[ruby/prism] Fix token incompatibility for `Prism::Translation::Parser::Lexer`

This PR fixes token incompatibility for `Prism::Translation::Parser::Lexer` when using backquoted heredoc indetiner:

```ruby
<<-`  FOO`
a
b
     FOO
```

## Parser gem (Expected)

Returns `tXSTRING_BEG` as the first token:

```console
$ bundle exec ruby -Ilib -rparser/ruby33 -ve \
'buf = Parser::Source::Buffer.new("example.rb"); buf.source = File.read("example.rb"); p Parser::Ruby33.new.tokenize(buf)'
ruby 3.3.0 (2023-12-25 revision https://github.com/ruby/prism/commit/5124f9ac75) [x86_64-darwin22]
[s(:xstr,
  s(:str, "a\n"),
  s(:str, "b\n")), [], [[:tXSTRING_BEG, ["<<`", #<Parser::Source::Range example.rb 0...10>]],
[:tSTRING_CONTENT, ["a\n", #<Parser::Source::Range example.rb 11...13>]],
[:tSTRING_CONTENT, ["b\n", #<Parser::Source::Range example.rb 13...15>]],
[:tSTRING_END, ["  FOO", #<Parser::Source::Range example.rb 15...23>]], [:tNL, [nil, #<Parser::Source::Range example.rb 10...11>]]]]
```

## `Prism::Translation::Parser` (Actual)

Previously, the tokens returned by the Parser gem were different. The escaped backslash does not match in the `tSTRING_BEG` token and
value of `tSTRING_END` token.

```console
$ bundle exec ruby -Ilib -rprism -rprism/translation/parser33 -ve \
'buf = Parser::Source::Buffer.new("example.rb"); buf.source = File.read("example.rb"); p Prism::Translation::Parser33.new.tokenize(buf)'
ruby 3.3.0 (2023-12-25 revision https://github.com/ruby/prism/commit/5124f9ac75) [x86_64-darwin22]
[s(:xstr,
  s(:str, "a\n"),
  s(:str, "b\n")), [], [[:tSTRING_BEG, ["<<\"", #<Parser::Source::Range example.rb 0...10>]],
[:tSTRING_CONTENT, ["a\n", #<Parser::Source::Range example.rb 11...13>]],
[:tSTRING_CONTENT, ["b\n", #<Parser::Source::Range example.rb 13...15>]],
[:tSTRING_END, ["`  FOO`", #<Parser::Source::Range example.rb 15...23>]], [:tNL, [nil, #<Parser::Source::Range example.rb 10...11>]]]]
```

After this correction, the AST and tokens returned by the Parser gem are the same:

```console
$ bunlde exec ruby -Ilib -rprism -rprism/translation/parser33 -ve \
'buf = Parser::Source::Buffer.new("example.rb"); buf.source = File.read("example.rb"); p Prism::Translation::Parser33.new.tokenize(buf)'
ruby 3.3.0 (2023-12-25 revision https://github.com/ruby/prism/commit/5124f9ac75) [x86_64-darwin22]
[s(:xstr,
  s(:str, "a\n"),
  s(:str, "b\n")), [], [[:tXSTRING_BEG, ["<<`", #<Parser::Source::Range example.rb 0...10>]],
[:tSTRING_CONTENT, ["a\n", #<Parser::Source::Range example.rb 11...13>]],
[:tSTRING_CONTENT, ["b\n", #<Parser::Source::Range example.rb 13...15>]],
[:tSTRING_END, ["  FOO", #<Parser::Source::Range example.rb 15...23>]], [:tNL, [nil, #<Parser::Source::Range example.rb 10...11>]]]]
```

https://github.com/ruby/prism/commit/308f8d85a1
This commit is contained in:
Koichi ITO 2024-03-16 15:21:36 +09:00 коммит произвёл git
Родитель 815c7e197c
Коммит 3605d6076d
3 изменённых файлов: 38 добавлений и 22 удалений

Просмотреть файл

@ -278,7 +278,7 @@ module Prism
value = nil value = nil
when :tSTRING_BEG when :tSTRING_BEG
if token.type == :HEREDOC_START if token.type == :HEREDOC_START
heredoc_identifier_stack.push(value.match(/<<[-~]?["']?(?<heredoc_identifier>.*?)["']?\z/)[:heredoc_identifier]) heredoc_identifier_stack.push(value.match(/<<[-~]?["'`]?(?<heredoc_identifier>.*?)["'`]?\z/)[:heredoc_identifier])
end end
if ["\"", "'"].include?(value) && (next_token = lexed[index][0]) && next_token.type == :STRING_END if ["\"", "'"].include?(value) && (next_token = lexed[index][0]) && next_token.type == :STRING_END
next_location = token.location.join(next_token.location) next_location = token.location.join(next_token.location)
@ -294,7 +294,12 @@ module Prism
index += 2 index += 2
elsif value.start_with?("<<") elsif value.start_with?("<<")
quote = value[2] == "-" || value[2] == "~" ? value[3] : value[2] quote = value[2] == "-" || value[2] == "~" ? value[3] : value[2]
value = "<<#{quote == "'" || quote == "\"" ? quote : "\""}" if quote == "`"
type = :tXSTRING_BEG
value = "<<`"
else
value = "<<#{quote == "'" || quote == "\"" ? quote : "\""}"
end
end end
when :tSTRING_CONTENT when :tSTRING_CONTENT
unless (lines = token.value.lines).one? unless (lines = token.value.lines).one?

Просмотреть файл

@ -8,6 +8,11 @@ a
b b
FOO FOO
<<-` FOO`
a
b
FOO
<<-' FOO' <<-' FOO'
a a
b b

Просмотреть файл

@ -1,8 +1,8 @@
@ ProgramNode (location: (1,0)-(21,10)) @ ProgramNode (location: (1,0)-(26,10))
├── locals: [] ├── locals: []
└── statements: └── statements:
@ StatementsNode (location: (1,0)-(21,10)) @ StatementsNode (location: (1,0)-(26,10))
└── body: (length: 5) └── body: (length: 6)
├── @ StringNode (location: (1,0)-(1,10)) ├── @ StringNode (location: (1,0)-(1,10))
│ ├── flags: ∅ │ ├── flags: ∅
│ ├── opening_loc: (1,0)-(1,10) = "<<-' FOO'" │ ├── opening_loc: (1,0)-(1,10) = "<<-' FOO'"
@ -15,41 +15,47 @@
│ ├── content_loc: (7,0)-(9,0) = "a\nb\n" │ ├── content_loc: (7,0)-(9,0) = "a\nb\n"
│ ├── closing_loc: (9,0)-(10,0) = " FOO\n" │ ├── closing_loc: (9,0)-(10,0) = " FOO\n"
│ └── unescaped: "a\nb\n" │ └── unescaped: "a\nb\n"
├── @ StringNode (location: (11,0)-(11,10)) ├── @ XStringNode (location: (11,0)-(11,10))
│ ├── flags: ∅ │ ├── flags: ∅
│ ├── opening_loc: (11,0)-(11,10) = "<<-' FOO'" │ ├── opening_loc: (11,0)-(11,10) = "<<-` FOO`"
│ ├── content_loc: (12,0)-(14,0) = "a\nb\n" │ ├── content_loc: (12,0)-(14,0) = "a\nb\n"
│ ├── closing_loc: (14,0)-(15,0) = " FOO\n" │ ├── closing_loc: (14,0)-(15,0) = " FOO\n"
│ └── unescaped: "a\nb\n" │ └── unescaped: "a\nb\n"
├── @ InterpolatedStringNode (location: (16,0)-(16,10)) ├── @ StringNode (location: (16,0)-(16,10))
│ ├── opening_loc: (16,0)-(16,10) = "<<~' FOO'" │ ├── flags: ∅
│ ├── opening_loc: (16,0)-(16,10) = "<<-' FOO'"
│ ├── content_loc: (17,0)-(19,0) = "a\nb\n"
│ ├── closing_loc: (19,0)-(20,0) = " FOO\n"
│ └── unescaped: "a\nb\n"
├── @ InterpolatedStringNode (location: (21,0)-(21,10))
│ ├── opening_loc: (21,0)-(21,10) = "<<~' FOO'"
│ ├── parts: (length: 2) │ ├── parts: (length: 2)
│ │ ├── @ StringNode (location: (17,0)-(18,0)) │ │ ├── @ StringNode (location: (22,0)-(23,0))
│ │ │ ├── flags: ∅ │ │ │ ├── flags: ∅
│ │ │ ├── opening_loc: ∅ │ │ │ ├── opening_loc: ∅
│ │ │ ├── content_loc: (17,0)-(18,0) = "a\n" │ │ │ ├── content_loc: (22,0)-(23,0) = "a\n"
│ │ │ ├── closing_loc: ∅ │ │ │ ├── closing_loc: ∅
│ │ │ └── unescaped: "a\n" │ │ │ └── unescaped: "a\n"
│ │ └── @ StringNode (location: (18,0)-(19,0)) │ │ └── @ StringNode (location: (23,0)-(24,0))
│ │ ├── flags: ∅ │ │ ├── flags: ∅
│ │ ├── opening_loc: ∅ │ │ ├── opening_loc: ∅
│ │ ├── content_loc: (18,0)-(19,0) = "b\n" │ │ ├── content_loc: (23,0)-(24,0) = "b\n"
│ │ ├── closing_loc: ∅ │ │ ├── closing_loc: ∅
│ │ └── unescaped: "b\n" │ │ └── unescaped: "b\n"
│ └── closing_loc: (19,0)-(20,0) = " FOO\n" │ └── closing_loc: (24,0)-(25,0) = " FOO\n"
└── @ InterpolatedStringNode (location: (21,0)-(21,10)) └── @ InterpolatedStringNode (location: (26,0)-(26,10))
├── opening_loc: (21,0)-(21,10) = "<<~' FOO'" ├── opening_loc: (26,0)-(26,10) = "<<~' FOO'"
├── parts: (length: 2) ├── parts: (length: 2)
│ ├── @ StringNode (location: (22,0)-(23,0)) │ ├── @ StringNode (location: (27,0)-(28,0))
│ │ ├── flags: ∅ │ │ ├── flags: ∅
│ │ ├── opening_loc: ∅ │ │ ├── opening_loc: ∅
│ │ ├── content_loc: (22,0)-(23,0) = "a\n" │ │ ├── content_loc: (27,0)-(28,0) = "a\n"
│ │ ├── closing_loc: ∅ │ │ ├── closing_loc: ∅
│ │ └── unescaped: "a\n" │ │ └── unescaped: "a\n"
│ └── @ StringNode (location: (23,0)-(24,0)) │ └── @ StringNode (location: (28,0)-(29,0))
│ ├── flags: ∅ │ ├── flags: ∅
│ ├── opening_loc: ∅ │ ├── opening_loc: ∅
│ ├── content_loc: (23,0)-(24,0) = "b\n" │ ├── content_loc: (28,0)-(29,0) = "b\n"
│ ├── closing_loc: ∅ │ ├── closing_loc: ∅
│ └── unescaped: "b\n" │ └── unescaped: "b\n"
└── closing_loc: (24,0)-(25,0) = " FOO\n" └── closing_loc: (29,0)-(30,0) = " FOO\n"