sportdb-formats 1.1.3 → 1.1.4

data/lib/sportdb/formats/match/match_status_parser.rb

@@ -1,86 +0,0 @@
-#####################
-# helpers for parsing & finding match status e.g.
-#    - cancelled / canceled
-#    - awarded
-#    - abandoned
-#    - replay
-#    etc.
-
-
-module SportDb
-
-  class Status
-# note: use a class as an "enum"-like namespace for now - why? why not?
-#   move class into Match e.g. Match::Status  - why? why not?
-    CANCELLED = 'CANCELLED'   # canceled (US spelling), cancelled (UK spelling) - what to use?
-    AWARDED   = 'AWARDED'
-    POSTPONED = 'POSTPONED'
-    ABANDONED = 'ABANDONED'
-    REPLAY    = 'REPLAY'
-  end # class Status
-
-
-
-  class StatusParser
-
-    def self.parse( str )
-      ## note: returns nil if no match found
-      ## note: english usage - cancelled (in UK), canceled (in US)
-      if str =~ /^(cancelled|
-                   canceled|
-                   can\.
-                  )/xi
-        Status::CANCELLED
-      elsif str =~ /^(awarded|
-                       awd\.
-                      )/xi
-        Status::AWARDED
-      elsif str =~ /^(postponed
-                      )/xi
-        Status::POSTPONED
-      elsif str =~ /^(abandoned|
-                       abd\.
-                      )/xi
-        Status::ABANDONED
-      elsif str =~ /^(replay
-                      )/xi
-        Status::REPLAY
-      else
-        # no match
-        nil
-      end
-    end
-
-
-    RUN_RE = /\[
-                 (?<text>[^\]]+)
-               \]
-             /x
-    def self.find!( line )
-      ## for now check all "protected" text run blocks e.g. []
-      ##  puts "line: >#{line}<"
-
-      status = nil
-
-      str = line
-      while m = str.match( RUN_RE )
-        str = m.post_match  ## keep on processing rest of line/str (a.k.a. post match string)
-
-        ## check for status match
-        match_str = m[0]  ## keep a copy of the match string (for later sub)
-        text = m[:text].strip
-        ## puts "  text: >#{text}<"
-
-        status = parse( text )
-
-        if status
-           line.sub!( match_str, "[STATUS.#{status}]" )
-           break
-        end
-      end  # while match
-
-      status
-    end # method find!
- end # class StatusParser
-
-end # module SportDb

data/lib/sportdb/formats/name_helper.rb

@@ -1,87 +0,0 @@
-
-module SportDb
-  module NameHelper
-
-
-  ## note: allow placeholder years to e.g. (-___) or (-????)
-  ##    for marking missing (to be filled in) years
-  ##  e.g. (1887-1911), (-2013),
-  ##      (1946-2001, 2013-) etc.
-  ##  todo/check: make more strict  e.g. only accept 4-digit years? - why? why not?
-  YEAR_RE =  %r{\(
-                  [0-9, ?_-]+?     # note: non-greedy (minimum/first) match
-              \)}x
-
-  def strip_year( name )
-    ## check for year(s) e.g. (1887-1911), (-2013),
-    ##                        (1946-2001, 2013-) etc.
-    ##  todo/check: only sub once (not global) - why? why not?
-    name.gsub( YEAR_RE, '' ).strip
-  end
-
-  def has_year?( name ) name =~ YEAR_RE; end
-
-
-  LANG_RE =  %r{\[
-                [a-z]{1,2}    # note also allow single-letter [a] or [d] or [e] - why? why not?
-                \]}x
-  def strip_lang( name )
-     name.gsub( LANG_RE, '' ).strip
-  end
-
-  def has_lang?( name ) name =~ LANG_RE; end
-
-
-  def sanitize( name )
-    ## check for year(s) e.g. (1887-1911), (-2013),
-    ##                        (1946-2001,2013-) etc.
-    name = strip_year( name )
-    ## check lang codes e.g. [en], [fr], etc.
-    name = strip_lang( name )
-    name
-  end
-
-
-  ## note: also add (),’,−  etc. e.g.
-  ##   Estudiantes (LP) => Estudiantes LP
-  ##   Saint Patrick’s Athletic FC => Saint Patricks Athletic FC
-  ##   Myllykosken Pallo −47 => Myllykosken Pallo 47
-  ##
-  ##  add & too!!
-  ##   e.g. Brighton & Hove Albion => Brighton Hove Albion  -- and others in England
-
-  NORM_RE =  %r{
-                    [.'’º/()&_−-]
-                  }x   # note: in [] dash (-) if last doesn't need to get escaped
-  ## note: remove all dots (.), dash (-), ', º, /, etc.
-  #   .  U+002E (46) - FULL STOP
-  #   '  U+0027 (39) - APOSTROPHE
-  #   ’  U+2019 (8217) - RIGHT SINGLE QUOTATION MARK
-  #   º  U+00BA (186) - MASCULINE ORDINAL INDICATOR
-  #   /  U+002F (47) - SOLIDUS
-  #   (  U+0028 (40) - LEFT PARENTHESIS
-  #   )  U+0029 (41) - RIGHT PARENTHESIS
-  #   −  U+2212 (8722) - MINUS SIGN
-  #   -  U+002D (45) - HYPHEN-MINUS
-
-  ##         for norm(alizing) names
-  def strip_norm( name )
-    name.gsub( NORM_RE, '' )
-  end
-
-  def normalize( name )
-    # note: do NOT call sanitize here (keep normalize "atomic" for reuse)
-    name = strip_norm( name )
-    name = name.gsub( ' ', '' )  # note: also remove all spaces!!!
-
-    ## todo/check: use our own downcase - why? why not?
-    name = downcase_i18n( name )     ## do NOT care about upper and lowercase for now
-    name
-  end
-
-
-  def variants( name )  Variant.find( name ); end
-
-  end  # module NameHelper
-end   # module SportDb
-

data/lib/sportdb/formats/score/score_formats.rb

@@ -1,239 +0,0 @@
-
-module ScoreFormats
-
-    ## todo/check: use ‹› (unicode chars) to mark optional parts in regex constant name - why? why not?
-
-    #####
-    #  english helpers (penalty, extra time, ...)
-    P_EN  =  '(?: p | pen\.? | pso )'     # e.g. p, pen, pen., PSO, etc.
-    ET_EN =  '(?: aet | a\.e\.t\.? )'     # note: make last . optional (e.g a.e.t) allowed too
-
-
-    ##  note: allow SPECIAL cases WITHOUT full time scores (just a.e.t or pen. + a.e.t.)
-    ##      3-4 pen. 2-2 a.e.t.
-    ##               2-2 a.e.t.
-    EN__P_ET__RE = /\b
-            (?:
-               (?<score1p>\d{1,2})
-                 [ ]* - [ ]*          # note: sep in optional block; CANNOT use a reference
-               (?<score2p>\d{1,2})
-                 [ ]* #{P_EN} [ ]*
-             )?            # note: make penalty (P) score optional for now
-            (?<score1et>\d{1,2})
-               [ ]* - [ ]*
-            (?<score2et>\d{1,2})
-               [ ]* #{ET_EN}
-               (?=[ \]]|$)/xi    ## 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
-    ##      3-4 pen. 2-2 a.e.t. (1-1, )     or
-    ##      3-4 pen. 2-2 a.e.t. (1-1)       or
-    ##               2-2 a.e.t. (1-1, 1-1)  or
-    ##               2-2 a.e.t. (1-1, )     or
-    ##               2-2 a.e.t. (1-1)
-
-    EN__P_ET_FT_HT__RE = /\b
-               (?:
-                (?<score1p>\d{1,2})
-                   [ ]* - [ ]*          # note: sep in optional block; CANNOT use a reference
-                (?<score2p>\d{1,2})
-                   [ ]* #{P_EN} [ ]*
-                )?            # note: make penalty (P) score optional for now
-               (?<score1et>\d{1,2})
-                   [ ]* - [ ]*
-               (?<score2et>\d{1,2})
-                   [ ]* #{ET_EN} [ ]*
-                   \(
-                   [ ]*
-              (?<score1>\d{1,2})
-                   [ ]* - [ ]*
-              (?<score2>\d{1,2})
-                   [ ]*
-                (?:
-                     , [ ]*
-                    (?: (?<score1i>\d{1,2})
-                        [ ]* - [ ]*
-                        (?<score2i>\d{1,2})
-                        [ ]*
-                    )?
-                )?              # note: make half time (HT) score optional for now
-              \)
-             (?=[ \]]|$)/xi    ## todo/check:  remove loakahead assertion here - why require space?
-                               ## note: \b works only after non-alphanum e.g. )
-
-    ###
-    ##   special case for case WITHOUT extra time!!
-    ##     same as above (but WITHOUT extra time and pen required)
-    EN__P_FT_HT__RE = /\b
-     (?<score1p>\d{1,2})
-        [ ]* - [ ]*          # note: sep in optional block; CANNOT use a reference
-     (?<score2p>\d{1,2})
-        [ ]* #{P_EN} [ ]*
-        \(
-        [ ]*
-   (?<score1>\d{1,2})
-        [ ]* - [ ]*
-   (?<score2>\d{1,2})
-        [ ]*
-     (?:
-          , [ ]*
-         (?: (?<score1i>\d{1,2})
-             [ ]* - [ ]*
-             (?<score2i>\d{1,2})
-             [ ]*
-         )?
-     )?              # note: make half time (HT) score optional for now
-   \)
-  (?=[ \]]|$)/xi    ## todo/check:  remove loakahead assertion here - why require space?
-                    ## note: \b works only after non-alphanum e.g. )
-
-
-
-    ## e.g. 2-1 (1-1) or
-    ##      2-1
-    ## note: for now add here used in Brazil / Portugal
-    ##  e.g 1x1 or 1X1 or 0x2 or 3x3  too
-    ##   todo/check/fix: move to its own use PT__FT_HT etc!!!!
-
-    EN__FT_HT__RE = /\b
-              (?<score1>\d{1,2})
-                [ ]* (?<sep>[x-]) [ ]*
-              (?<score2>\d{1,2})
-               (?:
-                   [ ]* \( [ ]*
-                (?<score1i>\d{1,2})
-                   [ ]* \k<sep> [ ]*
-                (?<score2i>\d{1,2})
-                   [ ]* \)
-               )?   # note: make half time (HT) score optional for now
-             (?=[ \]]|$)/xi    ## todo/check:  remove loakahead assertion here - why require space?
-                               ## note: \b works only after non-alphanum e.g. )
-
-
-    #####
-    #  deutsch / german helpers (penalty, extra time, ...)
-    ## todo add more marker e.g. im Elf. or such!!!
-    P_DE  =  '(?: ie | i\.e\.? )'     # e.g. iE, i.E., i.E etc.
-    ET_DE =  '(?: nv | n\.v\.? )'     # e.g. nV, n.V., n.V etc.
-
-
-    ## support alternate all-in-one score e.g.
-    ##     i.E. 2:4, n.V. 3:3 (1:1, 1:1)  or
-    ##               n.V. 3:2 (2:2, 1:2)
-    DE__P_ET_FT_HT__RE = /\b
-                     (?:
-                     #{P_DE}
-                      [ ]*
-                     (?<score1p>\d{1,2})
-                      [ ]* : [ ]*
-                     (?<score2p>\d{1,2})
-                      [ ]* (?:, [ ]*)?
-                     )?   # note: make penalty (P) score optional for now
-                      #{ET_DE}
-                      [ ]*
-                     (?<score1et>\d{1,2})
-                      [ ]* : [ ]*
-                     (?<score2et>\d{1,2})
-                      [ ]*
-                    \(
-                 [ ]*
-             (?<score1>\d{1,2})
-                  [ ]* : [ ]*
-             (?<score2>\d{1,2})
-                  [ ]*
-               (?:
-                   , [ ]*
-                   (?:
-                    (?<score1i>\d{1,2})
-                      [ ]* : [ ]*
-                    (?<score2i>\d{1,2})
-                      [ ]*
-                   )?
-               )?    # note: make half time (HT) score optional for now
-             \)
-            (?=[ \]]|$)
-              /xi
-
-    ## support all-in-one "literal form e.g.
-    #  2:2 (1:1, 1:0) n.V. 5:1 i.E.   or
-    #  2-2 (1-1, 1-0) n.V. 5-1 i.E.
-    DE__ET_FT_HT_P__RE = /\b
-               (?<score1et>\d{1,2})
-                   [ ]* (?<sep>[:-]) [ ]*     ## note: for now allow : or - as separator!!
-               (?<score2et>\d{1,2})
-                   [ ]*
-                   \(
-                  [ ]*
-              (?<score1>\d{1,2})
-                   [ ]* \k<sep> [ ]*
-              (?<score2>\d{1,2})
-                   [ ]*
-                (?:
-                    , [ ]*
-                    (?:
-                     (?<score1i>\d{1,2})
-                       [ ]* \k<sep> [ ]*
-                     (?<score2i>\d{1,2})
-                       [ ]*
-                    )?
-                )?    # note: make half time (HT) score optional for now
-              \)
-               [ ]*
-               #{ET_DE}
-              (?:
-                [ ]*
-                (?<score1p>\d{1,2})
-                 [ ]* \k<sep> [ ]*
-                (?<score2p>\d{1,2})
-                  [ ]*
-                #{P_DE}
-              )?       # note: make penalty (P) score optional for now
-             (?=[ \]]|$)
-               /xi    ## todo/check:  remove loakahead assertion here - why require space?
-                               ## note: \b works only after non-alphanum e.g. )
-
-
-    ## e.g. 2:1 (1:1)  or
-    ##      2-1 (1-1)  or
-    ##      2:1        or
-    ##      2-1
-    DE__FT_HT__RE = /\b
-              (?<score1>\d{1,2})
-                [ ]* (?<sep>[:-]) [ ]*
-              (?<score2>\d{1,2})
-               (?:
-                  [ ]* \( [ ]*
-                      (?<score1i>\d{1,2})
-                        [ ]* \k<sep> [ ]*
-                      (?<score2i>\d{1,2})
-                  [ ]* \)
-               )?   # note: make half time (HT) score optional for now
-             (?=[ \]]|$)/x    ## todo/check:  remove loakahead assertion here - why require space?
-                               ## note: \b works only after non-alphanum e.g. )
-
-
-#############################################
-# map tables - 1) regex,  2) tag - note: order matters; first come-first matched/served
-
-
-FORMATS_EN = [
-  [ EN__P_ET_FT_HT__RE, '[SCORE.EN__P?_ET_(FT_HT?)]' ], # e.g. 5-1 pen. 2-2 a.e.t. (1-1, 1-0)
-  [ EN__P_FT_HT__RE,    '[SCORE.EN__P_(FT_HT?)]'     ], # e.g. 5-1 pen. (1-1)
-  [ EN__P_ET__RE,       '[SCORE.EN__P?_ET]'          ], # e.g. 2-2 a.e.t.  or  5-1 pen. 2-2 a.e.t.
-  [ EN__FT_HT__RE,      '[SCORE.EN__FT_(HT)?]'       ], # e.g. 1-1 (1-0)
-]
-
-FORMATS_DE = [
-  [ DE__ET_FT_HT_P__RE, '[SCORE.DE__ET_(FT_HT?)_P?]' ], # e.g. 2:2 (1:1, 1:0) n.V. 5:1 i.E.
-  [ DE__P_ET_FT_HT__RE, '[SCORE.DE__P?_ET_(FT_HT?)]' ], # e.g. i.E. 2:4, n.V. 3:3 (1:1, 1:1)
-  [ DE__FT_HT__RE,      '[SCORE.DE__FT_(HT)?]'       ], # e.g. 1:1 (1:0)
-]
-
-FORMATS = {
-  en: FORMATS_EN,
-  de: FORMATS_DE,
-}
-
-end # module ScoreFormats

data/lib/sportdb/formats/score/score_parser.rb

@@ -1,204 +0,0 @@
-# encoding: utf-8
-
-
-## note: lets follow the model of DateFormats -see DateFormats gem for more!!!
-
-
-## note: make Score top-level and use like Date - why? why not?
-class Score
-
-  attr_reader :score1i,  :score2i,   # half time (ht) score
-              :score1,   :score2,    # full time (ft) score
-              :score1et, :score2et,  # extra time (et) score
-              :score1p,  :score2p    # penalty (p) score
-              ## todo/fix: add :score1agg, score2agg too - why? why not?!!!
-              ##  add state too e.g. canceled or abadoned etc - why? why not?
-
-  def initialize( *values )
-    ## note: for now always assumes integers
-    ##  todo/check - check/require integer args - why? why not?
-
-    @score1i  = values[0]    # half time (ht) score
-    @score2i  = values[1]
-
-    @score1   = values[2]    # full time (ft) score
-    @score2   = values[3]
-
-    @score1et = values[4]    # extra time (et) score
-    @score2et = values[5]
-
-    @score1p  = values[6]    # penalty (p) score
-    @score2p  = values[7]
-  end
-
-  def to_a
-    ## todo: how to handle game w/o extra time
-    #   but w/ optional penalty ???  e.g. used in copa liberatores, for example
-    #    retrun 0,0 or nil,nil for extra time score ?? or -1, -1 ??
-    #    for now use nil,nil
-    score = []
-    score += [score1i,  score2i]     if score1p || score2p || score1et || score2et || score1 || score2 || score1i || score2i
-    score += [score1,   score2]      if score1p || score2p || score1et || score2et || score1 || score2
-    score += [score1et, score2et]    if score1p || score2p || score1et || score2et
-    score += [score1p,  score2p]     if score1p || score2p
-    score
-  end
-
-end  # class Score
-
-
-
-module ScoreFormats
-
-  def self.lang
-    @@lang ||= :en            ## defaults to english (:en)
-  end
-  def self.lang=( value )
-    @@lang = value.to_sym    ## note: make sure lang is always a symbol for now (NOT a string)
-    @@lang      ## todo/check: remove  =() method always returns passed in value? double check
-  end
-
-
-  def self.parser( lang: )  ## find parser
-    lang = lang.to_sym  ## note: make sure lang is always a symbol for now (NOT a string)
-
-    ## note: cache all "built-in" lang versions (e.g. formats == nil)
-    @@parser ||= {}
-    parser = @@parser[ lang ] ||= ScoreParser.new( lang: lang )
-  end
-
-  def self.parse( line, lang: ScoreFormats.lang )
-    parser( lang: lang ).parse( line )
-  end
-
-  def self.find!( line, lang: ScoreFormats.lang )
-    parser( lang: lang ).find!( line )
-  end
-
-
-class ScoreParser
-
-  include LogUtils::Logging
-
-  def initialize( lang: )
-    @lang    = lang.to_sym   ## note: make sure lang is always a symbol for now (NOT a string)
-
-    ## fallback to english if lang not available
-    ##  todo/fix: add/issue warning - why? why not?
-    @formats = FORMATS[ @lang ] || FORMATS[ :en ]
-  end
-
-
-  def parse( line )
-
-     ##########
-     ## todo/fix/check: add unicode to regular dash conversion - why? why not?
-     ##  e.g. – becomes -  (yes, the letters a different!!!)
-     #############
-
-    score = nil
-    @formats.each do |format|
-      re = format[0]
-      m = re.match( line )
-      if m
-        score = parse_matchdata( m )
-        break
-      end
-      # no match; continue; try next regex pattern
-    end
-
-    ## todo/fix - raise ArgumentError - invalid score; no format match found
-    score  # note: nil if no match found
-  end # method parse
-
-
-  def find!( line )
-    ### fix: add and match all-in-one literal first, followed by
-
-    # note: always call after find_dates !!!
-    #  scores match date-like patterns!!  e.g. 10-11  or 10:00 etc.
-    #   -- note: score might have two digits too
-
-    ### fix: depending on language allow 1:1 or 1-1
-    ##   do NOT allow mix and match
-    ##  e.g. default to en is  1-1
-    ##    de is 1:1 etc.
-
-
-    # extract score from line
-    # and return it
-    # note: side effect - removes date from line string
-
-    score = nil
-    @formats.each do |format|
-      re  = format[0]
-      tag = format[1]
-      m = re.match( line )
-      if m
-        score = parse_matchdata( m )
-        line.sub!( m[0], tag )
-        break
-      end
-      # no match; continue; try next regex pattern
-    end
-
-    score  # note: nil if no match found
-  end # method find!
-
-private
-  def parse_matchdata( m )
-    # convert regex match_data captures to hash
-    # - note: cannont use match_data like a hash (e.g. raises exception if key/name not present/found)
-    h = {}
-    # - note: do NOT forget to turn name into symbol for lookup in new hash (name.to_sym)
-    m.names.each { |name| h[name.to_sym] = m[name] }  # or use match_data.names.zip( match_data.captures )  - more cryptic but "elegant"??
-
-    ## puts "[parse_date_time] match_data:"
-    ## pp h
-    logger.debug "   [parse_matchdata] hash: >#{h.inspect}<"
-
-    score1i   = nil    # half time (ht) scores
-    score2i   = nil
-
-    score1    = nil    # full time (ft) scores
-    score2    = nil
-
-    score1et  = nil    # extra time (et) scores
-    score2et  = nil
-
-    score1p   = nil   # penalty (p) scores
-    score2p   = nil
-
-
-    if h[:score1i] && h[:score2i]   ## note: half time (HT) score is optional now
-      score1i   = h[:score1i].to_i
-      score2i   = h[:score2i].to_i
-    end
-
-    if h[:score1] && h[:score2]     ## note: full time (FT) score can be optional too!!!
-      score1 = h[:score1].to_i
-      score2 = h[:score2].to_i
-    end
-
-    if h[:score1et] && h[:score2et]
-      score1et = h[:score1et].to_i
-      score2et = h[:score2et].to_i
-    end
-
-    if h[:score1p] && h[:score2p]
-      score1p   = h[:score1p].to_i
-      score2p   = h[:score2p].to_i
-    end
-
-    score = Score.new( score1i,  score2i,
-                       score1,   score2,
-                       score1et, score2et,
-                       score1p,  score2p   )
-    score
-  end  # method parse_matchdata
-
-
-
-end  # class ScoreParser
-end  # module ScoreFormats
-