt

@@ -1,94 +0,0 @@
-#!/usr/bin/env ruby
-
-## tip: to test run:
-##   ruby -I ./lib bin/fbt
-
-## our own code
-require 'sportdb/parser'
-
-
-
-require 'optparse'
-
-##
-## read textfile
-##   and dump tokens
-##
-##   fbt  ../openfootball/.../euro.txt
-
-
-
-
- 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?
-            [
-              '../../../openfootball/euro/2020--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.each_with_index do |path,i|
-    puts "==> [#{i+1}/#{paths.size}] reading >#{path}<..."
-    linter.read( path, parse: !opts[:metal] )
-
-    errors += linter.errors    if linter.errors?
-end
-
-if errors.size > 0
-    puts
-    pp errors
-    puts
-    puts "!!   #{errors.size} parse error(s) in #{paths.size} datafiles(s)"
-else
-    puts
-    puts "OK   no parse errors found in #{paths.size} datafile(s)"
-end
-
-puts "bye"
-

data/lib/sportdb/parser/linter.rb

@@ -1,149 +0,0 @@
-
-module SportDb
-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
-
-
-
-  ## note:  colon (:) MUST be followed by one (or more) spaces
-  ##      make sure mon feb 12 18:10 will not match
-  ##        allow 1. FC Köln etc.
-  ##               Mainz 05:
-  ##           limit to 30 chars max
-  ##          only allow  chars incl. intl buut (NOT ()[]/;)
-  ##
-  ##   Group A:
-  ##   Group B:   - remove colon
-  ##    or lookup first
-
-  ATTRIB_RE = %r{^  
-                   [ ]*?     # slurp leading spaces
-                (?<key>[^:|\]\[()\/; -]
-                       [^:|\]\[()\/;]{0,30}
-                 )
-                   [ ]*?     # slurp trailing spaces
-                   :[ ]+
-                (?<value>.+)
-                    [ ]*?   # slurp trailing spaces
-                   $
-                }ix
-
-
-#########
-## parse - false (default) - tokenize (only)
-##       - true            - tokenize & parse                
-def read( path, parse: false )
-  ## note: every (new) read call - resets errors list to empty
-  @errors = []
-
-  nodes = OutlineReader.read( path ) 
-
-  ##  process nodes
-  h1 = nil
-  orphans = 0    ## track paragraphs's with no heading
-
-  attrib_found = false
-
-
-  nodes.each do |node|
-    type = node[0]
-    
-    if type == :h1
-        h1 = node[1]  ## get heading text
-        puts
-        puts "  = Heading 1 >#{node[1]}<"
-    elsif type == :p
-
-       if h1.nil?
-         orphans += 1    ## only warn once 
-         puts "!! WARN - no heading for #{orphans} text paragraph(s); skipping parse"
-         next
-       end
-
-       lines = node[1]
-
-       tree = [] 
-       lines.each_with_index do |line,i|
-
-        if debug?
-         puts
-         puts "line >#{line}<"
-        end
-
-
-        ## skip new (experimental attrib syntax)
-        if attrib_found == false && 
-            ATTRIB_RE.match?( line )
-          ## note: check attrib regex AFTER group def e.g.:
-          ##         Group A: 
-          ##         Group B:  etc.
-          ##     todo/fix - change Group A: to Group A etc.
-          ##                       Group B: to Group B
-           attrib_found = true
-           ## logger.debug "skipping key/value line - >#{line}<"
-           next
-        end
-
-        if attrib_found
-          ## 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?( '.' ) 
-              # logger.debug "skipping key/value line (cont.) - >#{line}<"
-              next
-        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
-    else
-        pp node
-        raise ArgumentError, "unsupported (node) type >#{type}<"
-    end
-  end  # each node
-end  # read
-end  # class Linter
-
-
-end   # class Parser
-end   # module SportDb

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

data/lib/sportdb/parser/outline_reader.rb

@@ -1,97 +0,0 @@
-
-
-module SportDb
-
-class OutlineReader
-
-  def self.debug=(value) @@debug = value; end
-  def self.debug?() @@debug ||= false; end
-  def debug?()  self.class.debug?; end
-
-
-
-  def self.read( path )   ## use - rename to read_file or from_file etc. - why? why not?
-    txt = File.open( path, 'r:utf-8' ) {|f| f.read }
-    parse( txt )
-  end
-
-  def self.parse( txt )
-    new( txt ).parse
-  end
-
-  def initialize( txt )
-    @txt = txt
-  end
-
-  ## note: skip "decorative" only heading e.g. ========
-  ##  todo/check:  find a better name e.g. HEADING_EMPTY_RE or HEADING_LINE_RE or ???
-  HEADING_BLANK_RE = %r{\A
-                        ={1,}
-                        \z}x
-
-  ## note: like in wikimedia markup (and markdown) all optional trailing ==== too
-  HEADING_RE = %r{\A
-                  (?<marker>={1,})       ## 1. leading ======
-                    [ ]*
-                  (?<text>[^=]+)         ## 2. text   (note: for now no "inline" = allowed)
-                    [ ]*
-                    =*                   ## 3. (optional) trailing ====
-                  \z}x
-
-  def parse
-    outline=[]   ## outline structure
-    start_para = true      ## start new para(graph) on new text line?
-
-    @txt.each_line do |line|
-        line = line.strip      ## todo/fix: keep leading and trailing spaces - why? why not?
-
-        if line.empty?    ## todo/fix: keep blank line nodes?? and just remove comments and process headings?! - why? why not?
-          start_para = true
-          next
-        end
-
-        break if line == '__END__'
-
-        next if line.start_with?( '#' )   ## skip comments too
-        ## strip inline (until end-of-line) comments too
-        ##  e.g Eupen | KAS Eupen ## [de]
-        ##   => Eupen | KAS Eupen
-        ##  e.g bq   Bonaire,  BOE        # CONCACAF
-        ##   => bq   Bonaire,  BOE
-        line = line.sub( /#.*/, '' ).strip
-        pp line    if debug?
-
-        ## todo/check: also use heading blank as paragraph "breaker" or treat it like a comment ?? - why? why not?
-        next if HEADING_BLANK_RE.match( line )  # skip "decorative" only heading e.g. ========
-
-         ## note: like in wikimedia markup (and markdown) all optional trailing ==== too
-        if m=HEADING_RE.match( line )
-           start_para = true
-
-           heading_marker = m[:marker]
-           heading_level  = heading_marker.length   ## count number of = for heading level
-           heading        = m[:text].strip
-
-           puts "heading #{heading_level} >#{heading}<"   if debug?
-           outline << [:"h#{heading_level}", heading]
-        else    ## assume it's a (plain/regular) text line
-           if start_para
-             outline << [:p, [line]]
-             start_para = false
-           else
-             node = outline[-1]    ## get last entry
-             if node[0] == :p      ##  assert it's a p(aragraph) node!!!
-                node[1] << line    ## add line to p(aragraph)
-             else
-               puts "!! ERROR - invalid outline state / format - expected p(aragraph) node; got:"
-               pp node
-               exit 1
-             end
-           end
-        end
-    end
-    outline
-  end # method read
-end # class OutlineReader
-
-end # module SportDb