/
buffer.rb
332 lines (289 loc) · 8.79 KB
/
buffer.rb
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
# encoding: ascii-8bit
# frozen_string_literal: true
module Parser
module Source
##
# A buffer with source code. {Buffer} contains the source code itself,
# associated location information (name and first line), and takes care
# of encoding.
#
# A source buffer is immutable once populated.
#
# @!attribute [r] name
# Buffer name. If the buffer was created from a file, the name corresponds
# to relative path to the file.
# @return [String] buffer name
#
# @!attribute [r] first_line
# First line of the buffer, 1 by default.
# @return [Integer] first line
#
# @api public
#
class Buffer
attr_reader :name, :first_line
##
# @api private
#
ENCODING_RE =
/[\s#](en)?coding\s*[:=]\s*
(
# Special-case: there's a UTF8-MAC encoding.
(utf8-mac)
|
# Chew the suffix; it's there for emacs compat.
([A-Za-z0-9_-]+?)(-unix|-dos|-mac)
|
([A-Za-z0-9_-]+)
)
/x
##
# Try to recognize encoding of `string` as Ruby would, i.e. by looking for
# magic encoding comment or UTF-8 BOM. `string` can be in any encoding.
#
# @param [String] string
# @return [String, nil] encoding name, if recognized
#
def self.recognize_encoding(string)
return if string.empty?
# extract the first two lines in an efficient way
string =~ /\A(.*)\n?(.*\n)?/
first_line, second_line = $1, $2
if first_line.start_with?("\xef\xbb\xbf".freeze) # BOM
return Encoding::UTF_8
elsif first_line[0, 2] == '#!'.freeze
encoding_line = second_line
else
encoding_line = first_line
end
return nil if encoding_line.nil? || encoding_line[0] != '#'
if (result = ENCODING_RE.match(encoding_line))
Encoding.find(result[3] || result[4] || result[6])
else
nil
end
end
##
# Recognize encoding of `input` and process it so it could be lexed.
#
# * If `input` does not contain BOM or magic encoding comment, it is
# kept in the original encoding.
# * If the detected encoding is binary, `input` is kept in binary.
# * Otherwise, `input` is re-encoded into UTF-8 and returned as a
# new string.
#
# This method mutates the encoding of `input`, but not its content.
#
# @param [String] input
# @return [String]
# @raise [EncodingError]
#
def self.reencode_string(input)
original_encoding = input.encoding
detected_encoding = recognize_encoding(input.force_encoding(Encoding::BINARY))
if detected_encoding.nil?
input.force_encoding(original_encoding)
elsif detected_encoding == Encoding::BINARY
input
else
input.
force_encoding(detected_encoding).
encode(Encoding::UTF_8)
end
end
def initialize(name, first_line = 1, source: nil)
@name = name.to_s
@source = nil
@first_line = first_line
@lines = nil
@line_begins = nil
# UTF-32-reencoded source for O(1) slicing
@slice_source = nil
# Cache for fast lookup
@line_for_position = {}
@column_for_position = {}
self.source = source if source
end
##
# Populate this buffer from correspondingly named file.
#
# @example
# Parser::Source::Buffer.new('foo/bar.rb').read
#
# @return [Buffer] self
# @raise [ArgumentError] if already populated
#
def read
File.open(@name, 'rb') do |io|
self.source = io.read
end
self
end
##
# Source code contained in this buffer.
#
# @return [String] source code
# @raise [RuntimeError] if buffer is not populated yet
#
def source
if @source.nil?
raise RuntimeError, 'Cannot extract source from uninitialized Source::Buffer'
end
@source
end
##
# Populate this buffer from a string with encoding autodetection.
# `input` is mutated if not frozen.
#
# @param [String] input
# @raise [ArgumentError] if already populated
# @raise [EncodingError] if `input` includes invalid byte sequence for the encoding
# @return [String]
#
def source=(input)
input = input.dup if input.frozen?
input = self.class.reencode_string(input)
unless input.valid_encoding?
raise EncodingError, "invalid byte sequence in #{input.encoding.name}"
end
self.raw_source = input
end
##
# Populate this buffer from a string without encoding autodetection.
#
# @param [String] input
# @raise [ArgumentError] if already populated
# @return [String]
#
def raw_source=(input)
if @source
raise ArgumentError, 'Source::Buffer is immutable'
end
@source = input.gsub("\r\n".freeze, "\n".freeze).freeze
if !@source.ascii_only? &&
@source.encoding != Encoding::UTF_32LE &&
@source.encoding != Encoding::BINARY
@slice_source = @source.encode(Encoding::UTF_32LE)
end
end
def slice(range)
if @slice_source.nil?
@source[range]
else
@slice_source[range].encode(@source.encoding)
end
end
##
# Convert a character index into the source to a `[line, column]` tuple.
#
# @param [Integer] position
# @return [[Integer, Integer]] `[line, column]`
#
def decompose_position(position)
line_no, line_begin = line_for(position)
[ @first_line + line_no, position - line_begin ]
end
##
# Convert a character index into the source to a line number.
#
# @param [Integer] position
# @return [Integer] line
# @api private
#
def line_for_position(position)
@line_for_position[position] ||= begin
line_no, _ = line_for(position)
@first_line + line_no
end
end
##
# Convert a character index into the source to a column number.
#
# @param [Integer] position
# @return [Integer] column
# @api private
#
def column_for_position(position)
@column_for_position[position] ||= begin
_, line_begin = line_for(position)
position - line_begin
end
end
##
# Return an `Array` of source code lines.
#
# @return [Array<String>]
#
def source_lines
@lines ||= begin
lines = @source.lines.to_a
lines << ''.dup if @source.end_with?("\n".freeze)
lines.each do |line|
line.chomp!("\n".freeze)
line.freeze
end
lines.freeze
end
end
##
# Extract line `lineno` from source, taking `first_line` into account.
#
# @param [Integer] lineno
# @return [String]
# @raise [IndexError] if `lineno` is out of bounds
#
def source_line(lineno)
source_lines.fetch(lineno - @first_line).dup
end
##
# Extract line `lineno` as a new `Range`, taking `first_line` into account.
#
# @param [Integer] lineno
# @return [Range]
# @raise [IndexError] if `lineno` is out of bounds
#
def line_range(lineno)
index = lineno - @first_line + 1
if index <= 0 || index > line_begins.size
raise IndexError, 'Parser::Source::Buffer: range for line ' \
"#{lineno} requested, valid line numbers are #{@first_line}.." \
"#{@first_line + line_begins.size - 1}"
elsif index == line_begins.size
Range.new(self, line_begins[-index][1], @source.size)
else
Range.new(self, line_begins[-index][1], line_begins[-index - 1][1] - 1)
end
end
##
# @return [Range] A range covering the whole source
#
def source_range
@source_range ||= Range.new(self, 0, source.size)
end
##
# Number of last line in the buffer
#
# @return [Integer]
#
def last_line
line_begins.size + @first_line - 1
end
private
def line_begins
unless @line_begins
@line_begins, index = [ [ 0, 0 ] ], 0
while index = @source.index("\n".freeze, index)
index += 1
@line_begins.unshift [ @line_begins.length, index ]
end
end
@line_begins
end
def line_for(position)
line_begins.bsearch do |line, line_begin|
line_begin <= position
end
end
end
end
end