Class: Rsssf::PageConverter

Inherits:
Object
  • Object
show all
Defined in:
lib/rsssf/convert/errata.rb,
lib/rsssf/convert/convert.rb,
lib/rsssf/convert/html_to_txt.rb,
lib/rsssf/convert/html_entities.rb,
lib/rsssf/convert/html_to_txt/replace_hr.rb,
lib/rsssf/convert/html_to_txt/remove_emails.rb,
lib/rsssf/convert/html_to_txt/replace_a_href.rb,
lib/rsssf/convert/html_to_txt/replace_a_name.rb,
lib/rsssf/convert/html_to_txt/replace_heading.rb,
lib/rsssf/convert/html_to_txt/beautify_anchors.rb

Constant Summary collapse

ENTITIES = 
%w[
À   À
Á   Á
   Â
à   Ã
Ä   Ä
Å   Å

à   à
á   á
â   â
ã   ã
ä   ä
å   å
Æ   Æ
æ   æ
ß   ß
Ç   Ç
ç   ç
È   È
É   É
Ê   Ê
Ë   Ë
è   è
é   é
ê   ê
ë   ë

ð   ð

Ì   Ì
Í   Í
Î   Î
Ï   Ï
ì   ì
í   í
î   î
ï   ï
Ñ   Ñ
ñ   ñ
Ò   Ò
Ó   Ó
Ô   Ô
Õ   Õ
Ö   Ö
ò   ò
ó   ó
ô   ô
õ   õ
ö   ö
Ø   Ø
ø   ø
Ù   Ù
Ú   Ú
Û   Û
Ü   Ü
ù   ù
ú   ú
û   û
ü   ü
Ý   Ý
ý   ý
ÿ   ÿ

<    &lt;
>    &gt;
&    &amp;
©    &copy;
®    &reg;

]
HR_LINE_ASCII = 
"\n\n=-=-=-=-=-=-=-=-=-=-=-=-=-=-=\n\n"
EMAIL_RE = 
%r{ \s*
\(
 [a-z][a-z0-9_]+
   @[a-z]+(\.[a-z]+)+
 \)
}imx
A_HREF_RE =

<A href=“www.rsssf.org/”>Rec.Sport.Soccer

      Statistics Foundation</A> 
<A href="http://www.rsssfbrasil.com">RSSSF 
  Brazil</A> 

and Daniel Dalence (<A
 href="mailto:danielballack@terra.com.br">danielballack@terra.com.br</A>)

empty
<a>Primer Descenso – First Relegation</a>

%r{<A 
    (?: 
       \s+ HREF [ ]* = 
          (?<href>[^>]+?)
    )?
  >
    (?<title>.+?)
  <\/A>
}imx
A_NAME_OLD_RE =

note - for content use non-greedy to allow

match of tags inside content too

%r{<A [ ]+ NAME [ ]* =
     (?<name>[^>]+?) 
  >
     (?<title>.+?) 
  </A>
}imx
A_NAME_RE = 
%r{<A [ ]+ NAME [ ]* =
     (?<name>[^>]+?) 
  >
}imx
HEADING_RE =

note - for h1,h2,h3,h4,h5,h6

use a backref(erence) e.g. \1

note - include leading and trailing spaces (incl. newlines) !!!

note - for content inside use non-greedy to allow

match of tags inside content too

%r{ \s*
  <H(?<level>[1-6])>
    (?<title> .+?)
  </H\k<level>>
  \s*
}imx
BOLD_OR_UNDERLINE_LINE_HEADING_RE =

note - MUST be a one a single line (see make heading for more)

e.g.    "<h#{tag}>#{text}</h#{tag}>"

%r{^
       [ ]*
  <H (?<tag> [BU]) >
    (?<title> .+?)
  </H \k<tag> >
      [ ]*
       $
}ix

Class Method Summary collapse

Instance Method Summary collapse

Class Method Details

.convert(html, url:) ⇒ Object

convenience helper



6
7
8
9
 
# File 'lib/rsssf/convert/convert.rb', line 6

def self.convert( html, url: )
  @@converter ||= new   ## use a "shared" built-in converter
  @@converter.convert( html, url: url )
end

.convert_html_entities(html, url: nil) ⇒ Object 



81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
 
# File 'lib/rsssf/convert/html_entities.rb', line 81

def self.convert_html_entities( html, url: nil )
  ## check for html entities
  html = html.gsub( "&auml;", 'ä' )
  html = html.gsub( "&ouml;", 'ö' )
  html = html.gsub( "&uuml;", 'ü' )
  html = html.gsub( "&Auml;", 'Ä' )
  html = html.gsub( "&Ouml;", 'Ö' )
  html = html.gsub( "&Uuml;", 'Ü' )
  html = html.gsub( "&szlig;", 'ß' )


  html = errata_html_entities( html )


  ENTITIES.each_slice(2) do |str, entity|
     html = html.gsub( entity, str )
  end



  ##############
  ## check for more entities
  ##   limit &---; to length 10 - why? why not?


  ## check for decimal entities (mapping 1:1 to unicode) 
  html = html.gsub(/&#(\d+);/) do |match|
           uni =  if match == '&#307;'   ## use like Van D&#307;k  -> Van Dijk
                   'ij'
                  else
                   [$1.to_i].pack("U")          
                  end 
       
            ##puts "   converting numeric html entity #{match} to unicode char #{uni}"

           uni
        end

      
  html = html.gsub( /&[^; ]{1,10};/) do |match|
         ##   ignore weird edge case of &A;
         ##    e.g. [M&A; moved from pool B] - where M&A is name of club
         ##
         ##  in ital03.html:
         ###    [Eugenio Corini 22pen&36pen; Christian Vieri 69]
         ##     Francesco Totti 31, Vincenzo Montella 49&68; Antonio Di Natale 11] 

             if match == '&A;' ||
                match == '&36pen;' || match == '&68;'
             else
                msg = "found unencoded html entity #{match}"
                msg += " in >#{url}<"   if url

                puts "*** WARN - #{msg}"
                log( msg )  ## log too (see log.txt)
             end
             
             match   ## pass through as is (1:1)
  end
  
  html
end

.errata_html(html) ⇒ Object 



10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 
# File 'lib/rsssf/convert/errata.rb', line 10

def self.errata_html( html )
   ## auto-fix known typos / errors
   ###   kind of PRE-processing, see errata_txt for POST-processing
   ###  check - rename to errata_pre/post - why? why not?


     ## quick fix - rm </ADDRESS>
    ##  </ADDRESS>
    ##   tablesb/braz94.html
    html = html.gsub( '</ADDRESS>', ''  )

   ## quick fix   </a  => </a>   
   ##  <a href="#play6">Gold League (Calle 6)</a
   ##  <a href="#zpl">PBZ Premier League 2025/26</a
   ##  <a href="#lig1">Championnat National Ligue 1</a

   html = html.gsub( /<\/A
                          (?! [ ]*>)     ## negative lookahead
                      /ix, '</A>' )

   ## quick fix  </br>  => <br>
   html = html.gsub( /<\/BR>/i, '<BR>' )



  ## quick fix - change typo <H1></H2>
  ##  tables/58full.html
  html = html.gsub( '<H1>Quarterfinals</H2>', '<H2>Quarterfinals</H2>' ) 

  ## quick fix - change typo <M>,<N> to <B>
  ##   tables/54full.html
  html = html.gsub( '<M>MEX</B>', '<B>MEX</B>' ) 
  ##   tables/58full.html 
  html = html.gsub( '<N>CZE</B>', '<B>CZE</B>' ) 


  ## quick fix -
  ##   tablesb/braz88.html 
  html = html.gsub( '<</TITLE>', '</TITLE>' )

  
  ## quick fix
  ##   hr (horizontal rule) via img
  ##   in  tables/30full.html and others
  ##   
  ## <IMG SRC="xshadow.gif.pagespeed.ic.AbdeNVcmzw.png" ALT="-----------">
  ##   look  for
  ## <IMG   ALT="---">
  html = html.gsub(  /<IMG
                            [^>]+?
                           ALT="-{3,}"
                         >/ixm, '<HR>' )


  html
end

.errata_html_entities(html) ⇒ Object 



69
70
71
72
73
74
75
76
77
78
 
# File 'lib/rsssf/convert/errata.rb', line 69

def self.errata_html_entities( html )
    ########
    ## typos / autofix - keep - why? why not?
    html = html.gsub( "&oulm;", 'ö' )    ## support typo in entity (&ouml;)
    html = html.gsub( "&uml;",  'ü' )    ## support typo in entity (&uuml;) - why? why not?
    html = html.gsub( "&slig;", "ß" )    ## support typo in entity (&szlig;)
    html = html.gsub( "&aaacute;", "á" )  ## typo for &aacute; 
    html = html.gsub( "&nitlde;", "ñ" )  ## typ for &ntilde;   
    html
end

.log(msg) ⇒ Object

more helpers



58
59
60
61
62
63
64
65
 
# File 'lib/rsssf/convert/convert.rb', line 58

def self.log( msg )
  ## append msg to ./logs.txt
  ##     use ./errors.txt - why? why not?
  File.open( './logs.txt', 'a:utf-8' ) do |f|
    f.write( msg )
    f.write( "\n" )
  end
end

Instance Method Details

#beautify_anchors(html) ⇒ Object 



5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
 
# File 'lib/rsssf/convert/html_to_txt/beautify_anchors.rb', line 5

def beautify_anchors( html )
  ## beautify 
  ##  ‹§2fin›
  ##
  ## == Semifinals
  ##
  ##  merge anchor (a name) with heading into one line e.g.
  ##       => 
  ##  == Semifinals  ‹§2fin›

   html = html.gsub( /\s*
                          (?<name>ܤ
                                    [^›]+?
                                 ›)
                      \s*
                          (?<heading>={2,}
                              [^=\n]+? 
                          )
                       \n
                       \s*/ixm ) do |match|
   
           m = Regexp.last_match
 
           match = match.gsub( "\n", '$$' )  ## make newlines visible for debugging
           puts "   mergeing anchor (a name) with heading into one line - >#{match}<" 

           "\n\n#{m[:heading]}  #{m[:name]}\n\n"
    end

###
## 
## beautify 
##  ‹§argsquad›Argentine Squad Full Info
##  ‹§eng›ENGLAND
##
##
##  reformat anchor (a name) start line with text  e.g.
##       => 
##  Argentine Squad Full Info  ‹§argsquad›
##  ENGLAND  ‹§eng›

   html = html.gsub( /\n
                          (?<name>ܤ
                                    [^›]+?
                                 ›)
                      [ ]*
                          (?<text>[^\n]+? 
                          )
                       \n
                       /ixm ) do |match|
   
           m = Regexp.last_match
 
           match = match.gsub( "\n", '$$' )  ## make newlines visible for debugging
           puts "   move anchor (a name) starting line with text to end - >#{match}<"

           "\n#{m[:text]}  #{m[:name]}\n"
    end

###
## beautify heading
##   ==== ‹§gra›Group A
##     =>
##   ==== Group A  ‹§gra›

   html = html.gsub( /\n
                          (?<heading_marker>
                               ={2,})
                               [ ]*
                          (?<name>ܤ
                                    [^›]+?
                                 ›)
                             [ ]*
                          (?<heading_text>[^\n]+? 
                          )
                       \n
                       /ixm ) do |match|
   
           m = Regexp.last_match
 
           match = match.gsub( "\n", '$$' )  ## make newlines visible for debugging
           puts "   move anchor (a name) in heading to end - >#{match}<"

           "\n#{m[:heading_marker]} #{m[:heading_text]}  #{m[:name]}\n"
    end
    html
end

#convert(html, url:) ⇒ Object

add anchor: options or such

lets you toggle adding anchors (§premier etc.) - why? why not?



18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
 
# File 'lib/rsssf/convert/convert.rb', line 18

def convert( html, url: )
  ### todo/fix: first check if html is all ascii-7bit e.g.
  ## includes only chars from 64 to 127!!!

  ## normalize newlines
  ##   replace \r\n (form feed \r) used by Windows - ff+lf;
  ##         just use \n (new line a.k.a. line feed)
  html = html.gsub( "\r\n", "\n" )

  ##  convert tabs to two spaces (or use four??)
  html = html.gsub( "\t", '   ' )



  html = convert_html_entities( html, url: url )

 ###################################
 ### smart quotes quick fixes
 ### convert all "smart" quote to (standard) single and double quotes
 ##  D´Alessandro   =>  D'Alessandro
 ##    81´ and 88'   =>  81' and 88'


  html = html.gsub( /[´’‘]/, "'" )
  html = html.gsub( /[“”]/,  '"' )

### convert fancy (unicode) dashes/hyphens to plain dash/hyphen
   html = html.gsub( '–', '-' )



  txt   = html_to_txt( html, url: url )
  txt
end

#convert_html_entities(html, url: nil) ⇒ Object 



143
 
# File 'lib/rsssf/convert/html_entities.rb', line 143

def convert_html_entities( html, url: nil ) self.class.convert_html_entities( html, url: url ); end

#errata_html(html) ⇒ Object 



66
 
# File 'lib/rsssf/convert/errata.rb', line 66

def errata_html( html ) self.class.errata_html( html ); end

#errata_html_entities(html) ⇒ Object 



79
 
# File 'lib/rsssf/convert/errata.rb', line 79

def errata_html_entities( html ) self.class.errata_html_entities( html ); end

#errata_txt(txt) ⇒ Object 



83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
 
# File 'lib/rsssf/convert/errata.rb', line 83

def errata_txt( txt )
  ## kind-of POST-processing, see errata_html for PRE-processing

   ## quick fix - squish spaces (to single)
   ##   tables/82full.html
   txt = txt.gsub( 'Second  phase', 'Second phase' )


   ## quick fix - add (missing) closing bracket (])
   ##   tables/70q.html
   txt = txt.gsub(/^South America Group 10 \[Brazil$/,
                   'South America Group 10 [Brazil]' )


  txt
end

#html_to_txt(html, url:) ⇒ Object 



17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
 
# File 'lib/rsssf/convert/html_to_txt.rb', line 17

def html_to_txt( html, url: )

###
#   todo: check if any tags (still) present??


  ## cut off everything before body
  ##   
  ## note - might incl. attributes e.g.
  ## <body bgcolor="yellow">

  ## record / track (important) edits - sub(stitutions) etc.
  edits = []


  html = html.sub( /.+?
                      <BODY [^>]*? >
                      \s*
                   /xim, 
                   '' )
                  
  ## special case i)   no <body> - cut-off head if present
  ## cut off everything before <head/>
  ##   used in braz93.html, braz98.html
  html = html.sub( /.+?
                     <\/HEAD>
                       \s*
                   /xim, '' )    

  ## special case ii) no <body>, no </head> 
  ## cut off everything before <head/>
  ##   used in braz93.html, braz98.html
  html = html.sub( /.+?
                     <HEAD\/>
                       \s*
                   /xim, '' )    




  ## cut off everything after body (closing)
  html = html.sub( /<\/BODY>.*/im, '' )
  
  ## special case
  ## cut off everything after </html> (closing)
  ##   used in braz93.html, braz98.html
  html = html.sub( /<\/HTML>.*/im, '' )




  ## quick fix
  ## <title>World Cup 1950 qualifications</title>
  ## <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-2">
  
  ## remove title and meta
  html = html.sub( /<TITLE>.*?<\/TITLE>/i, '' )
  html = html.sub( /<META .*?>/i, '' )


  ## (auto-)fix known types / errors
  ##  todo - pass in/along filename/url too - why? why not?
  html = errata_html( html )



  ##
  ##   change ^<b><a name ...></a></b>$  or    <hb>  - heading "bold" - might be h5
  ##          ^<u><a name ...></a></u>$    to  <hu>  - heading "underscore" - might be h6
  html, more_edits = make_heading( html )
  edits += more_edits 






  ## remove cite
  html = html.gsub( /<CITE>([^<]+)<\/CITE>/im ) do |_|
    puts " remove cite >#{$1}<"
    "#{$1}"
  end


  html = replace_hr( html )




  ## replace break (br)
  ## note: do NOT use m/multiline for now - why? why not??
  html = html.gsub( /<BR>\s*/i ) do |match|    ## note: include (swallow) "extra" newline
    match = match.gsub( "\n", '$$' )  ## make newlines visible for debugging
    puts " replace break (br) - >#{match}<"
    "\n"
  end



  
  html = replace_a_name( html )

  html = replace_a_href( html )

  ## quickfix  remove trailing </a> left possibly by a_name
  html = html.gsub( /<\/A>/i, '' )



  ## replace paragrah (p)
  html = html.gsub( /\s*<P>\s*/im ) do |match|    ## note: include (swallow) "extra" newline
    match = match.gsub( "\n", '$$' )  ## make newlines visible for debugging
    puts " replace paragraph (p) - >#{match}<"
    "\n\n"
  end
  html = html.gsub( /<\/P>/i, '' )  ## replace paragraph (p) closing w/ nothing for now



 
  html = replace_heading( html )


  ## remove i(talics)
  ##    use non-greedy match as default? e.g. .*? - why? why not?
  ##    or greedy w/  ([^<]+)
  html = html.gsub( /<I>(.*?)<\/I>/im ) do |_|
    puts " remove italic (i) >#{$1}<"
    "#{$1}"
  end

  html = html.gsub( /<U>(.*?)<\/U>/im ) do |_|
    puts " remove underline (u) >#{$1}<"
    "#{$1}"
  end

  ## remove b   - note: might include anchors (thus, call after anchors)
  ###   use non-greedy match as default? e.g. .*? - why? why not?
  ## was - "**#{$1}**"
  html = html.gsub( /<B>(.*?)<\/B>/im ) do |_|
    puts " remove bold (b) >#{$1}<"
    "#{$1}"  
  end

  ## <strong></strong>
  html = html.gsub( /<STRONG>(.*?)<\/STRONG>/im ) do |_|
    puts " remove strong (strong) >#{$1}<"
    "#{$1}"  
  end



  ## replace preformatted (pre)
  html = html.gsub( /<PRE>|<\/PRE>/i ) do |match|
    puts " replace preformatted (pre)"

      ## note - replace preformatted blocks
      ##           with comments 
      ##  was:
      ##  ''  # replace w/ nothing for now (keep surrounding newlines)
      
     if match.downcase == '<pre>'
         '<!-- start pre -->'
     else
         '<!-- end pre -->'
     end
  end


=begin
  puts
  puts
  puts "html:"
  puts html[0..2000]
  puts "-- snip --"
  puts html[-1000..-1]   ## print last hundred chars
=end


  html = remove_emails( html )


  html = beautify_anchors( html )


  ## check for html tags
  ##  left
  ##  use scan instead of 
  html.gsub( /<
                \/?
                [A-Z]+ [^>]*
              > 
             /xim ) do |match|

          if ['<menu>', '<ul>', '<li>',
               '</menu>', '</ul>', '</li>'].include?(match.downcase) 
               ## do nothing
          else 
                  msg = "found unprocessed html tag #{match} in >#{url}<"
                  puts "*** WARN - #{msg}"
                  log( msg )  ## log too (see log.txt)
          end 
          match
    end


  ##
  ## todo/fix
  ##    move up-front - kind of preprocessing (not post) - why? why not?

  ## cleanup whitespaces
  ##   todo/fix:  convert newline in space first
  ##                and than collapse spaces etc.!!!
  txt = String.new
  html.each_line do |line|
     line = line.gsub( "\t", '  ' ) # replace all tabs w/ two spaces for nwo
     line = line.rstrip             # remove trailing whitespace (incl. newline/formfeed)

     txt << line
     txt << "\n"
  end

  txt = errata_txt( txt )

  [txt, edits]
end

#log(msg) ⇒ Object 



66
 
# File 'lib/rsssf/convert/convert.rb', line 66

def log( msg ) self.class.log( msg ); end

#remove_emails(html) ⇒ Object 



13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
 
# File 'lib/rsssf/convert/html_to_txt/remove_emails.rb', line 13

def remove_emails( html )
  ### remove converted ("blinded") mailto anchors
  ##  note   usually inside () e.g.
  ##    (‹mailto›) 
  ##   plus slurp up all leading whitespace (incl. newline) - why? why not?
  html = html.gsub( /\s*
                      \(‹mailto›\)
                     /xm, '' )
  
   ###
   ##  remove "regular emails too e.g.
   ##
   ## Thanks to Marcelo Leme de Arruda (___@___.__.br),
   ##  Ricardo FF Pontes (___@____.com), 
   ## Santiago Reis (____@____.com.br),
   ## Marcos Lacerda Queiroz (___@____.com.br)
   ##  etc.

  ## check for "free-standing e.g. on its own line" emails only for now
   html = html.gsub( EMAIL_RE ) do |match|
    puts "removing  email >#{match}<"
    ''   
   end
   html
end

#replace_a_href(html) ⇒ Object 



29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
 
# File 'lib/rsssf/convert/html_to_txt/replace_a_href.rb', line 29

def replace_a_href( html )
  ## remove anchors (a href)
  #    note: heading 4 includes anchor (thus, let anchors go first)
  #  note: <a \newline href is used for authors email - thus incl. support for newline as space
  html.gsub( A_HREF_RE ) do |match|   ## note: use .+? non-greedy match
    m = Regexp.last_match
    captures = m.named_captures
    href  = if m['href']
               m['href'].gsub( /["']/, '' ).strip   ## remove ("" or '')
            else
               nil
            end
    title = m['title'].strip   ## note: "save" caputure first; gets replaced by gsub (next regex call)


    if href.nil?
       ## report error - <a>hello</a> is useless
       puts " replace anchor w/ missing (!!) href (a) >#{title}<"
      "‹#{squish(title)}›"
  
    ## e.g.
    ##  ‹Larsen23@gmx.de, see page mailto:Larsen23@gmx.de›
    ##  ‹danielballack@terra.com.br, see page mailto:danielballack@terra.com.br›
    ##  ‹zja70@aol.com, see page mailto:zja70@aol.com›)
  
    elsif href.start_with?( 'mailto:')
      puts " blank mailto  -  anchor (a) href >#{href}, >#{title}<"
      '‹mailto›'   ## delete/remove email
    else
      puts " replace anchor (a) href >#{href}, >#{title}<"

      ## convert href to xref
      xref = if href.start_with?('#')    ## in-page ref
              ", see §#{href[1..-1]}"
             elsif href.start_with?( /https?:/ )            ## external page ref
               ## skip - keep empty - why? why not? (or add url domain?)
               ''
             else
               ## hack - check for some custom excludes  
               if title.start_with?( 'Rec.Sport.Soccer' )
                    ## skip - keep empty
                    '' 
               else   
                 ## strip (ending)  .htm|html
                 ", see page #{href.sub( /\.html?$/,'')}"
               end
             end

      "‹#{squish(title)}#{xref}›"
    end
  end
end

#replace_a_name(html) ⇒ Object 



56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
 
# File 'lib/rsssf/convert/html_to_txt/replace_a_name.rb', line 56

def replace_a_name( html )

  ## note - allows <a name=""> without closing </a>
  ##    <a name="semi"><H2>Semifinals</H2>
  ##   always put anchor on its own line for now

  ##
  ## remove (named) anchors
  html.gsub( A_NAME_RE ) do |match|   ## note: use .+? non-greedy match
    m = Regexp.last_match
    
    name = m[:name].gsub( /["']/, '' ).strip   ## remove ("" or '')
    match = match.gsub( "\n", '$$' )  ## make newlines visible for debugging
    puts " replace anchor (a) name >#{name}<    -    >#{match}<"
  
   ##
   ## todo - report WARN if title incl. tags
   ##    assumes text only for now - why? why not?
   ##  add a name inside heading !!!
   ##  do NOT add heading inside a name !!!

    "‹§#{name}›"   ## note - use two spaces min (between title & name)
  end
end

#replace_a_name_old(html) ⇒ Object 



33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
 
# File 'lib/rsssf/convert/html_to_txt/replace_a_name.rb', line 33

def replace_a_name_old( html )
  ##
  ## remove (named) anchors
  html.gsub( A_NAME_RE ) do |match|   ## note: use .+? non-greedy match
    m = Regexp.last_match
    name = m[:name].gsub( /["']/, '' ).strip   ## remove ("" or '')
    title = m[:title].strip   ## note: "save" caputure first; gets replaced by gsub (next regex call)
    match = match.gsub( "\n", '$$' )  ## make newlines visible for debugging
    puts " replace anchor (a) name >#{name}<, >#{title}<    -    >#{match}<"
  

   ##
   ## todo - report WARN if title incl. tags
   ##    assumes text only for now - why? why not?
   ##  add a name inside heading !!!
   ##  do NOT add heading inside a name !!!

    "#{title}  ‹§#{name}›"   ## note - use two spaces min (between title & name)
  end
end

#replace_heading(html) ⇒ Object 



34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
 
# File 'lib/rsssf/convert/html_to_txt/replace_heading.rb', line 34

def replace_heading( html )
   html = html.gsub( HEADING_RE ) do |_|
      m = Regexp.last_match

      level = m[:level].to_i(10) 
      title = m[:title]

      puts " replace heading #{level} (h#{level}) >#{title}<"

      ## note: make sure to always add two newlines before and after
      "\n\n#{'='*level} #{title}\n\n"    
      
   end

   html = html.gsub( BOLD_OR_UNDERLINE_LINE_HEADING_RE ) do |_|
      m = Regexp.last_match

      tag = m[:tag].downcase 
      title = m[:title]

      ## use heading 5 for b and heading 6 for underline for now
      ##   maybe later change to custom  ==_ or ==* or such
      ##     to mark the heading (sourced via bold/underscore) ???
      level =  if tag == 'b'
                    5
               elsif tag == 'u'
                    6 
               else
                   raise ArgumentError, "b(old)|u(underscore) tag expected; got #{tag}"
               end

      puts " replace #{tag}-heading #{level} (h#{level}) >#{title}<"

      ## note: do NOT add any newlines before and after
      "#{'='*level} #{title}"    
   end
 
  html
end

#replace_hr(html) ⇒ Object 



9
10
11
12
13
14
15
16
17
18
 
# File 'lib/rsssf/convert/html_to_txt/replace_hr.rb', line 9

def replace_hr( html )

  html = html.gsub( /\s*<HR>\s*/im ) do |match|
    match = match.gsub( "\n", '$$' )  ## make newlines visible for debugging
    puts " replace horizontal rule (hr) - >#{match}<"
    HR_LINE_ASCII    ## check what hr to use use  - . - . - or =-=-=-= or somehting distinct?
  end
 
  html
end

#squish(str) ⇒ Object 



9
10
11
12
 
# File 'lib/rsssf/convert/html_to_txt.rb', line 9

def squish( str )
   ## squish more than one white space to one space
   str.gsub( /[ \r\t\n]+/, ' ' )
end