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
|
# frozen_string_literal: true
require_relative "test_helper"
return if RUBY_VERSION < "3.1.0" || Prism::BACKEND == :FFI
module Prism
class UnescapeTest < TestCase
module Context
class Base
attr_reader :left, :right
def initialize(left, right)
@left = left
@right = right
end
def name
"#{left}#{right}".delete("\n")
end
private
def code(escape)
"#{left}\\#{escape}#{right}".b
end
def ruby(escape)
previous, $VERBOSE = $VERBOSE, nil
begin
yield eval(code(escape))
rescue SyntaxError
:error
ensure
$VERBOSE = previous
end
end
def prism(escape)
result = Prism.parse(code(escape))
if result.success?
yield result.value.statements.body.first
else
:error
end
end
def `(command)
command
end
end
class List < Base
def ruby_result(escape)
ruby(escape) { |value| value.first.to_s }
end
def prism_result(escape)
prism(escape) { |node| node.elements.first.unescaped }
end
end
class Symbol < Base
def ruby_result(escape)
ruby(escape, &:to_s)
end
def prism_result(escape)
prism(escape, &:unescaped)
end
end
class String < Base
def ruby_result(escape)
ruby(escape, &:itself)
end
def prism_result(escape)
prism(escape, &:unescaped)
end
end
class Heredoc < Base
def ruby_result(escape)
ruby(escape, &:itself)
end
def prism_result(escape)
prism(escape) do |node|
case node.type
when :interpolated_string_node, :interpolated_x_string_node
node.parts.flat_map(&:unescaped).join
else
node.unescaped
end
end
end
end
class RegExp < Base
def ruby_result(escape)
ruby(escape, &:source)
end
def prism_result(escape)
prism(escape, &:unescaped)
end
end
end
def test_char; assert_context(Context::String.new("?", "")); end
def test_sqte; assert_context(Context::String.new("'", "'")); end
def test_dqte; assert_context(Context::String.new("\"", "\"")); end
def test_lwrq; assert_context(Context::String.new("%q[", "]")); end
def test_uprq; assert_context(Context::String.new("%Q[", "]")); end
def test_dstr; assert_context(Context::String.new("%[", "]")); end
def test_xstr; assert_context(Context::String.new("`", "`")); end
def test_lwrx; assert_context(Context::String.new("%x[", "]")); end
def test_h0_1; assert_context(Context::String.new("<<H\n", "\nH")); end
def test_h0_2; assert_context(Context::String.new("<<'H'\n", "\nH")); end
def test_h0_3; assert_context(Context::String.new("<<\"H\"\n", "\nH")); end
def test_h0_4; assert_context(Context::String.new("<<`H`\n", "\nH")); end
def test_hd_1; assert_context(Context::String.new("<<-H\n", "\nH")); end
def test_hd_2; assert_context(Context::String.new("<<-'H'\n", "\nH")); end
def test_hd_3; assert_context(Context::String.new("<<-\"H\"\n", "\nH")); end
def test_hd_4; assert_context(Context::String.new("<<-`H`\n", "\nH")); end
def test_ht_1; assert_context(Context::Heredoc.new("<<~H\n", "\nH")); end
def test_ht_2; assert_context(Context::Heredoc.new("<<~'H'\n", "\nH")); end
def test_ht_3; assert_context(Context::Heredoc.new("<<~\"H\"\n", "\nH")); end
def test_ht_4; assert_context(Context::Heredoc.new("<<~`H`\n", "\nH")); end
def test_pw_1; assert_context(Context::List.new("%w[", "]")); end
def test_pw_2; assert_context(Context::List.new("%w<", ">")); end
def test_uprw; assert_context(Context::List.new("%W[", "]")); end
def test_lwri; assert_context(Context::List.new("%i[", "]")); end
def test_upri; assert_context(Context::List.new("%I[", "]")); end
def test_lwrs; assert_context(Context::Symbol.new("%s[", "]")); end
def test_sym1; assert_context(Context::Symbol.new(":'", "'")); end
def test_sym2; assert_context(Context::Symbol.new(":\"", "\"")); end
def test_reg1; assert_context(Context::RegExp.new("/", "/")); end
def test_reg2; assert_context(Context::RegExp.new("%r[", "]")); end
def test_reg3; assert_context(Context::RegExp.new("%r<", ">")); end
def test_reg4; assert_context(Context::RegExp.new("%r{", "}")); end
def test_reg5; assert_context(Context::RegExp.new("%r(", ")")); end
def test_reg6; assert_context(Context::RegExp.new("%r|", "|")); end
private
def assert_context(context)
octal = [*("0".."7")]
hex = [*("a".."f"), *("A".."F"), *("0".."9")]
(0...256).each do |ord|
# I think this might be a bug in Ruby.
next if context.name == "?" && ord == 0xFF
# We don't currently support scanning for the number of capture groups
# to validate backreferences so these are all going to fail.
next if (context.name == "//" || context.name.start_with?("%r")) && ord.chr.start_with?(/\d/)
# \a \b \c ...
assert_unescape(context, ord.chr)
end
# \\r\n
assert_unescape(context, "\r\n")
# We don't currently support scanning for the number of capture groups to
# validate backreferences so these are all going to fail.
if context.name != "//" && !context.name.start_with?("%r")
# \00 \01 \02 ...
octal.product(octal).each { |digits| assert_unescape(context, digits.join) }
# \000 \001 \002 ...
octal.product(octal).product(octal).each { |digits| assert_unescape(context, digits.join) }
end
# \x0 \x1 \x2 ...
hex.each { |digit| assert_unescape(context, "x#{digit}") }
# \x00 \x01 \x02 ...
hex.product(hex).each { |digits| assert_unescape(context, "x#{digits.join}") }
# \u0000 \u0001 \u0002 ...
assert_unescape(context, "u#{["5"].concat(hex.sample(3)).join}")
# The behavior of whitespace in the middle of these escape sequences
# changed in Ruby 3.3.0, so we only want to test against latest.
if RUBY_VERSION >= "3.3.0"
# \u{00 00} ...
assert_unescape(context, "u{00#{["5"].concat(hex.sample(3)).join} \t\v 00#{["5"].concat(hex.sample(3)).join}}")
end
(0...128).each do |ord|
chr = ord.chr
next if chr == "\\" || !chr.match?(/[[:print:]]/)
# \C-a \C-b \C-c ...
assert_unescape(context, "C-#{chr}")
# \ca \cb \cc ...
assert_unescape(context, "c#{chr}")
# \M-a \M-b \M-c ...
assert_unescape(context, "M-#{chr}")
# \M-\C-a \M-\C-b \M-\C-c ...
assert_unescape(context, "M-\\C-#{chr}")
# \M-\ca \M-\cb \M-\cc ...
assert_unescape(context, "M-\\c#{chr}")
# \c\M-a \c\M-b \c\M-c ...
assert_unescape(context, "c\\M-#{chr}")
end
end
def assert_unescape(context, escape)
expected = context.ruby_result(escape)
actual = context.prism_result(escape)
message = -> do
"Expected #{context.name} to unescape #{escape.inspect} to " \
"#{expected.inspect}, but got #{actual.inspect}"
end
if expected == :error || actual == :error
assert_equal expected, actual, message
else
assert_equal expected.bytes, actual.bytes, message
end
end
end
end
|