summaryrefslogtreecommitdiff
path: root/jni/ruby/lib/rdoc/markup/attribute_manager.rb
blob: ce4ac76d5b25bb2742e1f95f6f42fbf053550d74 (plain)
1
2
3
4
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
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
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
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
342
343
##
# Manages changes of attributes in a block of text

class RDoc::Markup::AttributeManager

  ##
  # The NUL character

  NULL = "\000".freeze

  #--
  # 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
  #++

  A_PROTECT = 004 # :nodoc:

  ##
  # Special mask character to prevent inline markup handling

  PROTECT_ATTR = A_PROTECT.chr # :nodoc:

  ##
  # The attributes enabled for this markup object.

  attr_reader :attributes

  ##
  # 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

  attr_reader :matching_word_pairs

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

  attr_reader :word_pair_map

  ##
  # This maps HTML tags to the corresponding attribute char

  attr_reader :html_tags

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

  attr_reader :protectable

  ##
  # And this maps _special_ sequences to a name. A special sequence is
  # something like a WikiWord

  attr_reader :special

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

  def initialize
    @html_tags = {}
    @matching_word_pairs = {}
    @protectable = %w[<]
    @special = []
    @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

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

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

  ##
  # Changes the current attribute from +current+ to +new+

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

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

  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

  ##
  # Copies +start_pos+ to +end_pos+ from the current string

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

  ##
  # Map attributes like <b>text</b>to the sequence
  # \001\002<char>\001\003<char>, where <char> is a per-attribute specific
  # character

  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

  ##
  # Converts HTML tags to RDoc attributes

  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

  ##
  # Converts special sequences to RDoc attributes

  def convert_specials str, attrs
    @special.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.special
      end
    end
  end

  ##
  # Escapes special sequences of text to prevent conversion to RDoc

  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

  ##
  # Unescapes special sequences of text

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

  ##
  # 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

  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

  ##
  # Adds a markup class with +name+ for words surrounded by HTML tag +tag+.
  # To process emphasis tags:
  #
  #   am.add_html 'em', :EM

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

  ##
  # Adds a special handler for +pattern+ with +name+.  A simple URL handler
  # would be:
  #
  #   @am.add_special(/((https?:)\S+\w)/, :HYPERLINK)

  def add_special pattern, name
    @special << [pattern, @attributes.bitmap_for(name)]
  end

  ##
  # Processes +str+ converting attributes, HTML and specials

  def flow str
    @str = str

    mask_protected_sequences

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

    convert_attrs    @str, @attrs
    convert_html     @str, @attrs
    convert_specials @str, @attrs

    unmask_protected_sequences

    split_into_flow
  end

  ##
  # Debug method that prints a string along with its attributes

  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

  ##
  # Splits the string into chunks by attribute change

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

          res << RDoc::Markup::Special.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

end