Class: RDoc::Markup::AttributeManager

Inherits:
Object
  • Object
show all
Defined in:
lib/rdoc/markup/attribute_manager.rb

Overview

Manages changes of attributes in a block of text

Constant Summary collapse

NULL =

The NUL character

"\000".freeze
A_PROTECT =

– We work by substituting non-printing characters in to the text. For now I’m assuming that I can substitute a character in the range 0..8 for a 7 bit character without damaging the encoded string, but this might be optimistic ++

004
PROTECT_ATTR =

Special mask character to prevent inline markup handling

A_PROTECT.chr

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initializeAttributeManager

Creates a new attribute manager that understands bold, emphasized and teletype text.



65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
# File 'lib/rdoc/markup/attribute_manager.rb', line 65

def initialize
  @html_tags = {}
  @matching_word_pairs = {}
  @protectable = %w[<]
  @regexp_handlings = []
  @word_pair_map = {}
  @attributes = RDoc::Markup::Attributes.new

  add_word_pair "*", "*", :BOLD
  add_word_pair "_", "_", :EM
  add_word_pair "+", "+", :TT

  add_html "em", :EM
  add_html "i",  :EM
  add_html "b",  :BOLD
  add_html "tt",   :TT
  add_html "code", :TT
end

Instance Attribute Details

#attributesObject (readonly)

The attributes enabled for this markup object.



29
30
31
# File 'lib/rdoc/markup/attribute_manager.rb', line 29

def attributes
  @attributes
end

#html_tagsObject (readonly)

This maps HTML tags to the corresponding attribute char



47
48
49
# File 'lib/rdoc/markup/attribute_manager.rb', line 47

def html_tags
  @html_tags
end

#matching_word_pairsObject (readonly)

This maps delimiters that occur around words (such as bold or tt) where the start and end delimiters and the same. This lets us optimize the regexp



36
37
38
# File 'lib/rdoc/markup/attribute_manager.rb', line 36

def matching_word_pairs
  @matching_word_pairs
end

#protectableObject (readonly)

A \ in front of a character that would normally be processed turns off processing. We do this by turning < into <#PROTECT



53
54
55
# File 'lib/rdoc/markup/attribute_manager.rb', line 53

def protectable
  @protectable
end

#regexp_handlingsObject (readonly)

And this maps _regexp handling_ sequences to a name. A regexp handling sequence is something like a WikiWord



59
60
61
# File 'lib/rdoc/markup/attribute_manager.rb', line 59

def regexp_handlings
  @regexp_handlings
end

#word_pair_mapObject (readonly)

And this is used when the delimiters aren’t the same. In this case the hash maps a pattern to the attribute character



42
43
44
# File 'lib/rdoc/markup/attribute_manager.rb', line 42

def word_pair_map
  @word_pair_map
end

Instance Method Details

#add_html(tag, name) ⇒ Object

Adds a markup class with name for words surrounded by HTML tag tag. To process emphasis tags:

am.add_html 'em', :EM


231
232
233
# File 'lib/rdoc/markup/attribute_manager.rb', line 231

def add_html(tag, name)
  @html_tags[tag.downcase] = @attributes.bitmap_for name
end

#add_regexp_handling(pattern, name) ⇒ Object

Adds a regexp handling for pattern with name. A simple URL handler would be:

@am.add_regexp_handling(/((https?:)\S+\w)/, :HYPERLINK)


241
242
243
# File 'lib/rdoc/markup/attribute_manager.rb', line 241

def add_regexp_handling pattern, name
  @regexp_handlings << [pattern, @attributes.bitmap_for(name)]
end

#add_word_pair(start, stop, name) ⇒ Object

Adds a markup class with name for words wrapped in the start and stop character. To make words wrapped with “*” bold:

am.add_word_pair '*', '*', :BOLD

Raises:

  • (ArgumentError)


208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
# File 'lib/rdoc/markup/attribute_manager.rb', line 208

def add_word_pair(start, stop, name)
  raise ArgumentError, "Word flags may not start with '<'" if
    start[0,1] == '<'

  bitmap = @attributes.bitmap_for name

  if start == stop then
    @matching_word_pairs[start] = bitmap
  else
    pattern = /(#{Regexp.escape start})(\S+)(#{Regexp.escape stop})/
    @word_pair_map[pattern] = bitmap
  end

  @protectable << start[0,1]
  @protectable.uniq!
end

#attribute(turn_on, turn_off) ⇒ Object

Return an attribute object with the given turn_on and turn_off bits set



87
88
89
# File 'lib/rdoc/markup/attribute_manager.rb', line 87

def attribute(turn_on, turn_off)
  RDoc::Markup::AttrChanger.new turn_on, turn_off
end

#change_attribute(current, new) ⇒ Object

Changes the current attribute from current to new



94
95
96
97
# File 'lib/rdoc/markup/attribute_manager.rb', line 94

def change_attribute current, new
  diff = current ^ new
  attribute(new & diff, current & diff)
end

#changed_attribute_by_name(current_set, new_set) ⇒ Object

Used by the tests to change attributes by name from current_set to new_set



103
104
105
106
107
108
109
110
111
112
113
114
# File 'lib/rdoc/markup/attribute_manager.rb', line 103

def changed_attribute_by_name current_set, new_set
  current = new = 0
  current_set.each do |name|
    current |= @attributes.bitmap_for(name)
  end

  new_set.each do |name|
    new |= @attributes.bitmap_for(name)
  end

  change_attribute(current, new)
end

#convert_attrs(str, attrs) ⇒ Object

Map attributes like textto the sequence 001002<char>001003<char>, where <char> is a per-attribute specific character



130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
# File 'lib/rdoc/markup/attribute_manager.rb', line 130

def convert_attrs(str, attrs)
  # first do matching ones
  tags = @matching_word_pairs.keys.join("")

  re = /(^|\W)([#{tags}])([#\\]?[\w:.\/-]+?\S?)\2(\W|$)/

  1 while str.gsub!(re) do
    attr = @matching_word_pairs[$2]
    attrs.set_attrs($`.length + $1.length + $2.length, $3.length, attr)
    $1 + NULL * $2.length + $3 + NULL * $2.length + $4
  end

  # then non-matching
  unless @word_pair_map.empty? then
    @word_pair_map.each do |regexp, attr|
      str.gsub!(regexp) {
        attrs.set_attrs($`.length + $1.length, $2.length, attr)
        NULL * $1.length + $2 + NULL * $3.length
      }
    end
  end
end

#convert_html(str, attrs) ⇒ Object

Converts HTML tags to RDoc attributes



156
157
158
159
160
161
162
163
164
165
166
# File 'lib/rdoc/markup/attribute_manager.rb', line 156

def convert_html(str, attrs)
  tags = @html_tags.keys.join '|'

  1 while str.gsub!(/<(#{tags})>(.*?)<\/\1>/i) {
    attr = @html_tags[$1.downcase]
    html_length = $1.length + 2
    seq = NULL * html_length
    attrs.set_attrs($`.length + html_length, $2.length, attr)
    seq + $2 + seq + NULL
  }
end

#convert_regexp_handlings(str, attrs) ⇒ Object

Converts regexp handling sequences to RDoc attributes



171
172
173
174
175
176
177
178
179
180
181
# File 'lib/rdoc/markup/attribute_manager.rb', line 171

def convert_regexp_handlings str, attrs
  @regexp_handlings.each do |regexp, attribute|
    str.scan(regexp) do
      capture = $~.size == 1 ? 0 : 1

      s, e = $~.offset capture

      attrs.set_attrs s, e - s, attribute | @attributes.regexp_handling
    end
  end
end

#copy_string(start_pos, end_pos) ⇒ Object

Copies start_pos to end_pos from the current string



119
120
121
122
123
# File 'lib/rdoc/markup/attribute_manager.rb', line 119

def copy_string(start_pos, end_pos)
  res = @str[start_pos...end_pos]
  res.gsub!(/\000/, '')
  res
end

#display_attributesObject

Debug method that prints a string along with its attributes



267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
# File 'lib/rdoc/markup/attribute_manager.rb', line 267

def display_attributes
  puts
  puts @str.tr(NULL, "!")
  bit = 1
  16.times do |bno|
    line = ""
    @str.length.times do |i|
      if (@attrs[i] & bit) == 0
        line << " "
      else
        if bno.zero?
          line << "S"
        else
          line << ("%d" % (bno+1))
        end
      end
    end
    puts(line) unless line =~ /^ *$/
    bit <<= 1
  end
end

#flow(str) ⇒ Object

Processes str converting attributes, HTML and regexp handlings



248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
# File 'lib/rdoc/markup/attribute_manager.rb', line 248

def flow str
  @str = str.dup

  mask_protected_sequences

  @attrs = RDoc::Markup::AttrSpan.new @str.length

  convert_attrs            @str, @attrs
  convert_html             @str, @attrs
  convert_regexp_handlings @str, @attrs

  unmask_protected_sequences

  split_into_flow
end

#mask_protected_sequencesObject

Escapes regexp handling sequences of text to prevent conversion to RDoc



186
187
188
189
190
191
192
193
# File 'lib/rdoc/markup/attribute_manager.rb', line 186

def mask_protected_sequences
  # protect __send__, __FILE__, etc.
  @str.gsub!(/__([a-z]+)__/i,
    "_#{PROTECT_ATTR}_#{PROTECT_ATTR}\\1_#{PROTECT_ATTR}_#{PROTECT_ATTR}")
  @str.gsub!(/(\A|[^\\])\\([#{Regexp.escape @protectable.join}])/m,
             "\\1\\2#{PROTECT_ATTR}")
  @str.gsub!(/\\(\\[#{Regexp.escape @protectable.join}])/m, "\\1")
end

#split_into_flowObject

Splits the string into chunks by attribute change



292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
# File 'lib/rdoc/markup/attribute_manager.rb', line 292

def split_into_flow
  res = []
  current_attr = 0

  str_len = @str.length

  # skip leading invisible text
  i = 0
  i += 1 while i < str_len and @str[i].chr == "\0"
  start_pos = i

  # then scan the string, chunking it on attribute changes
  while i < str_len
    new_attr = @attrs[i]
    if new_attr != current_attr
      if i > start_pos
        res << copy_string(start_pos, i)
        start_pos = i
      end

      res << change_attribute(current_attr, new_attr)
      current_attr = new_attr

      if (current_attr & @attributes.regexp_handling) != 0 then
        i += 1 while
          i < str_len and (@attrs[i] & @attributes.regexp_handling) != 0

        res << RDoc::Markup::RegexpHandling.new(current_attr,
                                                copy_string(start_pos, i))
        start_pos = i
        next
      end
    end

    # move on, skipping any invisible characters
    begin
      i += 1
    end while i < str_len and @str[i].chr == "\0"
  end

  # tidy up trailing text
  if start_pos < str_len
    res << copy_string(start_pos, str_len)
  end

  # and reset to all attributes off
  res << change_attribute(current_attr, 0) if current_attr != 0

  res
end

#unmask_protected_sequencesObject

Unescapes regexp handling sequences of text



198
199
200
# File 'lib/rdoc/markup/attribute_manager.rb', line 198

def unmask_protected_sequences
  @str.gsub!(/(.)#{PROTECT_ATTR}/, "\\1\000")
end