rusa 0.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +7 -0
- data/LICENSE.txt +21 -0
- data/README.md +175 -0
- data/Rakefile +26 -0
- data/Steepfile +9 -0
- data/examples/calc.rb +29 -0
- data/examples/json.rb +55 -0
- data/examples/mini_lang.rb +52 -0
- data/exe/rusa +6 -0
- data/lib/rusa/analysis/automaton.rb +60 -0
- data/lib/rusa/analysis/conflict_resolver.rb +211 -0
- data/lib/rusa/analysis/first_follow.rb +106 -0
- data/lib/rusa/analysis/item.rb +51 -0
- data/lib/rusa/analysis/item_set.rb +64 -0
- data/lib/rusa/analysis/lalr_table.rb +460 -0
- data/lib/rusa/analysis/parse_action.rb +81 -0
- data/lib/rusa/cli.rb +188 -0
- data/lib/rusa/errors.rb +12 -0
- data/lib/rusa/generator/code_generator.rb +334 -0
- data/lib/rusa/grammar/action_capture.rb +128 -0
- data/lib/rusa/grammar/dsl.rb +123 -0
- data/lib/rusa/grammar/grammar.rb +212 -0
- data/lib/rusa/grammar/precedence.rb +29 -0
- data/lib/rusa/grammar/rule.rb +55 -0
- data/lib/rusa/grammar/symbol.rb +71 -0
- data/lib/rusa/version.rb +5 -0
- data/lib/rusa.rb +31 -0
- data/sig/generated/rusa/analysis/automaton.rbs +25 -0
- data/sig/generated/rusa/analysis/conflict_resolver.rbs +57 -0
- data/sig/generated/rusa/analysis/first_follow.rbs +33 -0
- data/sig/generated/rusa/analysis/item.rbs +35 -0
- data/sig/generated/rusa/analysis/item_set.rbs +31 -0
- data/sig/generated/rusa/analysis/lalr_table.rbs +182 -0
- data/sig/generated/rusa/analysis/parse_action.rbs +58 -0
- data/sig/generated/rusa/cli.rbs +68 -0
- data/sig/generated/rusa/errors.rbs +24 -0
- data/sig/generated/rusa/generator/code_generator.rbs +82 -0
- data/sig/generated/rusa/grammar/action_capture.rbs +46 -0
- data/sig/generated/rusa/grammar/dsl.rbs +62 -0
- data/sig/generated/rusa/grammar/grammar.rbs +103 -0
- data/sig/generated/rusa/grammar/precedence.rbs +23 -0
- data/sig/generated/rusa/grammar/rule.rbs +35 -0
- data/sig/generated/rusa/grammar/symbol.rbs +51 -0
- data/sig/generated/rusa/version.rbs +5 -0
- data/sig/generated/rusa.rbs +6 -0
- data/test/test_automaton.rb +27 -0
- data/test/test_code_generator.rb +74 -0
- data/test/test_dsl.rb +77 -0
- data/test/test_e2e.rb +134 -0
- data/test/test_first_follow.rb +70 -0
- data/test/test_grammar_model.rb +60 -0
- data/test/test_helper.rb +6 -0
- data/test/test_lalr_table.rb +64 -0
- metadata +96 -0
checksums.yaml
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
1
|
+
---
|
|
2
|
+
SHA256:
|
|
3
|
+
metadata.gz: 3c28ea4a3cf8a916002f8824535cd4ff420762aa9ce83480852b58e1abc094dc
|
|
4
|
+
data.tar.gz: a69bd45ee7cb9b135b46904cedd492956fe8f8ccf02777632f17b8d0d336c810
|
|
5
|
+
SHA512:
|
|
6
|
+
metadata.gz: 7a26465250a4347e9645b78638dad53b71460a923fe52c366ebce6c9ef258fa7768cadb18040d24b674a406d8c1494a13a1bd2453094abd7c0ae6f2ceae4d52a
|
|
7
|
+
data.tar.gz: 4445972fbf14b5163f5b096903129248f16ca14cf01cb25b13478cb57bb818587dd77e4c9f9424d4e07c100ab979e360ea0ecd53d25f931abc9ef40ad4368f95
|
data/LICENSE.txt
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
1
|
+
The MIT License (MIT)
|
|
2
|
+
|
|
3
|
+
Copyright (c) 2026 Yudai Takada
|
|
4
|
+
|
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
|
7
|
+
in the Software without restriction, including without limitation the rights
|
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
|
10
|
+
furnished to do so, subject to the following conditions:
|
|
11
|
+
|
|
12
|
+
The above copyright notice and this permission notice shall be included in
|
|
13
|
+
all copies or substantial portions of the Software.
|
|
14
|
+
|
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
|
|
21
|
+
THE SOFTWARE.
|
data/README.md
ADDED
|
@@ -0,0 +1,175 @@
|
|
|
1
|
+
# Rusa
|
|
2
|
+
|
|
3
|
+
Rusa is a pure Ruby LALR(1) parser generator. You define a grammar with a Ruby DSL, and Rusa generates a standalone Ruby parser with no runtime gem dependency.
|
|
4
|
+
|
|
5
|
+
## Features
|
|
6
|
+
|
|
7
|
+
- Pure Ruby implementation
|
|
8
|
+
- Ruby-first DSL for tokens, precedence, and productions
|
|
9
|
+
- Standalone parser output as a single `.rb` file
|
|
10
|
+
- Conflict reporting for shift/reduce and reduce/reduce cases
|
|
11
|
+
- Line and column tracking in generated parse errors
|
|
12
|
+
|
|
13
|
+
## Installation
|
|
14
|
+
|
|
15
|
+
Add the gem to your Gemfile:
|
|
16
|
+
|
|
17
|
+
```ruby
|
|
18
|
+
gem "rusa"
|
|
19
|
+
```
|
|
20
|
+
|
|
21
|
+
Or install it directly:
|
|
22
|
+
|
|
23
|
+
```bash
|
|
24
|
+
gem install rusa
|
|
25
|
+
```
|
|
26
|
+
|
|
27
|
+
## Quick Start
|
|
28
|
+
|
|
29
|
+
Define a grammar:
|
|
30
|
+
|
|
31
|
+
```ruby
|
|
32
|
+
Rusa.grammar do
|
|
33
|
+
token :NUMBER, /\d+/
|
|
34
|
+
token :PLUS, "+"
|
|
35
|
+
token :STAR, "*"
|
|
36
|
+
skip /\s+/
|
|
37
|
+
|
|
38
|
+
left :PLUS
|
|
39
|
+
left :STAR
|
|
40
|
+
|
|
41
|
+
start :expr
|
|
42
|
+
|
|
43
|
+
rule :expr do
|
|
44
|
+
alt(:expr, :PLUS, :expr) { |left, _, right| [:add, left, right] }
|
|
45
|
+
alt(:expr, :STAR, :expr) { |left, _, right| [:mul, left, right] }
|
|
46
|
+
alt(:NUMBER) { |number| number.to_i }
|
|
47
|
+
end
|
|
48
|
+
end
|
|
49
|
+
```
|
|
50
|
+
|
|
51
|
+
Generate a parser:
|
|
52
|
+
|
|
53
|
+
```bash
|
|
54
|
+
exe/rusa generate examples/calc.rb -o calc_parser.rb --class CalcParser
|
|
55
|
+
```
|
|
56
|
+
|
|
57
|
+
Use the generated parser:
|
|
58
|
+
|
|
59
|
+
```ruby
|
|
60
|
+
require_relative "calc_parser"
|
|
61
|
+
|
|
62
|
+
parser = CalcParser.new
|
|
63
|
+
parser.parse("1 + 2 * 3")
|
|
64
|
+
# => [:add, 1, [:mul, 2, 3]]
|
|
65
|
+
```
|
|
66
|
+
|
|
67
|
+
## DSL Reference
|
|
68
|
+
|
|
69
|
+
### Tokens and Skips
|
|
70
|
+
|
|
71
|
+
```ruby
|
|
72
|
+
token :NUMBER, /\d+/
|
|
73
|
+
token :PLUS, "+"
|
|
74
|
+
skip /\s+/
|
|
75
|
+
```
|
|
76
|
+
|
|
77
|
+
- `token(name, pattern)` accepts a literal string or regular expression.
|
|
78
|
+
- String literals are escaped and anchored automatically.
|
|
79
|
+
- Identifier-like keywords such as `"if"` are emitted with a trailing word boundary.
|
|
80
|
+
|
|
81
|
+
### Precedence
|
|
82
|
+
|
|
83
|
+
```ruby
|
|
84
|
+
left :PLUS, :MINUS
|
|
85
|
+
left :STAR, :SLASH
|
|
86
|
+
right :UMINUS
|
|
87
|
+
nonassoc :EQ, :NEQ
|
|
88
|
+
```
|
|
89
|
+
|
|
90
|
+
Each call creates one precedence level. Later calls bind tighter.
|
|
91
|
+
|
|
92
|
+
### Rules
|
|
93
|
+
|
|
94
|
+
```ruby
|
|
95
|
+
start :expr
|
|
96
|
+
|
|
97
|
+
rule :expr do
|
|
98
|
+
alt(:expr, :PLUS, :expr) { |left, _, right| [:add, left, right] }
|
|
99
|
+
alt(:NUMBER) { |number| number.to_i }
|
|
100
|
+
end
|
|
101
|
+
```
|
|
102
|
+
|
|
103
|
+
Empty productions are written as `alt`.
|
|
104
|
+
|
|
105
|
+
### Semantic Actions
|
|
106
|
+
|
|
107
|
+
File-backed grammar blocks are embedded into generated code automatically:
|
|
108
|
+
|
|
109
|
+
```ruby
|
|
110
|
+
alt(:NUMBER) { |number| number.to_i }
|
|
111
|
+
```
|
|
112
|
+
|
|
113
|
+
For REPL or eval-driven grammars, use string actions:
|
|
114
|
+
|
|
115
|
+
```ruby
|
|
116
|
+
alt(:NUMBER).action("_1.to_i")
|
|
117
|
+
alt(:expr, :PLUS, :expr).action("[:add, _1, _3]")
|
|
118
|
+
```
|
|
119
|
+
|
|
120
|
+
Context precedence is available with `prec`:
|
|
121
|
+
|
|
122
|
+
```ruby
|
|
123
|
+
alt(:MINUS, :expr).prec(:UMINUS).action("[:neg, _2]")
|
|
124
|
+
```
|
|
125
|
+
|
|
126
|
+
## CLI
|
|
127
|
+
|
|
128
|
+
Generate a parser:
|
|
129
|
+
|
|
130
|
+
```bash
|
|
131
|
+
exe/rusa generate GRAMMAR.rb [-o OUTPUT.rb] [--class NAME] [--compact] [--verbose] [--no-line-tracking]
|
|
132
|
+
```
|
|
133
|
+
|
|
134
|
+
Inspect conflicts and warnings:
|
|
135
|
+
|
|
136
|
+
```bash
|
|
137
|
+
exe/rusa report GRAMMAR.rb
|
|
138
|
+
```
|
|
139
|
+
|
|
140
|
+
Validate a grammar:
|
|
141
|
+
|
|
142
|
+
```bash
|
|
143
|
+
exe/rusa check GRAMMAR.rb
|
|
144
|
+
```
|
|
145
|
+
|
|
146
|
+
## Examples
|
|
147
|
+
|
|
148
|
+
- `examples/calc.rb`: arithmetic expressions with precedence
|
|
149
|
+
- `examples/json.rb`: JSON parser returning Ruby objects
|
|
150
|
+
- `examples/mini_lang.rb`: small statement language with `if`, `while`, and assignment
|
|
151
|
+
|
|
152
|
+
## Development
|
|
153
|
+
|
|
154
|
+
Run the test suite:
|
|
155
|
+
|
|
156
|
+
```bash
|
|
157
|
+
rake test
|
|
158
|
+
```
|
|
159
|
+
|
|
160
|
+
Refresh generated signatures and run the type checker:
|
|
161
|
+
|
|
162
|
+
```bash
|
|
163
|
+
rake sig:generate
|
|
164
|
+
bundle exec steep check
|
|
165
|
+
```
|
|
166
|
+
|
|
167
|
+
Or directly:
|
|
168
|
+
|
|
169
|
+
```bash
|
|
170
|
+
ruby -Ilib -Itest -e "Dir['test/test_*.rb'].sort.each { |f| require_relative f }"
|
|
171
|
+
```
|
|
172
|
+
|
|
173
|
+
## License
|
|
174
|
+
|
|
175
|
+
The gem is available as open source under the terms of the [MIT License](https://opensource.org/licenses/MIT).
|
data/Rakefile
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
1
|
+
# frozen_string_literal: true
|
|
2
|
+
|
|
3
|
+
require "bundler/gem_tasks"
|
|
4
|
+
require "rake/testtask"
|
|
5
|
+
|
|
6
|
+
Rake::TestTask.new(:test) do |task|
|
|
7
|
+
task.libs << "lib"
|
|
8
|
+
task.libs << "test"
|
|
9
|
+
task.pattern = "test/test_*.rb"
|
|
10
|
+
task.warning = true
|
|
11
|
+
end
|
|
12
|
+
|
|
13
|
+
desc "Generate RBS files from inline annotations"
|
|
14
|
+
task :"sig:generate" do
|
|
15
|
+
sh "rbs-inline", "--opt-out", "--output", "sig/generated", "lib"
|
|
16
|
+
end
|
|
17
|
+
|
|
18
|
+
begin
|
|
19
|
+
require "steep/rake_task"
|
|
20
|
+
|
|
21
|
+
Steep::RakeTask.new(:steep)
|
|
22
|
+
rescue LoadError
|
|
23
|
+
nil
|
|
24
|
+
end
|
|
25
|
+
|
|
26
|
+
task default: :test
|
data/Steepfile
ADDED
data/examples/calc.rb
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
1
|
+
# frozen_string_literal: true
|
|
2
|
+
|
|
3
|
+
require_relative "../lib/rusa"
|
|
4
|
+
|
|
5
|
+
Rusa.grammar do
|
|
6
|
+
token :NUMBER,(/\d+(?:\.\d+)?/)
|
|
7
|
+
token :PLUS, "+"
|
|
8
|
+
token :MINUS, "-"
|
|
9
|
+
token :STAR, "*"
|
|
10
|
+
token :SLASH, "/"
|
|
11
|
+
token :LPAREN, "("
|
|
12
|
+
token :RPAREN, ")"
|
|
13
|
+
|
|
14
|
+
skip(/\s+/)
|
|
15
|
+
|
|
16
|
+
left :PLUS, :MINUS
|
|
17
|
+
left :STAR, :SLASH
|
|
18
|
+
|
|
19
|
+
start :expr
|
|
20
|
+
|
|
21
|
+
rule :expr do
|
|
22
|
+
alt(:expr, :PLUS, :expr) { |left, _, right| [:add, left, right] }
|
|
23
|
+
alt(:expr, :MINUS, :expr) { |left, _, right| [:sub, left, right] }
|
|
24
|
+
alt(:expr, :STAR, :expr) { |left, _, right| [:mul, left, right] }
|
|
25
|
+
alt(:expr, :SLASH, :expr) { |left, _, right| [:div, left, right] }
|
|
26
|
+
alt(:LPAREN, :expr, :RPAREN) { |_, expression, _| expression }
|
|
27
|
+
alt(:NUMBER) { |number| number.include?(".") ? number.to_f : number.to_i }
|
|
28
|
+
end
|
|
29
|
+
end
|
data/examples/json.rb
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
1
|
+
# frozen_string_literal: true
|
|
2
|
+
|
|
3
|
+
require_relative "../lib/rusa"
|
|
4
|
+
|
|
5
|
+
Rusa.grammar do
|
|
6
|
+
token :LBRACE, "{"
|
|
7
|
+
token :RBRACE, "}"
|
|
8
|
+
token :LBRACKET, "["
|
|
9
|
+
token :RBRACKET, "]"
|
|
10
|
+
token :COLON, ":"
|
|
11
|
+
token :COMMA, ","
|
|
12
|
+
token :TRUE, "true"
|
|
13
|
+
token :FALSE, "false"
|
|
14
|
+
token :NULL, "null"
|
|
15
|
+
token :STRING,(/"(?:\\.|[^"\\])*"/)
|
|
16
|
+
token :NUMBER,(/-?(?:0|[1-9]\d*)(?:\.\d+)?(?:[eE][+-]?\d+)?/)
|
|
17
|
+
|
|
18
|
+
skip(/\s+/)
|
|
19
|
+
|
|
20
|
+
start :value
|
|
21
|
+
|
|
22
|
+
rule :value do
|
|
23
|
+
alt(:object).action("_1")
|
|
24
|
+
alt(:array).action("_1")
|
|
25
|
+
alt(:STRING).action("_1[1..-2]")
|
|
26
|
+
alt(:NUMBER).action("_1.match?(/[.eE]/) ? _1.to_f : _1.to_i")
|
|
27
|
+
alt(:TRUE).action("true")
|
|
28
|
+
alt(:FALSE).action("false")
|
|
29
|
+
alt(:NULL).action("nil")
|
|
30
|
+
end
|
|
31
|
+
|
|
32
|
+
rule :object do
|
|
33
|
+
alt(:LBRACE, :members, :RBRACE).action("_2")
|
|
34
|
+
alt(:LBRACE, :RBRACE).action("{}")
|
|
35
|
+
end
|
|
36
|
+
|
|
37
|
+
rule :members do
|
|
38
|
+
alt(:pair).action("{ _1[0] => _1[1] }")
|
|
39
|
+
alt(:members, :COMMA, :pair).action("_1.merge(_3[0] => _3[1])")
|
|
40
|
+
end
|
|
41
|
+
|
|
42
|
+
rule :pair do
|
|
43
|
+
alt(:STRING, :COLON, :value).action("[_1[1..-2], _3]")
|
|
44
|
+
end
|
|
45
|
+
|
|
46
|
+
rule :array do
|
|
47
|
+
alt(:LBRACKET, :elements, :RBRACKET).action("_2")
|
|
48
|
+
alt(:LBRACKET, :RBRACKET).action("[]")
|
|
49
|
+
end
|
|
50
|
+
|
|
51
|
+
rule :elements do
|
|
52
|
+
alt(:value).action("[_1]")
|
|
53
|
+
alt(:elements, :COMMA, :value).action("_1 + [_3]")
|
|
54
|
+
end
|
|
55
|
+
end
|
|
@@ -0,0 +1,52 @@
|
|
|
1
|
+
# frozen_string_literal: true
|
|
2
|
+
|
|
3
|
+
require_relative "../lib/rusa"
|
|
4
|
+
|
|
5
|
+
Rusa.grammar do
|
|
6
|
+
token :IF, "if"
|
|
7
|
+
token :THEN, "then"
|
|
8
|
+
token :ELSE, "else"
|
|
9
|
+
token :WHILE, "while"
|
|
10
|
+
token :DO, "do"
|
|
11
|
+
token :END, "end"
|
|
12
|
+
token :IDENT,(/[a-z_]\w*/)
|
|
13
|
+
token :NUMBER,(/\d+/)
|
|
14
|
+
token :ASSIGN, "="
|
|
15
|
+
token :SEMI, ";"
|
|
16
|
+
token :PLUS, "+"
|
|
17
|
+
token :STAR, "*"
|
|
18
|
+
token :LPAREN, "("
|
|
19
|
+
token :RPAREN, ")"
|
|
20
|
+
|
|
21
|
+
skip(/\s+/)
|
|
22
|
+
|
|
23
|
+
left :PLUS
|
|
24
|
+
left :STAR
|
|
25
|
+
|
|
26
|
+
start :program
|
|
27
|
+
|
|
28
|
+
rule :program do
|
|
29
|
+
alt(:stmts).action("[:program, _1]")
|
|
30
|
+
end
|
|
31
|
+
|
|
32
|
+
rule :stmts do
|
|
33
|
+
alt(:stmt).action("[_1]")
|
|
34
|
+
alt(:stmts, :SEMI, :stmt).action("_1 + [_3]")
|
|
35
|
+
end
|
|
36
|
+
|
|
37
|
+
rule :stmt do
|
|
38
|
+
alt(:IDENT, :ASSIGN, :expr).action("[:assign, _1, _3]")
|
|
39
|
+
alt(:IF, :expr, :THEN, :stmts, :END).action("[:if, _2, _4]")
|
|
40
|
+
alt(:IF, :expr, :THEN, :stmts, :ELSE, :stmts, :END).action("[:if, _2, _4, _6]")
|
|
41
|
+
alt(:WHILE, :expr, :DO, :stmts, :END).action("[:while, _2, _4]")
|
|
42
|
+
alt(:expr).action("_1")
|
|
43
|
+
end
|
|
44
|
+
|
|
45
|
+
rule :expr do
|
|
46
|
+
alt(:expr, :PLUS, :expr).action("[:add, _1, _3]")
|
|
47
|
+
alt(:expr, :STAR, :expr).action("[:mul, _1, _3]")
|
|
48
|
+
alt(:LPAREN, :expr, :RPAREN).action("_2")
|
|
49
|
+
alt(:NUMBER).action("_1.to_i")
|
|
50
|
+
alt(:IDENT).action("[:var, _1]")
|
|
51
|
+
end
|
|
52
|
+
end
|
data/exe/rusa
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
1
|
+
# frozen_string_literal: true
|
|
2
|
+
|
|
3
|
+
module Rusa
|
|
4
|
+
module Analysis
|
|
5
|
+
# Automaton builds the canonical LR(0) state graph.
|
|
6
|
+
class Automaton
|
|
7
|
+
attr_reader :states #: Array[ItemSet]
|
|
8
|
+
attr_reader :transitions #: Hash[Integer, Hash[Symbol, Integer]]
|
|
9
|
+
|
|
10
|
+
#: (Grammar::Grammar) -> void
|
|
11
|
+
def initialize(grammar)
|
|
12
|
+
@grammar = grammar
|
|
13
|
+
@states = []
|
|
14
|
+
@transitions = Hash.new do |hash, key|
|
|
15
|
+
hash[key] = {} #: Hash[Symbol, Integer]
|
|
16
|
+
end
|
|
17
|
+
build
|
|
18
|
+
end
|
|
19
|
+
|
|
20
|
+
private
|
|
21
|
+
|
|
22
|
+
attr_reader :grammar #: Grammar::Grammar
|
|
23
|
+
|
|
24
|
+
#: () -> void
|
|
25
|
+
def build
|
|
26
|
+
start_item = Item.new(grammar.augmented_production, 0)
|
|
27
|
+
state0 = ItemSet.new(0, [start_item], grammar)
|
|
28
|
+
@states << state0
|
|
29
|
+
|
|
30
|
+
known_states = { kernel_key(state0.kernel_items) => state0 }
|
|
31
|
+
worklist = [state0]
|
|
32
|
+
|
|
33
|
+
until worklist.empty?
|
|
34
|
+
current = worklist.shift
|
|
35
|
+
symbols = current.items.map(&:next_symbol).compact.uniq
|
|
36
|
+
|
|
37
|
+
symbols.each do |symbol|
|
|
38
|
+
kernel_items = current.items.select { |item| item.next_symbol == symbol }.map(&:advance)
|
|
39
|
+
key = kernel_key(kernel_items)
|
|
40
|
+
target = known_states[key]
|
|
41
|
+
|
|
42
|
+
unless target
|
|
43
|
+
target = ItemSet.new(states.length, kernel_items, grammar)
|
|
44
|
+
known_states[key] = target
|
|
45
|
+
states << target
|
|
46
|
+
worklist << target
|
|
47
|
+
end
|
|
48
|
+
|
|
49
|
+
transitions[current.id][symbol] = target.id
|
|
50
|
+
end
|
|
51
|
+
end
|
|
52
|
+
end
|
|
53
|
+
|
|
54
|
+
#: (Enumerable[Item]) -> Array[Array[Integer?]]
|
|
55
|
+
def kernel_key(items)
|
|
56
|
+
items.map { |item| [item.production.id, item.dot] }.sort.freeze
|
|
57
|
+
end
|
|
58
|
+
end
|
|
59
|
+
end
|
|
60
|
+
end
|
|
@@ -0,0 +1,211 @@
|
|
|
1
|
+
# frozen_string_literal: true
|
|
2
|
+
|
|
3
|
+
module Rusa
|
|
4
|
+
module Analysis
|
|
5
|
+
class ConflictReport
|
|
6
|
+
attr_reader :state_id #: Integer
|
|
7
|
+
attr_reader :lookahead #: Symbol
|
|
8
|
+
attr_reader :existing_action #: Shift | Reduce | Accept
|
|
9
|
+
attr_reader :new_action #: Shift | Reduce | Accept
|
|
10
|
+
attr_reader :resolved_action #: Shift | Reduce | Accept | nil
|
|
11
|
+
attr_reader :message #: String
|
|
12
|
+
|
|
13
|
+
#: (state_id: Integer, lookahead: Symbol, existing_action: Shift | Reduce | Accept, new_action: Shift | Reduce | Accept, resolved_action: Shift | Reduce | Accept | nil, message: String) -> void
|
|
14
|
+
def initialize(
|
|
15
|
+
state_id:,
|
|
16
|
+
lookahead:,
|
|
17
|
+
existing_action:,
|
|
18
|
+
new_action:,
|
|
19
|
+
resolved_action:,
|
|
20
|
+
message:
|
|
21
|
+
)
|
|
22
|
+
@state_id = state_id
|
|
23
|
+
@lookahead = lookahead
|
|
24
|
+
@existing_action = existing_action
|
|
25
|
+
@new_action = new_action
|
|
26
|
+
@resolved_action = resolved_action
|
|
27
|
+
@message = message
|
|
28
|
+
freeze
|
|
29
|
+
end
|
|
30
|
+
end
|
|
31
|
+
|
|
32
|
+
# ConflictResolver applies precedence and associativity rules to ACTION conflicts.
|
|
33
|
+
class ConflictResolver
|
|
34
|
+
#: (Shift | Reduce | Accept, Shift | Reduce | Accept, Symbol, Grammar::Grammar, state_id: Integer, productions_by_id: Hash[Integer?, Grammar::Production]) -> [Shift | Reduce | Accept | nil, ConflictReport]
|
|
35
|
+
def resolve(existing_action, new_action, lookahead, grammar, state_id:, productions_by_id:)
|
|
36
|
+
report = build_report(
|
|
37
|
+
existing_action,
|
|
38
|
+
new_action,
|
|
39
|
+
lookahead,
|
|
40
|
+
grammar,
|
|
41
|
+
state_id: state_id,
|
|
42
|
+
productions_by_id: productions_by_id
|
|
43
|
+
)
|
|
44
|
+
|
|
45
|
+
[report.resolved_action, report]
|
|
46
|
+
end
|
|
47
|
+
|
|
48
|
+
private
|
|
49
|
+
|
|
50
|
+
#: (Shift | Reduce | Accept, Shift | Reduce | Accept, Symbol, Grammar::Grammar, state_id: Integer, productions_by_id: Hash[Integer?, Grammar::Production]) -> ConflictReport
|
|
51
|
+
def build_report(
|
|
52
|
+
existing_action,
|
|
53
|
+
new_action,
|
|
54
|
+
lookahead,
|
|
55
|
+
grammar,
|
|
56
|
+
state_id:,
|
|
57
|
+
productions_by_id:
|
|
58
|
+
)
|
|
59
|
+
return resolve_shift_reduce(
|
|
60
|
+
existing_action,
|
|
61
|
+
new_action,
|
|
62
|
+
lookahead,
|
|
63
|
+
grammar,
|
|
64
|
+
productions_by_id,
|
|
65
|
+
state_id
|
|
66
|
+
) if existing_action.is_a?(Shift) && new_action.is_a?(Reduce)
|
|
67
|
+
|
|
68
|
+
return resolve_shift_reduce(
|
|
69
|
+
new_action,
|
|
70
|
+
existing_action,
|
|
71
|
+
lookahead,
|
|
72
|
+
grammar,
|
|
73
|
+
productions_by_id,
|
|
74
|
+
state_id
|
|
75
|
+
) if existing_action.is_a?(Reduce) && new_action.is_a?(Shift)
|
|
76
|
+
|
|
77
|
+
return resolve_reduce_reduce(
|
|
78
|
+
existing_action,
|
|
79
|
+
new_action,
|
|
80
|
+
lookahead,
|
|
81
|
+
state_id
|
|
82
|
+
) if existing_action.is_a?(Reduce) && new_action.is_a?(Reduce)
|
|
83
|
+
|
|
84
|
+
duplicate_action_report(
|
|
85
|
+
state_id,
|
|
86
|
+
lookahead,
|
|
87
|
+
existing_action,
|
|
88
|
+
new_action
|
|
89
|
+
)
|
|
90
|
+
end
|
|
91
|
+
|
|
92
|
+
#: (Shift, Reduce, Symbol, Grammar::Grammar, Hash[Integer?, Grammar::Production], Integer) -> ConflictReport
|
|
93
|
+
def resolve_shift_reduce(
|
|
94
|
+
shift_action,
|
|
95
|
+
reduce_action,
|
|
96
|
+
lookahead,
|
|
97
|
+
grammar,
|
|
98
|
+
productions_by_id,
|
|
99
|
+
state_id
|
|
100
|
+
)
|
|
101
|
+
shift_precedence = grammar.precedence_for(lookahead)
|
|
102
|
+
reduce_production = productions_by_id.fetch(reduce_action.production_id)
|
|
103
|
+
reduce_precedence = reduce_production.precedence(grammar)
|
|
104
|
+
|
|
105
|
+
resolved_action, message = if shift_precedence && reduce_precedence
|
|
106
|
+
compare_precedences(
|
|
107
|
+
shift_action,
|
|
108
|
+
reduce_action,
|
|
109
|
+
shift_precedence,
|
|
110
|
+
reduce_precedence
|
|
111
|
+
)
|
|
112
|
+
else
|
|
113
|
+
default_shift_resolution(shift_action, lookahead)
|
|
114
|
+
end
|
|
115
|
+
|
|
116
|
+
build_conflict_report(
|
|
117
|
+
state_id: state_id,
|
|
118
|
+
lookahead: lookahead,
|
|
119
|
+
existing_action: shift_action,
|
|
120
|
+
new_action: reduce_action,
|
|
121
|
+
resolved_action: resolved_action,
|
|
122
|
+
message: message
|
|
123
|
+
)
|
|
124
|
+
end
|
|
125
|
+
|
|
126
|
+
#: (Shift, Reduce, Grammar::Precedence, Grammar::Precedence) -> [Shift | Reduce | nil, String]
|
|
127
|
+
def compare_precedences(shift_action, reduce_action, shift_precedence, reduce_precedence)
|
|
128
|
+
if shift_precedence.level > reduce_precedence.level
|
|
129
|
+
return higher_token_precedence_result(shift_action)
|
|
130
|
+
end
|
|
131
|
+
|
|
132
|
+
if shift_precedence.level < reduce_precedence.level
|
|
133
|
+
return higher_production_precedence_result(reduce_action)
|
|
134
|
+
end
|
|
135
|
+
|
|
136
|
+
case shift_precedence.associativity
|
|
137
|
+
when :left
|
|
138
|
+
[reduce_action, "shift/reduce conflict resolved by left associativity"]
|
|
139
|
+
when :right
|
|
140
|
+
[shift_action, "shift/reduce conflict resolved by right associativity"]
|
|
141
|
+
when :nonassoc
|
|
142
|
+
[nil, "shift/reduce conflict resolved to syntax error by nonassociativity"]
|
|
143
|
+
else
|
|
144
|
+
[shift_action, "shift/reduce conflict resolved by default shift"]
|
|
145
|
+
end
|
|
146
|
+
end
|
|
147
|
+
|
|
148
|
+
#: (Reduce, Reduce, Symbol, Integer) -> ConflictReport
|
|
149
|
+
def resolve_reduce_reduce(existing_action, new_action, lookahead, state_id)
|
|
150
|
+
resolved_action = [existing_action, new_action].min_by do |action|
|
|
151
|
+
action.production_id || -1
|
|
152
|
+
end
|
|
153
|
+
|
|
154
|
+
build_conflict_report(
|
|
155
|
+
state_id: state_id,
|
|
156
|
+
lookahead: lookahead,
|
|
157
|
+
existing_action: existing_action,
|
|
158
|
+
new_action: new_action,
|
|
159
|
+
resolved_action: resolved_action,
|
|
160
|
+
message: "reduce/reduce conflict on #{lookahead} resolved by earlier production"
|
|
161
|
+
)
|
|
162
|
+
end
|
|
163
|
+
|
|
164
|
+
#: (Integer, Symbol, Shift | Reduce | Accept, Shift | Reduce | Accept) -> ConflictReport
|
|
165
|
+
def duplicate_action_report(state_id, lookahead, existing_action, new_action)
|
|
166
|
+
build_conflict_report(
|
|
167
|
+
state_id: state_id,
|
|
168
|
+
lookahead: lookahead,
|
|
169
|
+
existing_action: existing_action,
|
|
170
|
+
new_action: new_action,
|
|
171
|
+
resolved_action: existing_action,
|
|
172
|
+
message: "duplicate action kept for #{lookahead}"
|
|
173
|
+
)
|
|
174
|
+
end
|
|
175
|
+
|
|
176
|
+
#: (Shift, Symbol) -> [Shift, String]
|
|
177
|
+
def default_shift_resolution(shift_action, lookahead)
|
|
178
|
+
[shift_action, "shift/reduce conflict on #{lookahead} resolved by default shift"]
|
|
179
|
+
end
|
|
180
|
+
|
|
181
|
+
#: (Shift) -> [Shift, String]
|
|
182
|
+
def higher_token_precedence_result(shift_action)
|
|
183
|
+
[shift_action, "shift/reduce conflict resolved by higher token precedence"]
|
|
184
|
+
end
|
|
185
|
+
|
|
186
|
+
#: (Reduce) -> [Reduce, String]
|
|
187
|
+
def higher_production_precedence_result(reduce_action)
|
|
188
|
+
[reduce_action, "shift/reduce conflict resolved by higher production precedence"]
|
|
189
|
+
end
|
|
190
|
+
|
|
191
|
+
#: (state_id: Integer, lookahead: Symbol, existing_action: Shift | Reduce | Accept, new_action: Shift | Reduce | Accept, resolved_action: Shift | Reduce | Accept | nil, message: String) -> ConflictReport
|
|
192
|
+
def build_conflict_report(
|
|
193
|
+
state_id:,
|
|
194
|
+
lookahead:,
|
|
195
|
+
existing_action:,
|
|
196
|
+
new_action:,
|
|
197
|
+
resolved_action:,
|
|
198
|
+
message:
|
|
199
|
+
)
|
|
200
|
+
ConflictReport.new(
|
|
201
|
+
state_id: state_id,
|
|
202
|
+
lookahead: lookahead,
|
|
203
|
+
existing_action: existing_action,
|
|
204
|
+
new_action: new_action,
|
|
205
|
+
resolved_action: resolved_action,
|
|
206
|
+
message: message
|
|
207
|
+
)
|
|
208
|
+
end
|
|
209
|
+
end
|
|
210
|
+
end
|
|
211
|
+
end
|