prism 0.19.0 → 0.24.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +102 -1
- data/Makefile +5 -0
- data/README.md +9 -6
- data/config.yml +236 -38
- data/docs/build_system.md +19 -2
- data/docs/cruby_compilation.md +27 -0
- data/docs/parser_translation.md +34 -0
- data/docs/parsing_rules.md +19 -0
- data/docs/releasing.md +84 -16
- data/docs/ruby_api.md +1 -1
- data/docs/ruby_parser_translation.md +19 -0
- data/docs/serialization.md +19 -5
- data/ext/prism/api_node.c +1989 -1525
- data/ext/prism/extension.c +130 -30
- data/ext/prism/extension.h +2 -2
- data/include/prism/ast.h +1700 -505
- data/include/prism/defines.h +8 -0
- data/include/prism/diagnostic.h +49 -7
- data/include/prism/encoding.h +17 -0
- data/include/prism/options.h +40 -14
- data/include/prism/parser.h +34 -18
- data/include/prism/util/pm_buffer.h +9 -0
- data/include/prism/util/pm_constant_pool.h +18 -0
- data/include/prism/util/pm_newline_list.h +4 -14
- data/include/prism/util/pm_strpbrk.h +4 -1
- data/include/prism/version.h +2 -2
- data/include/prism.h +19 -2
- data/lib/prism/debug.rb +11 -5
- data/lib/prism/desugar_compiler.rb +225 -80
- data/lib/prism/dot_visitor.rb +36 -14
- data/lib/prism/dsl.rb +302 -299
- data/lib/prism/ffi.rb +107 -76
- data/lib/prism/lex_compat.rb +17 -1
- data/lib/prism/node.rb +4580 -2607
- data/lib/prism/node_ext.rb +27 -4
- data/lib/prism/parse_result.rb +75 -29
- data/lib/prism/serialize.rb +633 -305
- data/lib/prism/translation/parser/compiler.rb +1838 -0
- data/lib/prism/translation/parser/lexer.rb +335 -0
- data/lib/prism/translation/parser/rubocop.rb +45 -0
- data/lib/prism/translation/parser.rb +190 -0
- data/lib/prism/translation/parser33.rb +12 -0
- data/lib/prism/translation/parser34.rb +12 -0
- data/lib/prism/translation/ripper.rb +696 -0
- data/lib/prism/translation/ruby_parser.rb +1521 -0
- data/lib/prism/translation.rb +11 -0
- data/lib/prism.rb +1 -1
- data/prism.gemspec +18 -7
- data/rbi/prism.rbi +150 -88
- data/rbi/prism_static.rbi +15 -3
- data/sig/prism.rbs +996 -961
- data/sig/prism_static.rbs +123 -46
- data/src/diagnostic.c +264 -219
- data/src/encoding.c +21 -26
- data/src/node.c +2 -6
- data/src/options.c +29 -5
- data/src/prettyprint.c +176 -44
- data/src/prism.c +1499 -564
- data/src/serialize.c +35 -21
- data/src/token_type.c +353 -4
- data/src/util/pm_buffer.c +11 -0
- data/src/util/pm_constant_pool.c +37 -11
- data/src/util/pm_newline_list.c +6 -15
- data/src/util/pm_string.c +0 -7
- data/src/util/pm_strpbrk.c +122 -14
- metadata +16 -5
- data/docs/building.md +0 -29
- data/lib/prism/ripper_compat.rb +0 -207
data/lib/prism/node_ext.rb
CHANGED
@@ -81,7 +81,7 @@ module Prism
|
|
81
81
|
class RationalNode < Node
|
82
82
|
# Returns the value of the node as a Ruby Rational.
|
83
83
|
def value
|
84
|
-
Rational(slice.chomp("r"))
|
84
|
+
Rational(numeric.is_a?(IntegerNode) ? numeric.value : slice.chomp("r"))
|
85
85
|
end
|
86
86
|
end
|
87
87
|
|
@@ -94,7 +94,7 @@ module Prism
|
|
94
94
|
|
95
95
|
# Returns the full name of this constant. For example: "Foo"
|
96
96
|
def full_name
|
97
|
-
name.
|
97
|
+
name.to_s
|
98
98
|
end
|
99
99
|
end
|
100
100
|
|
@@ -118,7 +118,7 @@ module Prism
|
|
118
118
|
current = current.parent
|
119
119
|
end
|
120
120
|
|
121
|
-
|
121
|
+
if !current.is_a?(ConstantReadNode) && !current.nil?
|
122
122
|
raise DynamicPartsInConstantPathError, "Constant path contains dynamic parts. Cannot compute full name"
|
123
123
|
end
|
124
124
|
|
@@ -135,7 +135,17 @@ module Prism
|
|
135
135
|
# Returns the list of parts for the full name of this constant path.
|
136
136
|
# For example: [:Foo, :Bar]
|
137
137
|
def full_name_parts
|
138
|
-
|
138
|
+
parts = case parent
|
139
|
+
when ConstantPathNode, ConstantReadNode
|
140
|
+
parent.full_name_parts
|
141
|
+
when nil
|
142
|
+
[:""]
|
143
|
+
else
|
144
|
+
raise ConstantPathNode::DynamicPartsInConstantPathError,
|
145
|
+
"Constant path target contains dynamic parts. Cannot compute full name"
|
146
|
+
end
|
147
|
+
|
148
|
+
parts.push(child.name)
|
139
149
|
end
|
140
150
|
|
141
151
|
# Returns the full name of this constant path. For example: "Foo::Bar"
|
@@ -144,6 +154,19 @@ module Prism
|
|
144
154
|
end
|
145
155
|
end
|
146
156
|
|
157
|
+
class ConstantTargetNode < Node
|
158
|
+
# Returns the list of parts for the full name of this constant.
|
159
|
+
# For example: [:Foo]
|
160
|
+
def full_name_parts
|
161
|
+
[name]
|
162
|
+
end
|
163
|
+
|
164
|
+
# Returns the full name of this constant. For example: "Foo"
|
165
|
+
def full_name
|
166
|
+
name.to_s
|
167
|
+
end
|
168
|
+
end
|
169
|
+
|
147
170
|
class ParametersNode < Node
|
148
171
|
# Mirrors the Method#parameters method.
|
149
172
|
def signature
|
data/lib/prism/parse_result.rb
CHANGED
@@ -9,18 +9,16 @@ module Prism
|
|
9
9
|
attr_reader :source
|
10
10
|
|
11
11
|
# The line number where this source starts.
|
12
|
-
|
12
|
+
attr_reader :start_line
|
13
13
|
|
14
14
|
# The list of newline byte offsets in the source code.
|
15
15
|
attr_reader :offsets
|
16
16
|
|
17
|
-
# Create a new source object with the given source code
|
18
|
-
|
19
|
-
# the source code.
|
20
|
-
def initialize(source, start_line = 1, offsets = compute_offsets(source))
|
17
|
+
# Create a new source object with the given source code.
|
18
|
+
def initialize(source, start_line = 1, offsets = [])
|
21
19
|
@source = source
|
22
|
-
@start_line = start_line
|
23
|
-
@offsets = offsets
|
20
|
+
@start_line = start_line # set after parsing is done
|
21
|
+
@offsets = offsets # set after parsing is done
|
24
22
|
end
|
25
23
|
|
26
24
|
# Perform a byteslice on the source code using the given byte offset and
|
@@ -56,6 +54,23 @@ module Prism
|
|
56
54
|
character_offset(byte_offset) - character_offset(line_start(byte_offset))
|
57
55
|
end
|
58
56
|
|
57
|
+
# Returns the offset from the start of the file for the given byte offset
|
58
|
+
# counting in code units for the given encoding.
|
59
|
+
#
|
60
|
+
# This method is tested with UTF-8, UTF-16, and UTF-32. If there is the
|
61
|
+
# concept of code units that differs from the number of characters in other
|
62
|
+
# encodings, it is not captured here.
|
63
|
+
def code_units_offset(byte_offset, encoding)
|
64
|
+
byteslice = source.byteslice(0, byte_offset).encode(encoding)
|
65
|
+
(encoding == Encoding::UTF_16LE || encoding == Encoding::UTF_16BE) ? (byteslice.bytesize / 2) : byteslice.length
|
66
|
+
end
|
67
|
+
|
68
|
+
# Returns the column number in code units for the given encoding for the
|
69
|
+
# given byte offset.
|
70
|
+
def code_units_column(byte_offset, encoding)
|
71
|
+
code_units_offset(byte_offset, encoding) - code_units_offset(line_start(byte_offset), encoding)
|
72
|
+
end
|
73
|
+
|
59
74
|
private
|
60
75
|
|
61
76
|
# Binary search through the offsets to find the line number for the given
|
@@ -77,21 +92,14 @@ module Prism
|
|
77
92
|
|
78
93
|
left - 1
|
79
94
|
end
|
80
|
-
|
81
|
-
# Find all of the newlines in the source code and return their byte offsets
|
82
|
-
# from the start of the string an array.
|
83
|
-
def compute_offsets(code)
|
84
|
-
offsets = [0]
|
85
|
-
code.b.scan("\n") { offsets << $~.end(0) }
|
86
|
-
offsets
|
87
|
-
end
|
88
95
|
end
|
89
96
|
|
90
97
|
# This represents a location in the source.
|
91
98
|
class Location
|
92
99
|
# A Source object that is used to determine more information from the given
|
93
100
|
# offset and length.
|
94
|
-
|
101
|
+
attr_reader :source
|
102
|
+
protected :source
|
95
103
|
|
96
104
|
# The byte offset from the beginning of the source where this location
|
97
105
|
# starts.
|
@@ -137,6 +145,11 @@ module Prism
|
|
137
145
|
source.character_offset(start_offset)
|
138
146
|
end
|
139
147
|
|
148
|
+
# The offset from the start of the file in code units of the given encoding.
|
149
|
+
def start_code_units_offset(encoding = Encoding::UTF_16LE)
|
150
|
+
source.code_units_offset(start_offset, encoding)
|
151
|
+
end
|
152
|
+
|
140
153
|
# The byte offset from the beginning of the source where this location ends.
|
141
154
|
def end_offset
|
142
155
|
start_offset + length
|
@@ -148,6 +161,11 @@ module Prism
|
|
148
161
|
source.character_offset(end_offset)
|
149
162
|
end
|
150
163
|
|
164
|
+
# The offset from the start of the file in code units of the given encoding.
|
165
|
+
def end_code_units_offset(encoding = Encoding::UTF_16LE)
|
166
|
+
source.code_units_offset(end_offset, encoding)
|
167
|
+
end
|
168
|
+
|
151
169
|
# The line number where this location starts.
|
152
170
|
def start_line
|
153
171
|
source.line(start_offset)
|
@@ -176,6 +194,12 @@ module Prism
|
|
176
194
|
source.character_column(start_offset)
|
177
195
|
end
|
178
196
|
|
197
|
+
# The column number in code units of the given encoding where this location
|
198
|
+
# starts from the start of the line.
|
199
|
+
def start_code_units_column(encoding = Encoding::UTF_16LE)
|
200
|
+
source.code_units_column(start_offset, encoding)
|
201
|
+
end
|
202
|
+
|
179
203
|
# The column number in bytes where this location ends from the start of the
|
180
204
|
# line.
|
181
205
|
def end_column
|
@@ -188,6 +212,12 @@ module Prism
|
|
188
212
|
source.character_column(end_offset)
|
189
213
|
end
|
190
214
|
|
215
|
+
# The column number in code units of the given encoding where this location
|
216
|
+
# ends from the start of the line.
|
217
|
+
def end_code_units_column(encoding = Encoding::UTF_16LE)
|
218
|
+
source.code_units_column(end_offset, encoding)
|
219
|
+
end
|
220
|
+
|
191
221
|
# Implement the hash pattern matching interface for Location.
|
192
222
|
def deconstruct_keys(keys)
|
193
223
|
{ start_offset: start_offset, end_offset: end_offset }
|
@@ -312,20 +342,24 @@ module Prism
|
|
312
342
|
# A Location object representing the location of this error in the source.
|
313
343
|
attr_reader :location
|
314
344
|
|
345
|
+
# The level of this error.
|
346
|
+
attr_reader :level
|
347
|
+
|
315
348
|
# Create a new error object with the given message and location.
|
316
|
-
def initialize(message, location)
|
349
|
+
def initialize(message, location, level)
|
317
350
|
@message = message
|
318
351
|
@location = location
|
352
|
+
@level = level
|
319
353
|
end
|
320
354
|
|
321
355
|
# Implement the hash pattern matching interface for ParseError.
|
322
356
|
def deconstruct_keys(keys)
|
323
|
-
{ message: message, location: location }
|
357
|
+
{ message: message, location: location, level: level }
|
324
358
|
end
|
325
359
|
|
326
360
|
# Returns a string representation of this error.
|
327
361
|
def inspect
|
328
|
-
"#<Prism::ParseError @message=#{@message.inspect} @location=#{@location.inspect}>"
|
362
|
+
"#<Prism::ParseError @message=#{@message.inspect} @location=#{@location.inspect} @level=#{@level.inspect}>"
|
329
363
|
end
|
330
364
|
end
|
331
365
|
|
@@ -337,20 +371,24 @@ module Prism
|
|
337
371
|
# A Location object representing the location of this warning in the source.
|
338
372
|
attr_reader :location
|
339
373
|
|
374
|
+
# The level of this warning.
|
375
|
+
attr_reader :level
|
376
|
+
|
340
377
|
# Create a new warning object with the given message and location.
|
341
|
-
def initialize(message, location)
|
378
|
+
def initialize(message, location, level)
|
342
379
|
@message = message
|
343
380
|
@location = location
|
381
|
+
@level = level
|
344
382
|
end
|
345
383
|
|
346
384
|
# Implement the hash pattern matching interface for ParseWarning.
|
347
385
|
def deconstruct_keys(keys)
|
348
|
-
{ message: message, location: location }
|
386
|
+
{ message: message, location: location, level: level }
|
349
387
|
end
|
350
388
|
|
351
389
|
# Returns a string representation of this warning.
|
352
390
|
def inspect
|
353
|
-
"#<Prism::ParseWarning @message=#{@message.inspect} @location=#{@location.inspect}>"
|
391
|
+
"#<Prism::ParseWarning @message=#{@message.inspect} @location=#{@location.inspect} @level=#{@level.inspect}>"
|
354
392
|
end
|
355
393
|
end
|
356
394
|
|
@@ -369,9 +407,9 @@ module Prism
|
|
369
407
|
# The list of magic comments that were encountered during parsing.
|
370
408
|
attr_reader :magic_comments
|
371
409
|
|
372
|
-
# An optional location that represents the location of the
|
373
|
-
#
|
374
|
-
# file being parsed is the main file being executed.
|
410
|
+
# An optional location that represents the location of the __END__ marker
|
411
|
+
# and the rest of the content of the file. This content is loaded into the
|
412
|
+
# DATA constant when the file being parsed is the main file being executed.
|
375
413
|
attr_reader :data_loc
|
376
414
|
|
377
415
|
# The list of errors that were generated during parsing.
|
@@ -414,17 +452,19 @@ module Prism
|
|
414
452
|
|
415
453
|
# This represents a token from the Ruby source.
|
416
454
|
class Token
|
455
|
+
# The Source object that represents the source this token came from.
|
456
|
+
attr_reader :source
|
457
|
+
private :source
|
458
|
+
|
417
459
|
# The type of token that this token is.
|
418
460
|
attr_reader :type
|
419
461
|
|
420
462
|
# A byteslice of the source that this token represents.
|
421
463
|
attr_reader :value
|
422
464
|
|
423
|
-
# A Location object representing the location of this token in the source.
|
424
|
-
attr_reader :location
|
425
|
-
|
426
465
|
# Create a new token object with the given type, value, and location.
|
427
|
-
def initialize(type, value, location)
|
466
|
+
def initialize(source, type, value, location)
|
467
|
+
@source = source
|
428
468
|
@type = type
|
429
469
|
@value = value
|
430
470
|
@location = location
|
@@ -435,6 +475,12 @@ module Prism
|
|
435
475
|
{ type: type, value: value, location: location }
|
436
476
|
end
|
437
477
|
|
478
|
+
# A Location object representing the location of this token in the source.
|
479
|
+
def location
|
480
|
+
return @location if @location.is_a?(Location)
|
481
|
+
@location = Location.new(source, @location >> 32, @location & 0xFFFFFFFF)
|
482
|
+
end
|
483
|
+
|
438
484
|
# Implement the pretty print interface for Token.
|
439
485
|
def pretty_print(q)
|
440
486
|
q.group do
|