rsssf 0.2.0 → 0.3.0

data/lib/rsssf/fmtfix/outline.rb

@@ -0,0 +1,228 @@
+module Rsssf
+class  Fmtfix    ## todo: find a better name e.g. Format or Fixer or ??
+
+
+     ##
+     ## note - ascii hr replacement is
+     ##            =-=-= (do NOT match) !!!!!!
+
+
+HX_RE = %r{          ## negative lookahead
+                     ##   do NOT match  =-=
+                     ##   do NOT match  ===========  (without any heading text!!)
+                     ##     e.g.
+                     ##       Fall season
+                     ##       ===========
+
+                    (?! ^[ ]* (?:    =-=
+                                 |  ={1,} [ ]* $
+                               )
+                     )
+
+                     ^
+                    [ ]*
+
+                  (?<marker> ={1,6})
+                     [ ]*
+                  (?<text> .+?)
+                     #{OPT_REF}
+                     [ ]*
+            $}x
+
+
+
+
+
+def autofix_outline( txt, title: )
+
+
+     hx =  txt.scan( HX_RE )
+
+     ### note - shortcircuit if no headings found!!!
+     return txt    if hx.size == 0
+
+
+
+
+     ## update counts/usage of h1,h2,h3,h4,h5,h6
+     counts = [nil,0,0,0,0,0,0]
+     hx.each do |marker,_|
+                   level = marker.size;
+                   counts[ level ] += 1
+     end
+
+      ## flatten level; only record levels with heading counts
+      levels = []
+      counts.each_with_index do  |count,level|
+            levels << level    if count && count > 0
+      end
+
+
+
+      #####
+      ### special case for first heading
+      ##    check if heading is matching title AND the only one in top level
+      htop_marker, htop_header = hx[0]
+      htop_level = htop_marker.size
+
+      ##  top heading MUST always be lowest (top)
+      if htop_level == levels[0]
+        if counts[htop_level] == 1
+          ##  check if same as title
+          ##    if yes pop (that is, remove too)
+          if htop_header == title
+             counts[htop_level] = 0  ## update/reset counter
+             levels.shift            ## remove first level (inline op)!!!
+
+             ### note - space in header must be replaces with [ ]!!!!
+             ##                    or \\   with Regex.escape!!!
+             ###  note - MUST escape string for regex e.g. [Bra..] or 1.
+             ###
+             ###   V COPA BRASIL - 1979 [Brazilian Championship]
+             ##    check if space works with escape??
+
+             htop_re = %r{
+                            ^
+                           [ ]* #{htop_marker}
+                           [ ]* #{Regexp.escape(htop_header)}
+                              .*?
+                           $    ## or use \n - why? why not?
+                         }x
+             ## remove line in txt too
+             txt = txt.sub( htop_re ) do |match|
+                   puts "   removing top heading matching title  -- >#{match}<"
+                                 ''
+                              end
+          else
+             ## warn/log  - heading top NOT matching page title
+             msg = "first top heading NOT matching page title  #{htop_header} <=> #{title}"
+             puts "!! WARN #{msg}"
+             log( msg )
+          end
+        else
+          ## warn/log   - more than one top level heading!!!
+          msg = "more than one (#{counts[htop_level]}) top heading #{htop_level} found " +
+                "in page with title #{title}"
+          ## maybe add headers in the future - why? why not?
+          log( msg )
+       end
+      else
+         ## warn/log   - top heading NOT top!!
+          msg = "top heading #{htop_level} not top (#{levels[0]}) " +
+                "in page with title #{title}"
+          log( msg )
+      end
+
+
+       mapping = {}
+       levels.each_with_index do |level,i|
+            from = level
+            to   = i+1
+            mapping[from] = to
+       end
+
+
+      # rewrite headings
+       txt = txt.gsub( HX_RE ) do
+                 m = Regexp.last_match
+                 old_marker = m[:marker]
+                 old_level  = m[:marker].size
+
+                 new_level = mapping[old_level]
+
+                 if new_level.nil?
+                    puts "!! no heading #{old_level} mapping found in page >#{title}<:"
+                    puts "match:"
+                    pp m
+                    puts "counts:"
+                    pp counts
+                    puts "levels:"
+                    pp levels
+                    puts "mapping:"
+                    pp mapping
+                    exit 1
+                 end
+
+                 new_marker =  '=' * new_level
+
+                 ## remove level diff from marker
+                 ##
+                 ##  maybe in the future use track trailing marker too
+                 ##   and rebuild heading/header instead of gsub
+
+                ## note -  always start at level 2 (page title like in wikipedia is level 1)
+                 ##                  thus, new_level+1
+
+                 if (old_level - new_level+1) > 0
+                    ## note - will remove diff from leading (and possibly trailing) marker too
+                    m[0].gsub( old_marker, new_marker+'=' )
+                 else
+                    m[0]
+                 end
+           end
+
+
+    txt
+
+end
+
+
+
+def _scan_outline( txt )   txt.scan( HX_RE );  end
+
+def build_outline( txt )
+
+     hx =  txt.scan( HX_RE )
+
+
+     counts = [nil,0,0,0,0,0,0]  ## note - index 0 is nil
+                                 ##  index 1 (h1) is 0 etc.
+
+     hx.each { |marker,_| counts[ marker.size ] +=1 }
+
+
+     buf = String.new
+     buf += "  outline:"
+     buf += " " +
+          "#{counts[1]==0 ? '-' : 'h1'}/" +
+          "#{counts[2]==0 ? '-' : 'h2'}/" +
+          "#{counts[3]==0 ? '-' : 'h3'}/" +
+          "#{counts[4]==0 ? '-' : 'h4'}/" +
+          "#{counts[5]==0 ? '-' : 'h5'}/" +
+          "#{counts[6]==0 ? '-' : 'h6'}" +
+          "\n"
+
+         buf += "           " +
+              "#{counts[1]==0 ? '-' : counts[1]}/" +
+               "#{counts[2]==0 ? '-' : counts[2]}/" +
+               "#{counts[3]==0 ? '-' : counts[3]}/" +
+               "#{counts[4]==0 ? '-' : counts[4]}/" +
+               "#{counts[5]==0 ? '-' : counts[5]}/" +
+               "#{counts[6]==0 ? '-' : counts[6]}" +
+               "\n"
+
+     hx.each do |marker,text|
+        buf << "    (%d) %-6s" % [marker.size, marker]
+        buf <<  "  "
+        buf << text
+        buf << "\n"
+     end
+
+
+     ## count anchors (aka a name)
+     ##  e.g
+       aname = txt.scan( /‹§  [^›]+  ›/x )
+
+        if aname.size > 0
+          buf << "\n"
+          buf << "  aname #{aname.size}: "
+          buf <<  aname.join( ',' )
+          buf << "\n"
+        end
+
+        buf
+end
+
+
+end    ## class Fmtfix
+end    ## module Rsssf

data/lib/rsssf/fmtfix/patch_headings.rb

@@ -0,0 +1,141 @@
+module Rsssf
+class  Fmtfix    ## todo: find a better name e.g. Format or Fixer or ??
+
+## e.g. 2008/09
+##   note: also support 1999/2000
+## note: use single quotes - quotes do NOT get escaped (e.g. '\d' => "\\d")
+##
+## SEASON  =   \d{4}/(?:\d{2}|\d{4})
+##  built-in for now
+
+
+def self.mkheading_regex( str )
+
+    str = str.strip
+    ##
+    ## change all spaces (other than [ ] and $$ ) to interpunkt
+    str = str.gsub( %r{
+                             (?<charclass> [ ]* \[ [^\[\]]+ \] [*?+]? [ ]*)
+                          |  (?<newline>   [ ]* \$\$ [ ]*)
+                          |  (?<spaces>    [ ]+)
+                        }x
+                       ) do
+              m = Regexp.last_match
+              if m[:spaces]
+                 ' [ ] '     ##  change space to [ ]
+              elsif m[:newline]
+                 ' \s+ '     ##  $$ => \s+  -- note - make sure \s incl. newline!!
+              else
+                 m[0]         ## keep as is
+              end
+            end
+
+  ##  escpape  .   to \.
+  ##  change  ~    to [ ]?  -- that is, optional space
+  ##  change  ( )  to \( \)
+   str = str.gsub(  '~', ' [ ]? ' )
+   str = str.gsub(  '.', '\.' )
+   str = str.gsub(  '(', '\(' )
+   str = str.gsub(  ')', '\)' )
+
+
+  ### last step change builtins
+  ##     '$SEASON$' => '\d{4}/(?:\d{2}|\d{4})',
+   str = str.gsub( '$SEASON$', '\d{4}/(?:\d{2}|\d{4})' )
+
+end
+
+
+
+def self.parse_heading_patches( txt )
+   patches = {}
+
+   header = nil
+
+   txt.each_line do |line|
+
+      line = line.strip
+      next  if line.empty? || line.start_with?('#')
+      break if line == '__END__'
+
+      ## check if heading
+      if m=%r{ ^
+              [ ]* =+ [ ]*
+                (?<text> .+?)
+              (?: [ ]* =+ )?
+               [ ]*
+                $
+              }x.match(line)
+
+          header = patches[m[:text]] = []
+      else
+          re =  mkheading_regex( line )
+          ## note - wrap in %r{^$}ix
+          header <<  %r{^ #{re} $}ix
+      end
+
+
+   end
+   patches
+end
+
+
+def self.read_heading_patches( path )  parse_heading_patches( read_text( path)); end
+
+
+
+
+
+def _patch_heading( txt, rxs, title )
+   found_match = false
+   rxs.each do |rx|
+     txt = txt.sub( rx ) do |match|
+               puts "  found heading match >#{match}< replace with >== #{title}<"
+
+               if title == '*'    ## use orginal title/ do NOT replace/normalize
+                  ## note - keep going with replacements here
+                  ##
+                  ##  note - autoremove (optional) trailing colon (:)
+                  ##                                    or  dot (.)
+                  ##              use .chomp(':') instead - why? why not?
+                  match = match.sub( /[.:]$/, '').rstrip
+                  "== #{match}\n"
+               else
+                  ## note - only short-circuit match if NOT generic replace
+                  found_match = true
+                  "== #{title}\n"
+               end
+             end
+     ## note - break on first match
+     break if found_match
+   end
+   txt
+end
+
+
+def patch_headings( txt, patches )
+
+     patches.each do |title, rxs|
+         txt = _patch_heading( txt, rxs, title )
+     end
+     txt
+end
+
+
+end    ## class Fmtfix
+end    ## module Rsssf
+
+
+
+
+
+__END__
+
+patches = read_headings( "./make/de_headings.txt" )
+pp patches
+
+txt = read_text( "../clubs/germany/pages/duit09.txt" )
+patch_headings( txt, patches )
+
+
+puts "bye"

data/lib/rsssf/fmtfix/rounds.rb

@@ -0,0 +1,74 @@
+module Rsssf
+class  Fmtfix    ## todo: find a better name e.g. Format or Fixer or ??
+
+
+##
+## note all "type" for round
+##  eg. round  26   -  todo/fix - later use squish to autofix!!
+
+
+## e.g. round 1, round 2, etc.
+##      matchday 1
+##      week 1
+#   note - add optional   Matchday 1 of 2 or such
+##      keep why? why not?
+#
+#  matchweek used by premerleague.com
+#  week used in msl/usa (no matchdays/rounds)
+#    note - matchweek might start on tuesday (e.g. tue to mon)
+#                or check if always 7day week?
+#
+# note - use 1-9 regex (cannot start with 0) - why? why not?
+#             make week 01 or round 01 or matchday 01 possible?
+
+ROUND_PAT_BASE = %q{
+         (   Round
+           | Matchday
+           | Matchweek
+           | Week )   [ ]{1,2}  [1-9][0-9]*
+
+        (?:    ## note - add optional   Matchday 1 of 2 or such
+               [ ] of [ ] [1-9][0-9]*
+        )?
+}
+
+
+
+##
+## add more pattern via config
+##
+##  todo/fix - check if .txt is empty
+##                 do NOT add ( || will match everything!!)
+##
+## rename names_misc to names_more - why? why not?
+
+ROUND_NAMES_EN   = read_patterns( "#{Rsssf.config_dir}/rounds_en.txt" )
+ROUND_NAMES_ES   = read_patterns( "#{Rsssf.config_dir}/rounds_es.txt" )
+ROUND_NAMES_MISC = read_patterns( "#{Rsssf.config_dir}/rounds_misc.txt" )
+
+ROUND_PAT = ROUND_PAT_BASE + ' | ' + ROUND_NAMES_EN.join( ' | ' ) +
+                             ' | ' + ROUND_NAMES_ES.join( ' | ' ) +
+                             ' | ' + ROUND_NAMES_MISC.join( ' | ' )
+## pp ROUND_PAT
+
+
+end    ## class Fmtfix
+end    ## module Rsssf
+
+
+__END__
+
+"\n" +
+"         (   Round\n" +
+"           | Matchday\n" +
+"           | Matchweek\n" +
+"           | Week )   [ ]{1,2}  [1-9][0-9]*\n" +
+"\n" +
+"        (?:    ## note - add optional   Matchday 1 of 2 or such\n" +
+"               [ ] of [ ] [1-9][0-9]*\n" +
+"        )?\n" +
+" | Preliminary [ ] round | (?:  First | Second | Third | Fourth | Fifth
+ | 1st | 2nd | 3rd | 4th | 5th ) [ ] (?:  round (?:  [ ] replays? )? | phase )
+ | Round [ ] one | (?:  1/32 | 1/16 | 16th | 1/8 | 8th | 1/4 | 1/2 ) [ ] finals?  (?:  [ ] replays? )?
+ | (?:  Eight | Quarter | Semi ) [ -]? finals? (?:  [ ] replays? )?
+ | Semis | Quarters | (?: Round [ ] of | Last ) [ ] (?:  4 | 8 | 16 | 32) | (?:  fifth | 5th ) [ -] place [ ] (?:  match | final | play[ -]?off ) | Match [ ] for [ ] (?:  fifth | 5th ) [ -] place | (?:  third | 3rd ) [ -] place [ ] (?:  match | final | play[ -]?off ) | Match [ ] for [ ] (?:  third | 3rd ) [ -] place | Finals? (?:  [ ] replays? )? | Final [ ] (?: round | group | pool) | (?: Minor | Major) [ ] Semi [ -]? finals? | Preliminary [ ] final | Final [ ] series | Grand [ ] final | Conference [ ] wild [ ] card [ ] round | Wild [ ] card [ ] games | (?: Eastern | Western) [ ] conference [ ] quarter [ -]? finals | (?: Eastern | Western) [ ] semi [ -]? finals | (?: Eastern | Western) [ ] conference [ ] final | Conference [ ] quarterfinals | Conference [ ] semifinals | Conference [ ] finals | Divisional [ ] finals | Championship [ ] final | MLS [ ] cup (?: [ ] final)? (?: [ ] \\d{4})? | Qualification [ ] MLS | Super [ ] semi [ -]? final | Super [ ]? cup [ ] final | Deciding [ ] match | Decider | play-?offs? [ ] round | Prime [ ] round | Round [ ] (?: 1|2|3) [ ] \\( no [ ] extra [ ] time \\) | Quarter [ -]? finals? [ ] \\( no [ ] extra [ ] time \\) | Moved [ ] match(?: es)? | match(?: es)? [ ] from [ ] round [ ] \\d{1,2} | Replays? | Replayed [ ] match(?: es)? | (?:  First | Second | Third | Fourth | Fifth | 1st | 2nd | 3rd | 4th | 5th ) [ ] stage | (?:  Regular | Group | League | Playoff ) [ ] stage | Knock-?out [ ] stage | Regular [ ] season | Tournament [ ] proper | Championship [ ] play-?offs? | Europa [ ] league [ ] play-?offs? | Conference [ ] league [ ] play-?offs? | Promotion [ ] (?: play-?offs? | match(?: es)?) | Relegation [ ] (?: play-?offs? | match(?: es)?) | Promotion/relegation [ ] (?: play-?offs? | match(?: es)?) | Promotion/relegation [ ] (?: play-?offs? | match(?: es)?) [ ] (?:  1st/2nd | 2nd/3rd | 3rd/4th) [ ] level | Relegation [ ] playout | Play-?out | Play-in [ ] round [ ] (?:  A | B | A-B ) | play-?offs? | play-?offs? [ ] (?: 1 | 2)? | (?:  First | Second | Third | Fourth | Fifth | 1st | 2nd | 3rd | 4th | 5th ) [ ] legs? (?:  [ ] replays? )? | Third [ ] leg [ ] minigame | Playoffs [ ] \\( Liguilla \\) | Recalificación | Round [ ] of [ ] 64 [ ] - [ ] 32 [ ] avos [ ] de [ ] final | Round [ ] of [ ] 32 [ ] - [ ] 16 [ ] avos [ ] de [ ] final | Round [ ] of [ ] 16 [ ] - [ ] Octavos [ ] de [ ] Final | Quarter [ ] finals [ ] - [ ] Cuartos [ ] de [ ] final | Semi [ ] finals [ ] - [ ] Semifinales | Primera [ ] fase [ ] de [ ] zonas [ ] - [ ] Phase [ ] of [ ] groups | Play-off [ ] o [ ] umístění | Skupina [ ] o [ ] záchranu"

data/lib/rsssf/fmtfix/score.rb

@@ -0,0 +1,92 @@
+module Rsssf
+class  Fmtfix    ## todo: find a better name e.g. Format or Fixer or ??
+
+
+def handle_score( txt )
+
+    ## fix typos - move to errata
+    txt = txt.gsub( 'paet, 3-4 pen]', '[aet, 3-4 pen]' )
+
+
+
+
+    ###  [aet]   => (aet)         -- after extra time
+    ##   [asdet] => (asdet)      -- after sudden death extra time
+    txt = txt.gsub( '[aet]',   '(aet)' )
+    txt = txt.gsub( '[asdet]', '(asdet)' )
+
+
+
+
+   ## [aet, 2-3 pen] => (aet, 2-3 pen)
+   ##  [aet, 9-10 pen]
+   ## [aet, 2-3pen]
+   ## [aet, 7-6pen]
+
+   txt = txt.gsub( %r{
+                         \[
+                            aet[,;.] [ ]?
+                             (\d{1,2}-\d{1,2}) [ ]? pen
+                         \]
+                    }ix,
+                    '(aet, \1 pen)')
+
+
+###  [aet, pen 4-3]
+##   [aet, pen 2-4]
+###    =>
+##      (aet, 2-4 pen)
+   txt = txt.gsub( %r{
+                         \[
+                            aet[,;.] [ ]?
+                             pen [ ] (\d{1,2}-\d{1,2})
+                         \]
+                    }ix,
+                    '(aet, \1 pen)')
+
+ ##   [5-4 pen]
+ ##   [3-4 pen]
+ ##   [1-3 pen], [1-3pen]
+   txt = txt.gsub( %r{
+                         \[
+                           (\d{1,2}-\d{1,2}) [ ]? pen
+                         \]
+                    }ix,
+                    '(\1 pen)')
+
+
+##   [Pen 4-1], [Pen 4-5], [Pen 1-3]
+##      =>
+##    (4-1 pen)
+   txt = txt.gsub( %r{
+                         \[
+                           pen [ ] (\d{1,2}-\d{1,2})
+                         \]
+                    }ix,
+                    '(\1 pen)')
+
+
+##  [5-3 PK], [6-5 PK]
+##      =>
+##   (6-5 pen)
+      txt = txt.gsub( %r{
+                           \[
+                              (\d{1,2}-\d{1,2}) [ ] PK
+                         \]
+                        }ix,
+                        '(\1 pen)'  )
+
+
+
+ ##
+ ## check special case usage - uniques?
+ ##    [8-7 pen(no extra time)]
+ ##    [Pen 2-4 (1-3?)]
+
+
+   txt
+end
+
+
+end    ## class Fmtfix
+end    ## module Rsssf