irb 1.7.1 → 1.13.2
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.document +1 -1
- data/Gemfile +10 -1
- data/README.md +265 -20
- data/Rakefile +13 -10
- data/doc/irb/irb.rd.ja +1 -3
- data/irb.gemspec +2 -1
- data/lib/irb/cmd/nop.rb +3 -52
- data/lib/irb/color.rb +4 -2
- data/lib/irb/command/backtrace.rb +17 -0
- data/lib/irb/command/base.rb +62 -0
- data/lib/irb/command/break.rb +17 -0
- data/lib/irb/command/catch.rb +17 -0
- data/lib/irb/command/chws.rb +40 -0
- data/lib/irb/command/context.rb +16 -0
- data/lib/irb/{cmd → command}/continue.rb +3 -3
- data/lib/irb/command/debug.rb +71 -0
- data/lib/irb/{cmd → command}/delete.rb +3 -3
- data/lib/irb/command/disable_irb.rb +19 -0
- data/lib/irb/command/edit.rb +63 -0
- data/lib/irb/command/exit.rb +18 -0
- data/lib/irb/{cmd → command}/finish.rb +3 -3
- data/lib/irb/command/force_exit.rb +18 -0
- data/lib/irb/command/help.rb +83 -0
- data/lib/irb/command/history.rb +45 -0
- data/lib/irb/command/info.rb +17 -0
- data/lib/irb/command/internal_helpers.rb +27 -0
- data/lib/irb/{cmd → command}/irb_info.rb +7 -7
- data/lib/irb/{cmd → command}/load.rb +23 -8
- data/lib/irb/{cmd → command}/ls.rb +42 -19
- data/lib/irb/{cmd → command}/measure.rb +18 -17
- data/lib/irb/{cmd → command}/next.rb +3 -3
- data/lib/irb/command/pushws.rb +65 -0
- data/lib/irb/command/show_doc.rb +51 -0
- data/lib/irb/command/show_source.rb +74 -0
- data/lib/irb/{cmd → command}/step.rb +3 -3
- data/lib/irb/command/subirb.rb +123 -0
- data/lib/irb/{cmd → command}/whereami.rb +3 -5
- data/lib/irb/command.rb +23 -0
- data/lib/irb/completion.rb +133 -102
- data/lib/irb/context.rb +182 -66
- data/lib/irb/debug/ui.rb +103 -0
- data/lib/irb/{cmd/debug.rb → debug.rb} +53 -59
- data/lib/irb/default_commands.rb +265 -0
- data/lib/irb/easter-egg.rb +16 -6
- data/lib/irb/ext/change-ws.rb +6 -8
- data/lib/irb/ext/{history.rb → eval_history.rb} +7 -7
- data/lib/irb/ext/loader.rb +4 -4
- data/lib/irb/ext/multi-irb.rb +5 -5
- data/lib/irb/ext/tracer.rb +12 -51
- data/lib/irb/ext/use-loader.rb +6 -8
- data/lib/irb/ext/workspaces.rb +10 -34
- data/lib/irb/frame.rb +1 -1
- data/lib/irb/help.rb +3 -3
- data/lib/irb/helper_method/base.rb +16 -0
- data/lib/irb/helper_method/conf.rb +11 -0
- data/lib/irb/helper_method.rb +29 -0
- data/lib/irb/{ext/save-history.rb → history.rb} +20 -58
- data/lib/irb/init.rb +154 -58
- data/lib/irb/input-method.rb +238 -203
- data/lib/irb/inspector.rb +3 -3
- data/lib/irb/lc/error.rb +1 -11
- data/lib/irb/lc/help-message +4 -0
- data/lib/irb/lc/ja/error.rb +1 -11
- data/lib/irb/lc/ja/help-message +13 -0
- data/lib/irb/locale.rb +2 -2
- data/lib/irb/nesting_parser.rb +13 -3
- data/lib/irb/notifier.rb +1 -1
- data/lib/irb/output-method.rb +2 -8
- data/lib/irb/pager.rb +91 -0
- data/lib/irb/ruby-lex.rb +391 -471
- data/lib/irb/ruby_logo.aa +43 -0
- data/lib/irb/source_finder.rb +139 -0
- data/lib/irb/statement.rb +80 -0
- data/lib/irb/version.rb +3 -3
- data/lib/irb/workspace.rb +24 -4
- data/lib/irb/ws-for-case-2.rb +1 -1
- data/lib/irb/xmp.rb +3 -3
- data/lib/irb.rb +1232 -604
- data/man/irb.1 +7 -0
- metadata +60 -32
- data/lib/irb/cmd/backtrace.rb +0 -21
- data/lib/irb/cmd/break.rb +0 -21
- data/lib/irb/cmd/catch.rb +0 -21
- data/lib/irb/cmd/chws.rb +0 -36
- data/lib/irb/cmd/edit.rb +0 -61
- data/lib/irb/cmd/help.rb +0 -23
- data/lib/irb/cmd/info.rb +0 -21
- data/lib/irb/cmd/pushws.rb +0 -45
- data/lib/irb/cmd/show_cmds.rb +0 -39
- data/lib/irb/cmd/show_doc.rb +0 -48
- data/lib/irb/cmd/show_source.rb +0 -113
- data/lib/irb/cmd/subirb.rb +0 -66
- data/lib/irb/extend-command.rb +0 -356
- data/lib/irb/src_encoding.rb +0 -7
data/lib/irb/ruby-lex.rb
CHANGED
@@ -1,4 +1,4 @@
|
|
1
|
-
# frozen_string_literal:
|
1
|
+
# frozen_string_literal: true
|
2
2
|
#
|
3
3
|
# irb/ruby-lex.rb - ruby lexcal analyzer
|
4
4
|
# by Keiju ISHITSUKA(keiju@ruby-lang.org)
|
@@ -8,547 +8,467 @@ require "ripper"
|
|
8
8
|
require "jruby" if RUBY_ENGINE == "jruby"
|
9
9
|
require_relative "nesting_parser"
|
10
10
|
|
11
|
-
|
12
|
-
|
13
|
-
|
14
|
-
|
15
|
-
|
16
|
-
|
11
|
+
module IRB
|
12
|
+
# :stopdoc:
|
13
|
+
class RubyLex
|
14
|
+
ASSIGNMENT_NODE_TYPES = [
|
15
|
+
# Local, instance, global, class, constant, instance, and index assignment:
|
16
|
+
# "foo = bar",
|
17
|
+
# "@foo = bar",
|
18
|
+
# "$foo = bar",
|
19
|
+
# "@@foo = bar",
|
20
|
+
# "::Foo = bar",
|
21
|
+
# "a::Foo = bar",
|
22
|
+
# "Foo = bar"
|
23
|
+
# "foo.bar = 1"
|
24
|
+
# "foo[1] = bar"
|
25
|
+
:assign,
|
26
|
+
|
27
|
+
# Operation assignment:
|
28
|
+
# "foo += bar"
|
29
|
+
# "foo -= bar"
|
30
|
+
# "foo ||= bar"
|
31
|
+
# "foo &&= bar"
|
32
|
+
:opassign,
|
33
|
+
|
34
|
+
# Multiple assignment:
|
35
|
+
# "foo, bar = 1, 2
|
36
|
+
:massign,
|
37
|
+
]
|
38
|
+
|
39
|
+
class TerminateLineInput < StandardError
|
40
|
+
def initialize
|
41
|
+
super("Terminate Line Input")
|
42
|
+
end
|
17
43
|
end
|
18
|
-
end
|
19
|
-
|
20
|
-
def initialize(context)
|
21
|
-
@context = context
|
22
|
-
@line_no = 1
|
23
|
-
@prompt = nil
|
24
|
-
end
|
25
44
|
|
26
|
-
|
27
|
-
|
28
|
-
|
29
|
-
|
30
|
-
|
31
|
-
|
32
|
-
|
33
|
-
|
34
|
-
|
35
|
-
|
36
|
-
|
37
|
-
|
38
|
-
|
45
|
+
def self.compile_with_errors_suppressed(code, line_no: 1)
|
46
|
+
begin
|
47
|
+
result = yield code, line_no
|
48
|
+
rescue ArgumentError
|
49
|
+
# Ruby can issue an error for the code if there is an
|
50
|
+
# incomplete magic comment for encoding in it. Force an
|
51
|
+
# expression with a new line before the code in this
|
52
|
+
# case to prevent magic comment handling. To make sure
|
53
|
+
# line numbers in the lexed code remain the same,
|
54
|
+
# decrease the line number by one.
|
55
|
+
code = ";\n#{code}"
|
56
|
+
line_no -= 1
|
57
|
+
result = yield code, line_no
|
58
|
+
end
|
59
|
+
result
|
39
60
|
end
|
40
|
-
result
|
41
|
-
end
|
42
|
-
|
43
|
-
def single_line_command?(code)
|
44
|
-
command = code.split(/\s/, 2).first
|
45
|
-
@context.symbol_alias?(command) || @context.transform_args?(command)
|
46
|
-
end
|
47
61
|
|
48
|
-
|
49
|
-
|
50
|
-
|
51
|
-
|
52
|
-
|
53
|
-
|
54
|
-
|
55
|
-
|
56
|
-
|
57
|
-
|
58
|
-
|
59
|
-
|
60
|
-
Reline.delete_text
|
61
|
-
rest.bytes.reverse_each do |c|
|
62
|
-
Reline.ungetc(c)
|
63
|
-
end
|
64
|
-
true
|
65
|
-
else
|
66
|
-
false
|
67
|
-
end
|
68
|
-
else
|
69
|
-
# Accept any single-line input for symbol aliases or commands that transform args
|
70
|
-
next true if single_line_command?(code)
|
62
|
+
ERROR_TOKENS = [
|
63
|
+
:on_parse_error,
|
64
|
+
:compile_error,
|
65
|
+
:on_assign_error,
|
66
|
+
:on_alias_error,
|
67
|
+
:on_class_name_error,
|
68
|
+
:on_param_error
|
69
|
+
]
|
70
|
+
|
71
|
+
def self.generate_local_variables_assign_code(local_variables)
|
72
|
+
"#{local_variables.join('=')}=nil;" unless local_variables.empty?
|
73
|
+
end
|
71
74
|
|
72
|
-
|
73
|
-
|
74
|
-
|
75
|
+
# Some part of the code is not included in Ripper's token.
|
76
|
+
# Example: DATA part, token after heredoc_beg when heredoc has unclosed embexpr.
|
77
|
+
# With interpolated tokens, tokens.map(&:tok).join will be equal to code.
|
78
|
+
def self.interpolate_ripper_ignored_tokens(code, tokens)
|
79
|
+
line_positions = [0]
|
80
|
+
code.lines.each do |line|
|
81
|
+
line_positions << line_positions.last + line.bytesize
|
75
82
|
end
|
76
|
-
|
77
|
-
|
78
|
-
|
79
|
-
|
80
|
-
|
81
|
-
|
82
|
-
|
83
|
-
|
84
|
-
|
85
|
-
|
86
|
-
|
87
|
-
end
|
88
|
-
continue = should_continue?(tokens_until_line)
|
89
|
-
prompt(next_opens, continue, line_num_offset)
|
83
|
+
prev_byte_pos = 0
|
84
|
+
interpolated = []
|
85
|
+
prev_line = 1
|
86
|
+
tokens.each do |t|
|
87
|
+
line, col = t.pos
|
88
|
+
byte_pos = line_positions[line - 1] + col
|
89
|
+
if prev_byte_pos < byte_pos
|
90
|
+
tok = code.byteslice(prev_byte_pos...byte_pos)
|
91
|
+
pos = [prev_line, prev_byte_pos - line_positions[prev_line - 1]]
|
92
|
+
interpolated << Ripper::Lexer::Elem.new(pos, :on_ignored_by_ripper, tok, 0)
|
93
|
+
prev_line += tok.count("\n")
|
90
94
|
end
|
95
|
+
interpolated << t
|
96
|
+
prev_byte_pos = byte_pos + t.tok.bytesize
|
97
|
+
prev_line += t.tok.count("\n")
|
91
98
|
end
|
92
|
-
|
93
|
-
|
94
|
-
if @io.respond_to?(:auto_indent) and @context.auto_indent_mode
|
95
|
-
@io.auto_indent do |lines, line_index, byte_pointer, is_newline|
|
96
|
-
next nil if lines == [nil] # Workaround for exit IRB with CTRL+d
|
97
|
-
next nil if !is_newline && lines[line_index]&.byteslice(0, byte_pointer)&.match?(/\A\s*\z/)
|
98
|
-
|
99
|
-
code = lines[0..line_index].map { |l| "#{l}\n" }.join
|
100
|
-
tokens = self.class.ripper_lex_without_warning(code, context: @context)
|
101
|
-
process_indent_level(tokens, lines, line_index, is_newline)
|
102
|
-
end
|
103
|
-
end
|
104
|
-
end
|
105
|
-
|
106
|
-
def set_prompt(&block)
|
107
|
-
@prompt = block
|
108
|
-
end
|
109
|
-
|
110
|
-
ERROR_TOKENS = [
|
111
|
-
:on_parse_error,
|
112
|
-
:compile_error,
|
113
|
-
:on_assign_error,
|
114
|
-
:on_alias_error,
|
115
|
-
:on_class_name_error,
|
116
|
-
:on_param_error
|
117
|
-
]
|
118
|
-
|
119
|
-
def self.generate_local_variables_assign_code(local_variables)
|
120
|
-
"#{local_variables.join('=')}=nil;" unless local_variables.empty?
|
121
|
-
end
|
122
|
-
|
123
|
-
# Some part of the code is not included in Ripper's token.
|
124
|
-
# Example: DATA part, token after heredoc_beg when heredoc has unclosed embexpr.
|
125
|
-
# With interpolated tokens, tokens.map(&:tok).join will be equal to code.
|
126
|
-
def self.interpolate_ripper_ignored_tokens(code, tokens)
|
127
|
-
line_positions = [0]
|
128
|
-
code.lines.each do |line|
|
129
|
-
line_positions << line_positions.last + line.bytesize
|
130
|
-
end
|
131
|
-
prev_byte_pos = 0
|
132
|
-
interpolated = []
|
133
|
-
prev_line = 1
|
134
|
-
tokens.each do |t|
|
135
|
-
line, col = t.pos
|
136
|
-
byte_pos = line_positions[line - 1] + col
|
137
|
-
if prev_byte_pos < byte_pos
|
138
|
-
tok = code.byteslice(prev_byte_pos...byte_pos)
|
99
|
+
if prev_byte_pos < code.bytesize
|
100
|
+
tok = code.byteslice(prev_byte_pos..)
|
139
101
|
pos = [prev_line, prev_byte_pos - line_positions[prev_line - 1]]
|
140
102
|
interpolated << Ripper::Lexer::Elem.new(pos, :on_ignored_by_ripper, tok, 0)
|
141
|
-
prev_line += tok.count("\n")
|
142
103
|
end
|
143
|
-
interpolated
|
144
|
-
prev_byte_pos = byte_pos + t.tok.bytesize
|
145
|
-
prev_line += t.tok.count("\n")
|
104
|
+
interpolated
|
146
105
|
end
|
147
|
-
if prev_byte_pos < code.bytesize
|
148
|
-
tok = code.byteslice(prev_byte_pos..)
|
149
|
-
pos = [prev_line, prev_byte_pos - line_positions[prev_line - 1]]
|
150
|
-
interpolated << Ripper::Lexer::Elem.new(pos, :on_ignored_by_ripper, tok, 0)
|
151
|
-
end
|
152
|
-
interpolated
|
153
|
-
end
|
154
106
|
|
155
|
-
|
156
|
-
|
157
|
-
|
158
|
-
|
159
|
-
|
160
|
-
|
161
|
-
|
162
|
-
|
163
|
-
|
164
|
-
|
107
|
+
def self.ripper_lex_without_warning(code, local_variables: [])
|
108
|
+
verbose, $VERBOSE = $VERBOSE, nil
|
109
|
+
lvars_code = generate_local_variables_assign_code(local_variables)
|
110
|
+
original_code = code
|
111
|
+
if lvars_code
|
112
|
+
code = "#{lvars_code}\n#{code}"
|
113
|
+
line_no = 0
|
114
|
+
else
|
115
|
+
line_no = 1
|
116
|
+
end
|
165
117
|
|
166
|
-
|
167
|
-
|
168
|
-
|
169
|
-
|
170
|
-
|
171
|
-
|
172
|
-
|
173
|
-
|
174
|
-
|
175
|
-
|
176
|
-
|
118
|
+
compile_with_errors_suppressed(code, line_no: line_no) do |inner_code, line_no|
|
119
|
+
lexer = Ripper::Lexer.new(inner_code, '-', line_no)
|
120
|
+
tokens = []
|
121
|
+
lexer.scan.each do |t|
|
122
|
+
next if t.pos.first == 0
|
123
|
+
prev_tk = tokens.last
|
124
|
+
position_overlapped = prev_tk && t.pos[0] == prev_tk.pos[0] && t.pos[1] < prev_tk.pos[1] + prev_tk.tok.bytesize
|
125
|
+
if position_overlapped
|
126
|
+
tokens[-1] = t if ERROR_TOKENS.include?(prev_tk.event) && !ERROR_TOKENS.include?(t.event)
|
127
|
+
else
|
128
|
+
tokens << t
|
129
|
+
end
|
177
130
|
end
|
131
|
+
interpolate_ripper_ignored_tokens(original_code, tokens)
|
178
132
|
end
|
179
|
-
|
133
|
+
ensure
|
134
|
+
$VERBOSE = verbose
|
180
135
|
end
|
181
|
-
ensure
|
182
|
-
$VERBOSE = verbose
|
183
|
-
end
|
184
|
-
|
185
|
-
def prompt(opens, continue, line_num_offset)
|
186
|
-
ltype = ltype_from_open_tokens(opens)
|
187
|
-
indent_level = calc_indent_level(opens)
|
188
|
-
@prompt&.call(ltype, indent_level, opens.any? || continue, @line_no + line_num_offset)
|
189
|
-
end
|
190
|
-
|
191
|
-
def check_code_state(code)
|
192
|
-
check_target_code = code.gsub(/\s*\z/, '').concat("\n")
|
193
|
-
tokens = self.class.ripper_lex_without_warning(check_target_code, context: @context)
|
194
|
-
opens = IRB::NestingParser.open_tokens(tokens)
|
195
|
-
[tokens, opens, code_terminated?(code, tokens, opens)]
|
196
|
-
end
|
197
136
|
|
198
|
-
|
199
|
-
|
200
|
-
|
201
|
-
|
202
|
-
when :recoverable_error
|
203
|
-
false
|
204
|
-
when :other_error
|
205
|
-
opens.empty? && !should_continue?(tokens)
|
206
|
-
when :valid
|
207
|
-
!should_continue?(tokens)
|
137
|
+
def check_code_state(code, local_variables:)
|
138
|
+
tokens = self.class.ripper_lex_without_warning(code, local_variables: local_variables)
|
139
|
+
opens = NestingParser.open_tokens(tokens)
|
140
|
+
[tokens, opens, code_terminated?(code, tokens, opens, local_variables: local_variables)]
|
208
141
|
end
|
209
|
-
end
|
210
|
-
|
211
|
-
def save_prompt_to_context_io(opens, continue, line_num_offset)
|
212
|
-
# Implicitly saves prompt string to `@context.io.prompt`. This will be used in the next `@input.call`.
|
213
|
-
prompt(opens, continue, line_num_offset)
|
214
|
-
end
|
215
142
|
|
216
|
-
|
217
|
-
|
218
|
-
|
219
|
-
|
220
|
-
|
221
|
-
|
222
|
-
|
223
|
-
|
224
|
-
|
225
|
-
|
226
|
-
line = @input.call
|
227
|
-
unless line
|
228
|
-
return code.empty? ? nil : code
|
143
|
+
def code_terminated?(code, tokens, opens, local_variables:)
|
144
|
+
case check_code_syntax(code, local_variables: local_variables)
|
145
|
+
when :unrecoverable_error
|
146
|
+
true
|
147
|
+
when :recoverable_error
|
148
|
+
false
|
149
|
+
when :other_error
|
150
|
+
opens.empty? && !should_continue?(tokens)
|
151
|
+
when :valid
|
152
|
+
!should_continue?(tokens)
|
229
153
|
end
|
230
|
-
|
231
|
-
code << line
|
232
|
-
# Accept any single-line input for symbol aliases or commands that transform args
|
233
|
-
return code if single_line_command?(code)
|
234
|
-
|
235
|
-
tokens, opens, terminated = check_code_state(code)
|
236
|
-
return code if terminated
|
237
|
-
|
238
|
-
line_offset += 1
|
239
|
-
continue = should_continue?(tokens)
|
240
|
-
save_prompt_to_context_io(opens, continue, line_offset)
|
241
154
|
end
|
242
|
-
end
|
243
155
|
|
244
|
-
|
245
|
-
|
246
|
-
|
247
|
-
break unless code
|
156
|
+
def assignment_expression?(code, local_variables:)
|
157
|
+
# Try to parse the code and check if the last of possibly multiple
|
158
|
+
# expressions is an assignment type.
|
248
159
|
|
249
|
-
|
250
|
-
|
251
|
-
|
252
|
-
|
253
|
-
|
254
|
-
|
160
|
+
# If the expression is invalid, Ripper.sexp should return nil which will
|
161
|
+
# result in false being returned. Any valid expression should return an
|
162
|
+
# s-expression where the second element of the top level array is an
|
163
|
+
# array of parsed expressions. The first element of each expression is the
|
164
|
+
# expression's type.
|
165
|
+
verbose, $VERBOSE = $VERBOSE, nil
|
166
|
+
code = "#{RubyLex.generate_local_variables_assign_code(local_variables) || 'nil;'}\n#{code}"
|
167
|
+
# Get the last node_type of the line. drop(1) is to ignore the local_variables_assign_code part.
|
168
|
+
node_type = Ripper.sexp(code)&.dig(1)&.drop(1)&.dig(-1, 0)
|
169
|
+
ASSIGNMENT_NODE_TYPES.include?(node_type)
|
170
|
+
ensure
|
171
|
+
$VERBOSE = verbose
|
255
172
|
end
|
256
|
-
end
|
257
173
|
|
258
|
-
|
259
|
-
|
260
|
-
|
261
|
-
|
262
|
-
|
263
|
-
|
264
|
-
|
265
|
-
|
266
|
-
|
267
|
-
|
268
|
-
|
269
|
-
|
270
|
-
|
271
|
-
|
272
|
-
|
273
|
-
|
174
|
+
def should_continue?(tokens)
|
175
|
+
# Look at the last token and check if IRB need to continue reading next line.
|
176
|
+
# Example code that should continue: `a\` `a +` `a.`
|
177
|
+
# Trailing spaces, newline, comments are skipped
|
178
|
+
return true if tokens.last&.event == :on_sp && tokens.last.tok == "\\\n"
|
179
|
+
|
180
|
+
tokens.reverse_each do |token|
|
181
|
+
case token.event
|
182
|
+
when :on_sp, :on_nl, :on_ignored_nl, :on_comment, :on_embdoc_beg, :on_embdoc, :on_embdoc_end
|
183
|
+
# Skip
|
184
|
+
when :on_regexp_end, :on_heredoc_end, :on_semicolon
|
185
|
+
# State is EXPR_BEG but should not continue
|
186
|
+
return false
|
187
|
+
else
|
188
|
+
# Endless range should not continue
|
189
|
+
return false if token.event == :on_op && token.tok.match?(/\A\.\.\.?\z/)
|
274
190
|
|
275
|
-
|
276
|
-
|
191
|
+
# EXPR_DOT and most of the EXPR_BEG should continue
|
192
|
+
return token.state.anybits?(Ripper::EXPR_BEG | Ripper::EXPR_DOT)
|
193
|
+
end
|
277
194
|
end
|
195
|
+
false
|
278
196
|
end
|
279
|
-
false
|
280
|
-
end
|
281
197
|
|
282
|
-
|
283
|
-
|
284
|
-
|
198
|
+
def check_code_syntax(code, local_variables:)
|
199
|
+
lvars_code = RubyLex.generate_local_variables_assign_code(local_variables)
|
200
|
+
code = "#{lvars_code}\n#{code}"
|
285
201
|
|
286
|
-
|
287
|
-
|
288
|
-
|
289
|
-
|
290
|
-
|
291
|
-
|
202
|
+
begin # check if parser error are available
|
203
|
+
verbose, $VERBOSE = $VERBOSE, nil
|
204
|
+
case RUBY_ENGINE
|
205
|
+
when 'ruby'
|
206
|
+
self.class.compile_with_errors_suppressed(code) do |inner_code, line_no|
|
207
|
+
RubyVM::InstructionSequence.compile(inner_code, nil, nil, line_no)
|
208
|
+
end
|
209
|
+
when 'jruby'
|
210
|
+
JRuby.compile_ir(code)
|
211
|
+
else
|
212
|
+
catch(:valid) do
|
213
|
+
eval("BEGIN { throw :valid, true }\n#{code}")
|
214
|
+
false
|
215
|
+
end
|
292
216
|
end
|
293
|
-
|
294
|
-
|
295
|
-
|
296
|
-
|
297
|
-
|
298
|
-
|
217
|
+
rescue EncodingError
|
218
|
+
# This is for a hash with invalid encoding symbol, {"\xAE": 1}
|
219
|
+
:unrecoverable_error
|
220
|
+
rescue SyntaxError => e
|
221
|
+
case e.message
|
222
|
+
when /unexpected keyword_end/
|
223
|
+
# "syntax error, unexpected keyword_end"
|
224
|
+
#
|
225
|
+
# example:
|
226
|
+
# if (
|
227
|
+
# end
|
228
|
+
#
|
229
|
+
# example:
|
230
|
+
# end
|
231
|
+
return :unrecoverable_error
|
232
|
+
when /unexpected '\.'/
|
233
|
+
# "syntax error, unexpected '.'"
|
234
|
+
#
|
235
|
+
# example:
|
236
|
+
# .
|
237
|
+
return :unrecoverable_error
|
238
|
+
when /unexpected tREGEXP_BEG/
|
239
|
+
# "syntax error, unexpected tREGEXP_BEG, expecting keyword_do or '{' or '('"
|
240
|
+
#
|
241
|
+
# example:
|
242
|
+
# method / f /
|
243
|
+
return :unrecoverable_error
|
244
|
+
when /unterminated (?:string|regexp) meets end of file/
|
245
|
+
# "unterminated regexp meets end of file"
|
246
|
+
#
|
247
|
+
# example:
|
248
|
+
# /
|
249
|
+
#
|
250
|
+
# "unterminated string meets end of file"
|
251
|
+
#
|
252
|
+
# example:
|
253
|
+
# '
|
254
|
+
return :recoverable_error
|
255
|
+
when /unexpected end-of-input/
|
256
|
+
# "syntax error, unexpected end-of-input, expecting keyword_end"
|
257
|
+
#
|
258
|
+
# example:
|
259
|
+
# if true
|
260
|
+
# hoge
|
261
|
+
# if false
|
262
|
+
# fuga
|
263
|
+
# end
|
264
|
+
return :recoverable_error
|
265
|
+
else
|
266
|
+
return :other_error
|
299
267
|
end
|
268
|
+
ensure
|
269
|
+
$VERBOSE = verbose
|
300
270
|
end
|
301
|
-
|
302
|
-
# This is for a hash with invalid encoding symbol, {"\xAE": 1}
|
303
|
-
:unrecoverable_error
|
304
|
-
rescue SyntaxError => e
|
305
|
-
case e.message
|
306
|
-
when /unterminated (?:string|regexp) meets end of file/
|
307
|
-
# "unterminated regexp meets end of file"
|
308
|
-
#
|
309
|
-
# example:
|
310
|
-
# /
|
311
|
-
#
|
312
|
-
# "unterminated string meets end of file"
|
313
|
-
#
|
314
|
-
# example:
|
315
|
-
# '
|
316
|
-
return :recoverable_error
|
317
|
-
when /syntax error, unexpected end-of-input/
|
318
|
-
# "syntax error, unexpected end-of-input, expecting keyword_end"
|
319
|
-
#
|
320
|
-
# example:
|
321
|
-
# if true
|
322
|
-
# hoge
|
323
|
-
# if false
|
324
|
-
# fuga
|
325
|
-
# end
|
326
|
-
return :recoverable_error
|
327
|
-
when /syntax error, unexpected keyword_end/
|
328
|
-
# "syntax error, unexpected keyword_end"
|
329
|
-
#
|
330
|
-
# example:
|
331
|
-
# if (
|
332
|
-
# end
|
333
|
-
#
|
334
|
-
# example:
|
335
|
-
# end
|
336
|
-
return :unrecoverable_error
|
337
|
-
when /syntax error, unexpected '\.'/
|
338
|
-
# "syntax error, unexpected '.'"
|
339
|
-
#
|
340
|
-
# example:
|
341
|
-
# .
|
342
|
-
return :unrecoverable_error
|
343
|
-
when /unexpected tREGEXP_BEG/
|
344
|
-
# "syntax error, unexpected tREGEXP_BEG, expecting keyword_do or '{' or '('"
|
345
|
-
#
|
346
|
-
# example:
|
347
|
-
# method / f /
|
348
|
-
return :unrecoverable_error
|
349
|
-
else
|
350
|
-
return :other_error
|
351
|
-
end
|
352
|
-
ensure
|
353
|
-
$VERBOSE = verbose
|
271
|
+
:valid
|
354
272
|
end
|
355
|
-
:valid
|
356
|
-
end
|
357
273
|
|
358
|
-
|
359
|
-
|
360
|
-
|
361
|
-
|
362
|
-
|
363
|
-
|
364
|
-
|
365
|
-
|
366
|
-
|
367
|
-
|
274
|
+
def calc_indent_level(opens)
|
275
|
+
indent_level = 0
|
276
|
+
opens.each_with_index do |t, index|
|
277
|
+
case t.event
|
278
|
+
when :on_heredoc_beg
|
279
|
+
if opens[index + 1]&.event != :on_heredoc_beg
|
280
|
+
if t.tok.match?(/^<<[~-]/)
|
281
|
+
indent_level += 1
|
282
|
+
else
|
283
|
+
indent_level = 0
|
284
|
+
end
|
368
285
|
end
|
286
|
+
when :on_tstring_beg, :on_regexp_beg, :on_symbeg, :on_backtick
|
287
|
+
# No indent: "", //, :"", ``
|
288
|
+
# Indent: %(), %r(), %i(), %x()
|
289
|
+
indent_level += 1 if t.tok.start_with? '%'
|
290
|
+
when :on_embdoc_beg
|
291
|
+
indent_level = 0
|
292
|
+
else
|
293
|
+
indent_level += 1 unless t.tok == 'alias' || t.tok == 'undef'
|
369
294
|
end
|
370
|
-
when :on_tstring_beg, :on_regexp_beg, :on_symbeg, :on_backtick
|
371
|
-
# can be indented if t.tok starts with `%`
|
372
|
-
when :on_words_beg, :on_qwords_beg, :on_symbols_beg, :on_qsymbols_beg, :on_embexpr_beg
|
373
|
-
# can be indented but not indented in current implementation
|
374
|
-
when :on_embdoc_beg
|
375
|
-
indent_level = 0
|
376
|
-
else
|
377
|
-
indent_level += 1
|
378
295
|
end
|
296
|
+
indent_level
|
379
297
|
end
|
380
|
-
indent_level
|
381
|
-
end
|
382
298
|
|
383
|
-
|
299
|
+
FREE_INDENT_TOKENS = %i[on_tstring_beg on_backtick on_regexp_beg on_symbeg]
|
384
300
|
|
385
|
-
|
386
|
-
|
387
|
-
|
301
|
+
def free_indent_token?(token)
|
302
|
+
FREE_INDENT_TOKENS.include?(token&.event)
|
303
|
+
end
|
388
304
|
|
389
|
-
|
390
|
-
|
391
|
-
|
392
|
-
|
393
|
-
|
394
|
-
|
395
|
-
|
396
|
-
|
397
|
-
|
398
|
-
|
399
|
-
|
400
|
-
|
401
|
-
|
305
|
+
# Calculates the difference of pasted code's indent and indent calculated from tokens
|
306
|
+
def indent_difference(lines, line_results, line_index)
|
307
|
+
loop do
|
308
|
+
_tokens, prev_opens, _next_opens, min_depth = line_results[line_index]
|
309
|
+
open_token = prev_opens.last
|
310
|
+
if !open_token || (open_token.event != :on_heredoc_beg && !free_indent_token?(open_token))
|
311
|
+
# If the leading whitespace is an indent, return the difference
|
312
|
+
indent_level = calc_indent_level(prev_opens.take(min_depth))
|
313
|
+
calculated_indent = 2 * indent_level
|
314
|
+
actual_indent = lines[line_index][/^ */].size
|
315
|
+
return actual_indent - calculated_indent
|
316
|
+
elsif open_token.event == :on_heredoc_beg && open_token.tok.match?(/^<<[^-~]/)
|
317
|
+
return 0
|
318
|
+
end
|
319
|
+
# If the leading whitespace is not an indent but part of a multiline token
|
320
|
+
# Calculate base_indent of the multiline token's beginning line
|
321
|
+
line_index = open_token.pos[0] - 1
|
402
322
|
end
|
403
|
-
# If the leading whitespace is not an indent but part of a multiline token
|
404
|
-
# Calculate base_indent of the multiline token's beginning line
|
405
|
-
line_index = open_token.pos[0] - 1
|
406
323
|
end
|
407
|
-
end
|
408
324
|
|
409
|
-
|
410
|
-
|
411
|
-
|
412
|
-
|
413
|
-
|
414
|
-
|
415
|
-
|
416
|
-
|
417
|
-
|
418
|
-
|
325
|
+
def process_indent_level(tokens, lines, line_index, is_newline)
|
326
|
+
line_results = NestingParser.parse_by_line(tokens)
|
327
|
+
result = line_results[line_index]
|
328
|
+
if result
|
329
|
+
_tokens, prev_opens, next_opens, min_depth = result
|
330
|
+
else
|
331
|
+
# When last line is empty
|
332
|
+
prev_opens = next_opens = line_results.last[2]
|
333
|
+
min_depth = next_opens.size
|
334
|
+
end
|
419
335
|
|
420
|
-
|
421
|
-
|
422
|
-
|
336
|
+
# To correctly indent line like `end.map do`, we use shortest open tokens on each line for indent calculation.
|
337
|
+
# Shortest open tokens can be calculated by `opens.take(min_depth)`
|
338
|
+
indent = 2 * calc_indent_level(prev_opens.take(min_depth))
|
423
339
|
|
424
|
-
|
340
|
+
preserve_indent = lines[line_index - (is_newline ? 1 : 0)][/^ */].size
|
425
341
|
|
426
|
-
|
427
|
-
|
342
|
+
prev_open_token = prev_opens.last
|
343
|
+
next_open_token = next_opens.last
|
428
344
|
|
429
|
-
|
430
|
-
|
431
|
-
|
432
|
-
|
433
|
-
|
434
|
-
|
435
|
-
else
|
436
|
-
base_indent = 0
|
437
|
-
end
|
438
|
-
|
439
|
-
if free_indent_token?(prev_open_token)
|
440
|
-
if is_newline && prev_open_token.pos[0] == line_index
|
441
|
-
# First newline inside free-indent token
|
442
|
-
base_indent + indent
|
443
|
-
else
|
444
|
-
# Accept any number of indent inside free-indent token
|
445
|
-
preserve_indent
|
446
|
-
end
|
447
|
-
elsif prev_open_token&.event == :on_embdoc_beg || next_open_token&.event == :on_embdoc_beg
|
448
|
-
if prev_open_token&.event == next_open_token&.event
|
449
|
-
# Accept any number of indent inside embdoc content
|
450
|
-
preserve_indent
|
345
|
+
# Calculates base indent for pasted code on the line where prev_open_token is located
|
346
|
+
# irb(main):001:1* if a # base_indent is 2, indent calculated from tokens is 0
|
347
|
+
# irb(main):002:1* if b # base_indent is 6, indent calculated from tokens is 2
|
348
|
+
# irb(main):003:0> c # base_indent is 6, indent calculated from tokens is 4
|
349
|
+
if prev_open_token
|
350
|
+
base_indent = [0, indent_difference(lines, line_results, prev_open_token.pos[0] - 1)].max
|
451
351
|
else
|
452
|
-
|
453
|
-
0
|
352
|
+
base_indent = 0
|
454
353
|
end
|
455
|
-
|
456
|
-
|
457
|
-
|
458
|
-
|
459
|
-
|
460
|
-
tok.match?(/^<<[-~]/) ? base_indent + indent : indent
|
461
|
-
elsif tok.match?(/^<<~/)
|
462
|
-
# Accept extra indent spaces inside `<<~` heredoc
|
463
|
-
[base_indent + indent, preserve_indent].max
|
354
|
+
|
355
|
+
if free_indent_token?(prev_open_token)
|
356
|
+
if is_newline && prev_open_token.pos[0] == line_index
|
357
|
+
# First newline inside free-indent token
|
358
|
+
base_indent + indent
|
464
359
|
else
|
465
|
-
# Accept any number of indent inside
|
360
|
+
# Accept any number of indent inside free-indent token
|
361
|
+
preserve_indent
|
362
|
+
end
|
363
|
+
elsif prev_open_token&.event == :on_embdoc_beg || next_open_token&.event == :on_embdoc_beg
|
364
|
+
if prev_open_token&.event == next_open_token&.event
|
365
|
+
# Accept any number of indent inside embdoc content
|
466
366
|
preserve_indent
|
367
|
+
else
|
368
|
+
# =begin or =end
|
369
|
+
0
|
370
|
+
end
|
371
|
+
elsif prev_open_token&.event == :on_heredoc_beg
|
372
|
+
tok = prev_open_token.tok
|
373
|
+
if prev_opens.size <= next_opens.size
|
374
|
+
if is_newline && lines[line_index].empty? && line_results[line_index - 1][1].last != next_open_token
|
375
|
+
# First line in heredoc
|
376
|
+
tok.match?(/^<<[-~]/) ? base_indent + indent : indent
|
377
|
+
elsif tok.match?(/^<<~/)
|
378
|
+
# Accept extra indent spaces inside `<<~` heredoc
|
379
|
+
[base_indent + indent, preserve_indent].max
|
380
|
+
else
|
381
|
+
# Accept any number of indent inside other heredoc
|
382
|
+
preserve_indent
|
383
|
+
end
|
384
|
+
else
|
385
|
+
# Heredoc close
|
386
|
+
prev_line_indent_level = calc_indent_level(prev_opens)
|
387
|
+
tok.match?(/^<<[~-]/) ? base_indent + 2 * (prev_line_indent_level - 1) : 0
|
467
388
|
end
|
468
389
|
else
|
469
|
-
|
470
|
-
prev_line_indent_level = calc_indent_level(prev_opens)
|
471
|
-
tok.match?(/^<<[~-]/) ? base_indent + 2 * (prev_line_indent_level - 1) : 0
|
390
|
+
base_indent + indent
|
472
391
|
end
|
473
|
-
else
|
474
|
-
base_indent + indent
|
475
392
|
end
|
476
|
-
end
|
477
393
|
|
478
|
-
|
479
|
-
|
480
|
-
|
481
|
-
|
482
|
-
|
483
|
-
|
394
|
+
LTYPE_TOKENS = %i[
|
395
|
+
on_heredoc_beg on_tstring_beg
|
396
|
+
on_regexp_beg on_symbeg on_backtick
|
397
|
+
on_symbols_beg on_qsymbols_beg
|
398
|
+
on_words_beg on_qwords_beg
|
399
|
+
]
|
484
400
|
|
485
|
-
|
486
|
-
|
487
|
-
|
488
|
-
end
|
489
|
-
return nil unless start_token
|
490
|
-
|
491
|
-
case start_token&.event
|
492
|
-
when :on_tstring_beg
|
493
|
-
case start_token&.tok
|
494
|
-
when ?" then ?"
|
495
|
-
when /^%.$/ then ?"
|
496
|
-
when /^%Q.$/ then ?"
|
497
|
-
when ?' then ?'
|
498
|
-
when /^%q.$/ then ?'
|
401
|
+
def ltype_from_open_tokens(opens)
|
402
|
+
start_token = opens.reverse_each.find do |tok|
|
403
|
+
LTYPE_TOKENS.include?(tok.event)
|
499
404
|
end
|
500
|
-
|
501
|
-
|
502
|
-
|
503
|
-
|
504
|
-
|
505
|
-
|
506
|
-
|
507
|
-
|
508
|
-
|
509
|
-
|
510
|
-
else
|
511
|
-
nil
|
512
|
-
end
|
513
|
-
end
|
514
|
-
|
515
|
-
def check_termination_in_prev_line(code)
|
516
|
-
tokens = self.class.ripper_lex_without_warning(code, context: @context)
|
517
|
-
past_first_newline = false
|
518
|
-
index = tokens.rindex do |t|
|
519
|
-
# traverse first token before last line
|
520
|
-
if past_first_newline
|
521
|
-
if t.tok.include?("\n")
|
522
|
-
true
|
405
|
+
return nil unless start_token
|
406
|
+
|
407
|
+
case start_token&.event
|
408
|
+
when :on_tstring_beg
|
409
|
+
case start_token&.tok
|
410
|
+
when ?" then ?"
|
411
|
+
when /^%.$/ then ?"
|
412
|
+
when /^%Q.$/ then ?"
|
413
|
+
when ?' then ?'
|
414
|
+
when /^%q.$/ then ?'
|
523
415
|
end
|
524
|
-
|
525
|
-
|
526
|
-
|
416
|
+
when :on_regexp_beg then ?/
|
417
|
+
when :on_symbeg then ?:
|
418
|
+
when :on_backtick then ?`
|
419
|
+
when :on_qwords_beg then ?]
|
420
|
+
when :on_words_beg then ?]
|
421
|
+
when :on_qsymbols_beg then ?]
|
422
|
+
when :on_symbols_beg then ?]
|
423
|
+
when :on_heredoc_beg
|
424
|
+
start_token&.tok =~ /<<[-~]?(['"`])\w+\1/
|
425
|
+
$1 || ?"
|
527
426
|
else
|
528
|
-
|
427
|
+
nil
|
529
428
|
end
|
530
429
|
end
|
531
430
|
|
532
|
-
|
533
|
-
|
534
|
-
|
535
|
-
|
536
|
-
|
537
|
-
|
538
|
-
|
431
|
+
def check_termination_in_prev_line(code, local_variables:)
|
432
|
+
tokens = self.class.ripper_lex_without_warning(code, local_variables: local_variables)
|
433
|
+
past_first_newline = false
|
434
|
+
index = tokens.rindex do |t|
|
435
|
+
# traverse first token before last line
|
436
|
+
if past_first_newline
|
437
|
+
if t.tok.include?("\n")
|
438
|
+
true
|
439
|
+
end
|
440
|
+
elsif t.tok.include?("\n")
|
441
|
+
past_first_newline = true
|
442
|
+
false
|
443
|
+
else
|
444
|
+
false
|
539
445
|
end
|
540
446
|
end
|
541
447
|
|
542
|
-
if
|
543
|
-
|
544
|
-
|
545
|
-
|
546
|
-
|
547
|
-
|
448
|
+
if index
|
449
|
+
first_token = nil
|
450
|
+
last_line_tokens = tokens[(index + 1)..(tokens.size - 1)]
|
451
|
+
last_line_tokens.each do |t|
|
452
|
+
unless [:on_sp, :on_ignored_sp, :on_comment].include?(t.event)
|
453
|
+
first_token = t
|
454
|
+
break
|
455
|
+
end
|
456
|
+
end
|
457
|
+
|
458
|
+
if first_token && first_token.state != Ripper::EXPR_DOT
|
459
|
+
tokens_without_last_line = tokens[0..index]
|
460
|
+
code_without_last_line = tokens_without_last_line.map(&:tok).join
|
461
|
+
opens_without_last_line = NestingParser.open_tokens(tokens_without_last_line)
|
462
|
+
if code_terminated?(code_without_last_line, tokens_without_last_line, opens_without_last_line, local_variables: local_variables)
|
463
|
+
return last_line_tokens.map(&:tok).join
|
464
|
+
end
|
548
465
|
end
|
549
466
|
end
|
467
|
+
false
|
550
468
|
end
|
551
|
-
false
|
552
469
|
end
|
470
|
+
# :startdoc:
|
553
471
|
end
|
554
|
-
|
472
|
+
|
473
|
+
RubyLex = IRB::RubyLex
|
474
|
+
Object.deprecate_constant(:RubyLex)
|