rsssf 0.2.0 → 0.3.0

data/lib/rsssf/fmtfix/tables.rb

@@ -0,0 +1,316 @@
+module Rsssf
+class  Fmtfix    ## todo: find a better name e.g. Format or Fixer or ??
+
+
+
+def self.table_heading_( line )
+  ## M   W  T  L  GF  GA  PTS  AVGE
+  ##  =>
+  ## (?:
+  ##   [ ]+ M  [ ]+  W [ ]+ T [ ]+ L [ ]+ GF [ ]+ GA [ ]+ PTS [ ]+ AVGE  [ ]*
+  ##  )
+
+   cols = line.strip.split( /[ ]+/ )
+
+   "(?: [ ]+ #{cols.join(' [ ]+ ')} [ ]*)"
+end
+
+
+
+
+
+##
+###   note - may start with blank line OR
+##              header
+##             followed by optional heading (e.g. M   W  T  L  GF  GA  PTS)
+##        and table lines ( 1. rapid 38 17 ...)
+
+
+
+##
+## note simple/compact table standing format needs more thinking
+##   will match
+## FC Schalke 04        1-3  1. FSV Mainz 05
+## Hannover 96          3-1  1. FC Nürnberg
+## FC Schalke 04        0-1  1. FC Kaiserslautern
+## Hannover 96          2-0  1. FSV Mainz 05
+## FSV Mainz 05             3-1 1. FC Köln
+##
+##   - add a required ranking  in the beginning e.g. 1., 2. or such?
+
+=begin
+                   | (?:  ##  or compact/min form  -- 22  37-15  51
+                          ##     maybe allow spaces later inbetween 37- 15 - why? why not?
+                          ##    1. 1. FC Köln   30 17 11  2  78- 40  45
+
+                         [ ]+ \d{1,3}                 ## played
+                          [ ]+ \d{1,3} [ ]? -[ ]? \d{1,3}      ##  gf-ga
+                          [ ]+  \d{1,3} \b            ##  pts
+                     )
+=end
+
+
+=begin
+             ## (i) table header
+             ##
+             ## fix - make header match more strict!!!
+             ##   e.g. do NOT match ---  or more than three spaces or such
+             ## exlcude in header
+             ##   NB:
+             ##    [*]
+             ##    [1]
+             ## exclude heading === e.g.
+             ##    ==== USL Premier Development
+                 ^
+                [ ]*
+
+
+              ## exclude comma (,) - why? why not?
+              ##   and numbers  - unless group 1
+              ## e.g. Kaczor 78 - Dreßel 19, Steinkogler 50,
+              ## B'schweig  2-1 Schalke    (Handschuh 38, Popivoda 55 - Fischer 82)
+              ##  M'gladbach 2-1 1. FC Köln (Jensen 6, Wittkamp 35 - D.Müller 78)
+              ##   Kraft 3, E.Kremers 38)
+              ##  Schalke     4-0 Tasmania    (Klose 2, 78, Herrmann 40, Kreuz 82)
+              ##
+              ## allow name such as
+              ##    USL - 1ST DIVISION (2nd Division)
+
+
+             (?<header>  [^=*:,0-9\[\]\n]+?
+                          ([ ] \d{1,2} \b)?   ## optional number only at the end e.g. group 1
+                     )
+                  :?  ## optional colon (:) e.g. final table:
+=end
+
+
+TABLE_HEADER_RE = %r{
+      ############
+      ## negative & positive lookaheads
+
+##         (?!
+##               .* [ ]{2,}       ## no (inline)  double (or more) spaces allowed
+##         )
+
+      (?:
+      ## (i)  can only start with non-zero number
+      ##      or alpha
+      ##
+      ##  A.  or
+      ##  1.  or
+      ##   mixed with dot  1A. yes/no?, A1. yes/no?,  1B1. ?
+      ##   1.K    - 1.Klasse
+
+      ##
+      ##  note - \b(oundary) - to always get complete tokens (alphanum) tokens
+      ##            note - \b includes [a-z0-9_] PLUS underscore (_)
+      ##                          check if underscore is \b
+      ##                              e.g.   09_  or _09 or  match \b[0-9]\b  ???
+      ##   use our own asserts?
+      ##      BNUM (boundary number) e.g. [^0-9]
+      ##      BALPHA (boundary alpha) e.g. [^a-z]
+      ##      BALNUM (boundary alphanum) e.g. [^a-z0-9]
+      ##    classic is   [^a-z0-9_]
+
+         (?<header>
+          (?=
+              .*  \p{L}+    ## must incl. alpha character - not only numbers!!
+          )
+
+  ## note
+  ##   order matters
+  ##   move specific first!!
+
+            \b
+             (?:     [0-9]+\p{L}  [0-9\p{L}]* \b    ## (ii) mixed alphanum (starting w/ num)
+                |    [0-9]+  \b  \.?  (?! \d)    ## (i)  num
+                |  \p{L}+[0-9]  [0-9\p{L}]* \b     ## (iiii) mixed alphanum (starting w/ alpha)
+                |  \p{L}+  \b    \.?              ## (iii) alpha
+             )
+             (?:
+                ## " (i-iiii) connector options  (a) single space
+                ##                                   -- exclude numbers on numbers (FIX)
+                ##                               (b) dash (-) or slash (/)
+                ##                                  --  must be alpha(.?)-alpha
+                ##                                        incl.  K.-H.  with trailing dot
+                ##                              add ampersand (&) too
+                ##                                    w/ leading & trailing opt space?
+                ##                                                incl.  K.&H., K. & H.
+               (?:   [ ]?
+                   | (?<! \d)  -   ## add negative lookbehind&ahead (no numbers please)
+                     (?! \d)
+                   |  /
+                )
+                    ## repeat (i-iiii) see above
+                       ## todo - do NOT allow numbers followed by numbers
+                \b
+                (?:  [0-9]+ \b    (?! [ ] \d)     ## (i) num - no more ordinals - why? why not?
+                  |  [0-9]+\p{L}  [0-9\p{L}]* \b     ## (ii) mixed alphanum (starting w/ num)
+                                                  ##     group 1a 1FC?? - why? why not?
+                  |  \p{L}+  \b   \.?              ## (iii) alpha
+                  |  \p{L}+[0-9]  [0-9\p{L}]* \b   ## (iiii) mixed alphanum (starting w/ alpha)
+               )
+            )*
+            (?:
+                 [ ]
+                \(  [^:()\[\]]+?  \)
+            )?
+        )  ## end-of-capture header
+   )
+   :?    ## optional colon (:) e.g. final table:
+}ix
+
+
+TABLE_RE = %r{
+
+         ### optional table header
+          (?:
+             ### negative lookahead
+             ##    MUST NOT match  standing line e.g.  10  3  4
+             ##      or         table heading (see below)
+             ##      or   -----  (old style structured heading left overs)
+                    (?! ^[ ]* (?:   [^\n]+?  [ ]+ \d{1,3} [ ]+ \d{1,3} [ ]+ \d{1,3}
+                                |   (?: GP | M | Team ) [ ]
+                                |  -{3,}
+                              )
+                     )
+
+             ## (i) table header
+             ##
+             ## fix - make header match more strict!!!
+             ##   e.g. do NOT match ---  or more than three spaces or such
+             ## exlcude in header
+             ##   NB:
+             ##    [*]
+             ##    [1]
+             ## exclude heading === e.g.
+             ##    ==== USL Premier Development
+                 ^
+                [ ]*
+
+
+              ## exclude comma (,) - why? why not?
+              ##   and numbers  - unless group 1
+              ## e.g. Kaczor 78 - Dreßel 19, Steinkogler 50,
+              ## B'schweig  2-1 Schalke    (Handschuh 38, Popivoda 55 - Fischer 82)
+              ##  M'gladbach 2-1 1. FC Köln (Jensen 6, Wittkamp 35 - D.Müller 78)
+              ##   Kraft 3, E.Kremers 38)
+              ##  Schalke     4-0 Tasmania    (Klose 2, 78, Herrmann 40, Kreuz 82)
+              ##
+              ## allow name such as
+              ##    USL - 1ST DIVISION (2nd Division)
+
+
+             (?<header>  [^=*:,0-9\[\]\n]+?
+                          ([ ] \d{1,2} \b)?   ## optional number only at the end e.g. group 1
+                     )
+                  :?  ## optional colon (:) e.g. final table:
+                  ## cut-off everything separated by more than three spaces
+                  ##   e.g. might be "inline" table heading (follow table header name)
+                  ##  e.g. Group 1                  M     W     T     L    GF    GA    DIF   PTS
+                  (?: [ ]{4,} (?: GP | M |Team ) [ ]  [^\n]+? )?
+              [ ]*
+             ## note - allow optional blank line - why? why not?
+             (?:  \n ^[ ]* )?
+             \n
+          )?
+
+
+      #### optional  table heading line
+      (?:  ^(?:
+          #{table_heading_( 'GP  W   L   D  GF  GA  PTS?' )}
+        | #{table_heading_( 'GP  W   L   T  GF  GA  PTS?' )}
+        | #{table_heading_( 'GP  W   T   L  GF  GA  PTS?' )}
+        | #{table_heading_( 'GP  W   D   L  GF  GA  PTS?' )}
+        ##  SW  sudden death win, SL sudden death lose
+        | #{table_heading_( 'GP  W   L  SW  GF  GA  PTS?' )}
+        | #{table_heading_( 'GP  W SW  SL   L   GF  GA  PTS?' )}
+        | #{table_heading_( 'GP  W SOW SOL  L  GF  GA PTS?'   )}
+        ##  mx/spanish
+        | #{table_heading_( 'M   W   T   L  GF  GC  DIF  PTS' )}
+        | #{table_heading_( 'M   W   T   L  GF  GA PTS AVGE' )}
+        | #{table_heading_( 'Team  M  W  T  L  GF-GA  PTS')}
+        | #{table_heading_( 'Team   M  W  T  L  GF-GA  PTS EP  TP')}
+        )
+       ## note - allow optional blank line - why? why not?
+          (?: \n ^[ ]* )?
+            \n
+      )?
+
+
+  ## MUST be followed by a table (standing) line
+  ## e.g.  1.FC Cincinnati    34  20  9  5  57-39  69
+  ##
+  ##   note - allow "run-on" e.g. LB14 on first number
+  ## Hudson Valley Quickstrike LB14  12   0   2   40   9   38
+  ## Hudson Valley Quickstrike LB12  11   1   0   26   9   33
+  ##
+  ##    17    11     5     1    40    16    +24    38
+  ##  or
+  ###  + 1.DC United                       32 17  6/ 3  6 65-43 57
+
+         ^
+         (?:
+               [^\n]+?
+                 (?:
+                    (?:
+
+                      \d{1,3}
+                 [ ]+ \d{1,3}  ## win
+ (?: [ ]+ | [ ]* / [ ]* ) \d{1,3}  ## draw
+                 [ ]+ \d{1,3}  ## lose
+                 [ ]+ \d{1,3}  (?:  [ ]* [:-] [ ]*
+                                  | [ ]+ )  \d{1,3}
+                 [ ]+ [+-]? \d{1,3} \b  # might be diff or point allow +/-!!
+                   )
+                 )
+               [^\n]*?
+          )
+         \n
+
+         ## eat-up the rest
+         .*?   ## non-greedy - match everything (incl. newline!) until
+                 (?:   \n (?= \n)    ## break on blank line (\n\n) or end-of-string/file
+                          | \z
+                 )
+
+}ixm
+
+
+
+
+def handle_tables( txt, tables: [] )
+
+
+   txt = txt.gsub( TABLE_RE ) do |match|
+
+                 m = Regexp.last_match
+
+                 puts "  proc table >#{m[:header]}< block:"
+                 puts ">>> (begin)"
+                 puts match
+                 puts "<<< (end)"
+
+                    ## remove everyting
+                    ##  or put in comment block later with command line option/switch!!
+                    ##    ''
+
+                     ## replace with "collapsed" marker
+
+
+
+                    tables << match
+                    table_id = tables.size
+                    if m[:header]   ## note - header might be missing
+                                    ##   table starting w/ blank line
+                       "<!-- $table#{table_id}$ - #{m[:header]} -->\n"
+                    else
+                       "<!-- $table#{table_id}$ -->\n"
+                    end
+                  end
+   txt
+end
+
+
+end    ## class Fmtfix
+end    ## module Rsssf

data/lib/rsssf/fmtfix/topscorers.rb

@@ -0,0 +1,50 @@
+module Rsssf
+class  Fmtfix    ## todo: find a better name e.g. Format or Fixer or ??
+
+##
+## process/handle Topscoreres: ... to first blank line (\n\n)
+
+
+## e.g.
+##  topscorer, topscorers
+##  top scorer, top scorers
+##  scorer, scorers
+
+TOPSCORERS_RE = %r{^     [ ]*
+                        (?<header>
+                           (?: top [ ]?)?  ## note - optional top
+                              scorers?      ## singular or plural
+                         )
+                          (?: [ ]* :)?    ## note - optional colon
+                           [ ]*
+                          \n{0,2}       ## note - optional leading blank line!!
+
+                        .*?             ## non-greedy - match everything until
+                      (?:   \n (?= \n)    ## blank line (\n\n) or end-of-string/file
+                          | \z
+                      )
+                    }ixm
+
+
+def handle_topscorers( txt, topscorers: [], opts: {} )
+   txt = txt.gsub( TOPSCORERS_RE ) do |match|
+                 if opts[:topscorers]
+                   puts "  proc topscorers block:"
+                   puts match
+                 end
+
+                    ## remove everyting
+                    ##  or put in comment block later with command line option/switch!!
+                    ##    ''
+
+                    ## replace with "collapsed" marker
+                      topscorers << match
+                    topscorers_id = topscorers.size
+                    "<!-- $topscorers#{topscorers_id}$ -->\n\n"
+                  end
+   txt
+end
+
+
+end    ## class Fmtfix
+end    ## module Rsssf

data/lib/rsssf/page-find_schedule.rb

@@ -0,0 +1,127 @@
+module Rsssf
+class Page
+
+
+###
+#  (experimental)
+#  machinery to split document by leagues & cups
+
+
+##
+## for now simply split
+##    on headings
+
+
+###  fix - support match with trailing ==== too!!!
+
+### note - starts at
+HEADER_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 _split_sections( txt, level: 2 )
+
+  sections = {}
+  current  = nil
+
+  txt.each_line do |line|
+    if m=HEADER_RE.match( line )
+        header_level  = m[:marker].size
+        header_text   = m[:text]
+        if header_level == level
+           current = String.new
+           sections[ header_text ] = current
+           next
+        end
+    end
+
+    current << line    if current
+  end
+
+  sections
+end
+
+
+
+
+
+## make header required - yes
+##   change to build_schedule - why? why not???
+##    add level: 2 or such - why? why not?
+def find_schedule!( header: )
+    _find_schedule( header: header, strict: true )
+end
+
+
+def _find_schedule( header:, strict: false )
+    ## make sure header is an array
+    header = [header]    if header.is_a?( String )
+
+    txt = _walk_sections( @txt, header: header,
+                                depth:  0,
+                                strict: strict )
+
+    if txt
+        ## wrap in schedule class - why? why not?
+        schedule = Schedule.new( txt )
+        schedule
+    else
+       nil
+    end
+end
+
+
+def _walk_sections( txt, header:,
+                         depth:,
+                         strict: false )
+
+   query      =  header[depth]
+   query_next =  header[depth+1]
+
+   ## note - start at level 2
+   sections = _split_sections( txt, level: depth+2 )
+
+   txt = sections[ query ]
+   if txt
+       if query_next
+         txt = _walk_sections( txt, header: header,
+                                    depth: depth+1,
+                                    strict: strict )
+         txt
+       else
+         txt
+       end
+   else
+      if strict
+        ## note - return nil if not found!!!
+        raise ArgumentError, "section with header >#{query}< not found; sections incl. #{sections.keys}"
+      else
+        nil
+      end
+   end
+end  # method _find_schedule
+
+
+
+end # class Page
+end # module Rsssf

data/lib/rsssf/page-meta.rb

@@ -0,0 +1,68 @@
+
+
+module Rsssf
+class Page
+
+
+##
+##  note -  \A - start of string
+##            comment must start .txt document!!!
+
+HTML_COMMENT_HEADER_RE = %r{  \A
+                            [ \n]*  ## trailing spaces and blank lines
+                       <!--
+                            [ \n]*
+                          (?<text> .+?)
+                            [ \n]*
+                        -->
+                   }imx
+
+
+
+
+###
+##   find meta data block (via html-style comment header )
+##    incl.   title, autor(s), url,  updated
+##  e.g.
+##    <!--
+##       title:   Austria 2024/25
+##       source:  https://rsssf.org/tableso/oost2025.html
+##       author:  Hans Schöggl
+##       updated: 7 Jul 2025
+##      -->
+##  -or-
+##      authors: Hans Schöggl and Karel Stokkermans
+
+
+
+
+
+def self.parse_meta( txt )
+     meta = {}
+     m = HTML_COMMENT_HEADER_RE.match( txt )
+     if m
+        text = m[:text]
+        text.each_line do |line|
+            line = line.strip
+
+            ## note - allow "inline" blank lines and comment lines (starting w/ #)
+            next if line.empty?  || line.start_with?('#')
+
+            ## split line on first colon (:) (only)
+            ##   note - limit split to two pieces!!!
+            key, value = line.split( /[ ]*:[ ]*/, 2)
+            ## use a symbol (not string) as key - why? why not?
+            meta[ key.to_sym ] = value
+        end
+        meta
+     else
+        nil ## no meta data (comment header) found
+     end
+end
+def parse_meta( txt ) self.class.parse_meta( txt ); end
+
+
+
+
+end # class Page
+end # module Rsssf