dolos 0.2.0 → 0.2.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/README.md +14 -5
- data/benchmarks/json/json.rb +31 -8
- data/benchmarks/json/nested_json_1m.json +20557 -0
- data/benchmarks/letter.rb +78 -0
- data/examples/letter.rb +2 -2
- data/lib/dolos/parsers.rb +13 -15
- data/lib/dolos/result.rb +12 -3
- data/lib/dolos/string_io_wrapper.rb +3 -8
- data/lib/dolos/version.rb +1 -1
- data/lib/dolos.rb +13 -39
- data/lib/dolos_common_parsers/common_parsers.rb +4 -0
- data/sig/dolos/common_parsers.rbs +11 -0
- data/sig/dolos/parser.rbs +6 -2
- data/sig/dolos/parser_state.rbs +1 -1
- data/sig/dolos/parsers.rbs +4 -0
- data/sig/dolos/result.rbs +7 -0
- metadata +5 -3
- /data/benchmarks/json/{nested_json.json → nested_json_166.json} +0 -0
@@ -0,0 +1,78 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
require 'bundler/setup'
|
3
|
+
require 'dolos'
|
4
|
+
require 'dolos_common_parsers/common_parsers'
|
5
|
+
require 'benchmark/ips'
|
6
|
+
|
7
|
+
include Dolos
|
8
|
+
|
9
|
+
# Include common parsers
|
10
|
+
# In future this can be more structured, moved them to separate module to prevent breaking changes
|
11
|
+
include Dolos::CommonParsers
|
12
|
+
|
13
|
+
# Library usage example
|
14
|
+
# Parse out a name and address from a letter
|
15
|
+
# For higher difficulty, we will not split this into multiple lines, but instead parse it all at once
|
16
|
+
letter = <<-LETTER
|
17
|
+
Mr. Vardeniui Pavardeniui
|
18
|
+
AB „Lietuvos Paštas“
|
19
|
+
Totorių g. 8
|
20
|
+
01121 Vilnius
|
21
|
+
LETTER
|
22
|
+
|
23
|
+
# Combine with 'or'
|
24
|
+
honorific = c("Mr. ") | c("Mrs. ") | c("Ms. ")
|
25
|
+
|
26
|
+
# Can be parsed any_char which will include needed letters
|
27
|
+
# Or combine LT letters with latin alphabet
|
28
|
+
alpha_with_lt = char_in("ąčęėįšųūžĄČĘĖĮŠŲŪŽ") | alpha
|
29
|
+
|
30
|
+
# Capture all letters in a row and join them,
|
31
|
+
# because they are captured as elements of array by each alpha_with_lt parser.
|
32
|
+
first_name = alpha_with_lt.rep.map(&:join).capture!
|
33
|
+
last_name = alpha_with_lt.rep.map(&:join).capture!
|
34
|
+
|
35
|
+
# Combine first line parsers
|
36
|
+
# Consume zero or more whitespace, after that honorific must follow and so on
|
37
|
+
name_line = ws_rep0 & honorific & first_name & ws & last_name & eol
|
38
|
+
|
39
|
+
# Next line is company info
|
40
|
+
# We could choose to accept UAB and AB or just AB and etc.
|
41
|
+
# 'c("AB")' is for case-sensitive string. 'string' can also be used
|
42
|
+
company_type = c("AB")
|
43
|
+
quote_open = c("„")
|
44
|
+
quote_close = c("“")
|
45
|
+
|
46
|
+
# Consume LT alphabet with whitespace
|
47
|
+
company_name = (alpha_with_lt | ws).rep.map(&:join).capture!
|
48
|
+
company_info = company_type & ws_rep0 & quote_open & company_name & quote_close
|
49
|
+
second_line = ws_rep0 & company_info & eol
|
50
|
+
|
51
|
+
# Address line
|
52
|
+
# 'char_while' will consume characters while passed predicate is true
|
53
|
+
# This could be an alternative to previous 'alpha_with_lt' approach
|
54
|
+
# After that result is captured and mapped to hash
|
55
|
+
# Mapping to hash so at the end its easy to tell tuples apart
|
56
|
+
# Also while mapping, doing some cleaning with '.strip'
|
57
|
+
street_name = char_while(->(char) { !char.match(/\d/) }).map { |s| { street: s.strip } }.capture!
|
58
|
+
building = digits.map { |s| { building: s.strip } }.capture!
|
59
|
+
address_line = ws_rep0 & street_name & building & eol
|
60
|
+
|
61
|
+
# City line
|
62
|
+
# All digits can be matched here or 'digits.rep(5)' could be used. Also joining with map.
|
63
|
+
postcode = digits.map { |s| { postcode: s.strip } }.capture!
|
64
|
+
city = alpha_with_lt.rep.map(&:join).map { |s| { city: s.strip } }.capture!
|
65
|
+
city_line = ws_rep0 & postcode & ws & city & eol
|
66
|
+
|
67
|
+
# Full letter parser which is combined from all previous parsers. All previous parsers can be ran separately.
|
68
|
+
letter_parser = name_line & second_line & address_line & city_line
|
69
|
+
result = letter_parser.run(letter)
|
70
|
+
|
71
|
+
puts result.success?
|
72
|
+
|
73
|
+
Benchmark.ips do |x|
|
74
|
+
x.report('letter benchmark') do
|
75
|
+
letter_parser.run(letter)
|
76
|
+
end
|
77
|
+
x.compare!
|
78
|
+
end
|
data/examples/letter.rb
CHANGED
data/lib/dolos/parsers.rb
CHANGED
@@ -3,9 +3,10 @@
|
|
3
3
|
module Dolos
|
4
4
|
module Parsers
|
5
5
|
def string(str)
|
6
|
+
utf8_str = str.encode('UTF-8')
|
7
|
+
|
6
8
|
Parser.new do |state|
|
7
9
|
state.input.mark_offset
|
8
|
-
utf8_str = str.encode('UTF-8')
|
9
10
|
if state.input.matches?(utf8_str)
|
10
11
|
Success.new(utf8_str, str.bytesize)
|
11
12
|
else
|
@@ -13,7 +14,7 @@ module Dolos
|
|
13
14
|
got_error = state.input.io.string.byteslice(state.input.backup, advanced)
|
14
15
|
state.input.rollback
|
15
16
|
Failure.new(
|
16
|
-
"Expected #{str.inspect} but got #{got_error.inspect}",
|
17
|
+
-> { "Expected #{str.inspect} but got #{got_error.inspect}" },
|
17
18
|
advanced,
|
18
19
|
state
|
19
20
|
)
|
@@ -32,7 +33,7 @@ module Dolos
|
|
32
33
|
advanced = state.input.offset
|
33
34
|
state.input.rollback
|
34
35
|
Failure.new(
|
35
|
-
"Expected pattern #{pattern.inspect} but got #{state.input.io.string.inspect}",
|
36
|
+
-> { "Expected pattern #{pattern.inspect} but got #{state.input.io.string.inspect}" },
|
36
37
|
advanced,
|
37
38
|
state
|
38
39
|
)
|
@@ -52,7 +53,7 @@ module Dolos
|
|
52
53
|
advanced = state.input.offset
|
53
54
|
state.input.rollback
|
54
55
|
Failure.new(
|
55
|
-
'Expected any character but got end of input',
|
56
|
+
-> { 'Expected any character but got end of input' },
|
56
57
|
advanced,
|
57
58
|
state
|
58
59
|
)
|
@@ -64,20 +65,20 @@ module Dolos
|
|
64
65
|
# Example:
|
65
66
|
# char_in('abc').run('b') # => Success.new('b', 1)
|
66
67
|
def char_in(characters_string)
|
67
|
-
|
68
|
+
characters_set = characters_string.chars
|
68
69
|
|
69
70
|
Parser.new do |state|
|
70
71
|
state.input.mark_offset
|
71
72
|
|
72
73
|
char, bytesize = state.input.peek(1)
|
73
74
|
|
74
|
-
if char &&
|
75
|
+
if char && characters_set.include?(char)
|
75
76
|
Success.new(char, bytesize)
|
76
77
|
else
|
77
78
|
advanced = state.input.offset
|
78
79
|
state.input.rollback
|
79
80
|
Failure.new(
|
80
|
-
"Expected one of #{
|
81
|
+
-> { "Expected one of #{characters_set.to_a.inspect} but got #{char.inspect}" },
|
81
82
|
advanced,
|
82
83
|
state
|
83
84
|
)
|
@@ -90,18 +91,18 @@ module Dolos
|
|
90
91
|
state.input.mark_offset
|
91
92
|
|
92
93
|
buffer = String.new
|
93
|
-
|
94
|
-
char, bytesize = state.input.peek(1)
|
95
|
-
break if char.nil? || !predicate.call(char)
|
94
|
+
char, bytesize = state.input.peek(1)
|
96
95
|
|
96
|
+
while char && predicate.call(char)
|
97
97
|
buffer << char
|
98
98
|
state.input.advance(bytesize)
|
99
|
+
char, bytesize = state.input.peek(1)
|
99
100
|
end
|
100
101
|
|
101
102
|
if buffer.empty?
|
102
103
|
advanced = state.input.offset
|
103
104
|
Failure.new(
|
104
|
-
"Predicate never returned true",
|
105
|
+
-> { "Predicate never returned true" },
|
105
106
|
advanced,
|
106
107
|
state
|
107
108
|
)
|
@@ -111,7 +112,6 @@ module Dolos
|
|
111
112
|
end
|
112
113
|
end
|
113
114
|
|
114
|
-
# Unstable API
|
115
115
|
def recursive(&block)
|
116
116
|
recursive_parser = nil
|
117
117
|
|
@@ -120,7 +120,7 @@ module Dolos
|
|
120
120
|
|
121
121
|
recursive_parser.call.run_with_state(state).tap do |result|
|
122
122
|
if result.failure?
|
123
|
-
error_msg = "Error in recursive structure around position #{state.input.offset}: #{result.message}"
|
123
|
+
error_msg = -> { "Error in recursive structure around position #{state.input.offset}: #{result.message}" }
|
124
124
|
Failure.new(error_msg, state.input.offset, state)
|
125
125
|
end
|
126
126
|
end
|
@@ -130,7 +130,5 @@ module Dolos
|
|
130
130
|
placeholder
|
131
131
|
end
|
132
132
|
|
133
|
-
|
134
|
-
|
135
133
|
end
|
136
134
|
end
|
data/lib/dolos/result.rb
CHANGED
@@ -55,12 +55,21 @@ module Dolos
|
|
55
55
|
end
|
56
56
|
|
57
57
|
class Failure < Result
|
58
|
-
attr_reader
|
58
|
+
attr_reader :error_position, :state
|
59
59
|
|
60
|
-
def initialize(
|
61
|
-
@
|
60
|
+
def initialize(message_proc, error_position, state)
|
61
|
+
@message_proc = message_proc
|
62
62
|
@error_position = error_position
|
63
63
|
@state = state
|
64
|
+
@message_evaluated = false
|
65
|
+
end
|
66
|
+
|
67
|
+
def message
|
68
|
+
unless @message_evaluated
|
69
|
+
@message_value = @message_proc.call
|
70
|
+
@message_evaluated = true
|
71
|
+
end
|
72
|
+
@message_value
|
64
73
|
end
|
65
74
|
|
66
75
|
def inspect
|
@@ -22,12 +22,7 @@ module Dolos
|
|
22
22
|
|
23
23
|
def matches?(utf8_str)
|
24
24
|
read = io.read(utf8_str.bytesize)
|
25
|
-
|
26
|
-
if read.nil?
|
27
|
-
false
|
28
|
-
else
|
29
|
-
read.force_encoding('UTF-8') == utf8_str
|
30
|
-
end
|
25
|
+
!read.nil? && read.force_encoding('UTF-8') == utf8_str
|
31
26
|
end
|
32
27
|
|
33
28
|
def advance(bytesize)
|
@@ -61,8 +56,8 @@ module Dolos
|
|
61
56
|
remaining_data = io.read
|
62
57
|
io.seek(current_position)
|
63
58
|
|
64
|
-
if
|
65
|
-
matched_string =
|
59
|
+
if remaining_data =~ /\A#{pattern}/
|
60
|
+
matched_string = $&
|
66
61
|
io.seek(current_position + matched_string.bytesize)
|
67
62
|
return matched_string
|
68
63
|
end
|
data/lib/dolos/version.rb
CHANGED
data/lib/dolos.rb
CHANGED
@@ -10,9 +10,7 @@ module Dolos
|
|
10
10
|
include Parsers
|
11
11
|
|
12
12
|
class Parser
|
13
|
-
|
14
13
|
attr_accessor :parser_proc
|
15
|
-
|
16
14
|
def initialize(&block)
|
17
15
|
@parser_proc = block
|
18
16
|
end
|
@@ -22,33 +20,23 @@ module Dolos
|
|
22
20
|
end
|
23
21
|
|
24
22
|
def run_with_state(state)
|
25
|
-
result = parser_proc.call(state)
|
26
|
-
if result.success?
|
27
|
-
state.last_success_position = state.input.offset
|
28
|
-
end
|
23
|
+
result = @parser_proc.call(state)
|
24
|
+
state.last_success_position = state.input.offset if result.success?
|
29
25
|
result
|
30
26
|
end
|
31
27
|
|
32
28
|
def capture!(wrap_in = nil)
|
33
29
|
Parser.new do |state|
|
34
30
|
result = run_with_state(state)
|
35
|
-
|
36
|
-
result.capture!(wrap_in)
|
37
|
-
else
|
38
|
-
result
|
39
|
-
end
|
31
|
+
result.success? ? result.capture!(wrap_in) : result
|
40
32
|
end
|
41
33
|
end
|
42
34
|
|
43
|
-
# Will call
|
35
|
+
# Will call `map` on captures
|
44
36
|
def map_captures(&block)
|
45
37
|
Parser.new do |state|
|
46
38
|
result = run_with_state(state)
|
47
|
-
|
48
|
-
Success.new(result.value, result.length, block.call(result.captures))
|
49
|
-
else
|
50
|
-
result
|
51
|
-
end
|
39
|
+
result.success? ? Success.new(result.value, result.length, block.call(result.captures)) : result
|
52
40
|
end
|
53
41
|
end
|
54
42
|
|
@@ -56,22 +44,18 @@ module Dolos
|
|
56
44
|
def map(&block)
|
57
45
|
Parser.new do |state|
|
58
46
|
result = run_with_state(state)
|
59
|
-
|
60
|
-
Success.new(block.call(result.value), result.length, result.captures)
|
61
|
-
else
|
62
|
-
result
|
63
|
-
end
|
47
|
+
result.success? ? Success.new(block.call(result.value), result.length, result.captures) : result
|
64
48
|
end
|
65
49
|
end
|
66
50
|
|
67
51
|
def combine(&block)
|
68
52
|
Parser.new do |state|
|
69
53
|
result = run_with_state(state)
|
54
|
+
|
70
55
|
if result.success?
|
56
|
+
state.input.advance(result.length)
|
71
57
|
new_parser = block.call(result.value, result.captures)
|
72
|
-
|
73
|
-
new_state.input.advance(result.length)
|
74
|
-
new_parser.run_with_state(new_state)
|
58
|
+
new_parser.run_with_state(state)
|
75
59
|
else
|
76
60
|
result
|
77
61
|
end
|
@@ -140,10 +124,9 @@ module Dolos
|
|
140
124
|
values = []
|
141
125
|
captures = []
|
142
126
|
count = 0
|
143
|
-
state.input.mark_offset
|
144
127
|
|
145
128
|
loop do
|
146
|
-
result = run_with_state(state.dup
|
129
|
+
result = run_with_state(state) # Removing .dup for performance. Be cautious of side effects.
|
147
130
|
|
148
131
|
if result.failure? || count >= n_max
|
149
132
|
break
|
@@ -155,7 +138,7 @@ module Dolos
|
|
155
138
|
count += 1
|
156
139
|
|
157
140
|
if separator && count < n_max
|
158
|
-
sep_result = separator.run_with_state(state.dup
|
141
|
+
sep_result = separator.run_with_state(state) # Removing .dup for performance. Be cautious of side effects.
|
159
142
|
break if sep_result.failure?
|
160
143
|
|
161
144
|
state.input.advance(sep_result.length)
|
@@ -163,10 +146,9 @@ module Dolos
|
|
163
146
|
end
|
164
147
|
|
165
148
|
if count < n_min
|
166
|
-
error_pos = state.input.offset
|
167
149
|
Failure.new(
|
168
|
-
"Expected parser to match at least #{n_min} times but matched only #{count} times",
|
169
|
-
|
150
|
+
-> { "Expected parser to match at least #{n_min} times but matched only #{count} times" },
|
151
|
+
state.input.offset,
|
170
152
|
state
|
171
153
|
)
|
172
154
|
else
|
@@ -174,7 +156,6 @@ module Dolos
|
|
174
156
|
end
|
175
157
|
end
|
176
158
|
end
|
177
|
-
|
178
159
|
def zero_or_more
|
179
160
|
repeat(n_min: 0, n_max: Float::INFINITY)
|
180
161
|
end
|
@@ -201,7 +182,6 @@ module Dolos
|
|
201
182
|
end
|
202
183
|
alias_method :opt, :optional
|
203
184
|
|
204
|
-
# Unstable API
|
205
185
|
# Used to declare lazy parser to avoid infinite loops in recursive parsers
|
206
186
|
def lazy
|
207
187
|
parser_memo = nil
|
@@ -212,11 +192,5 @@ module Dolos
|
|
212
192
|
end
|
213
193
|
end
|
214
194
|
|
215
|
-
private
|
216
|
-
|
217
|
-
def combine_and_discard_empty(*arrays)
|
218
|
-
arrays.compact.reject { |arr| arr.is_a?(Array) && arr.empty? }
|
219
|
-
end
|
220
|
-
|
221
195
|
end
|
222
196
|
end
|
@@ -1,5 +1,16 @@
|
|
1
1
|
module Dolos
|
2
2
|
module CommonParsers
|
3
|
+
def digit: -> Parser[String]
|
4
|
+
def digits: -> Parser[String]
|
5
|
+
|
6
|
+
def int: -> Parser[Integer]
|
7
|
+
|
8
|
+
def eol: -> Parser[String]
|
9
|
+
|
3
10
|
def ws: -> Parser[String]
|
11
|
+
def ws_rep0: -> Parser[String]
|
12
|
+
|
13
|
+
def alpha: -> Parser[String]
|
14
|
+
def alphanum: -> Parser[String]
|
4
15
|
end
|
5
16
|
end
|
data/sig/dolos/parser.rbs
CHANGED
@@ -4,16 +4,20 @@ module Dolos
|
|
4
4
|
def initialize: (^(ParserState) -> Result[A]) -> Parser[A]
|
5
5
|
def capture!: -> Parser[A]
|
6
6
|
def choice: [B](Parser[B])-> Parser[A | B]
|
7
|
+
def combine: [B](^(A, B) -> Parser[B]) -> Parser[B]
|
7
8
|
def flat_map: [B](Parser[A], ^(A) -> Parser[B]) -> Parser[B]
|
8
9
|
def flatten: -> Parser[A]
|
9
10
|
def map: [B](^(A) -> B) -> Parser[B]
|
10
|
-
def
|
11
|
+
def map_captures: [B](^(A) -> B) -> Parser[B]
|
11
12
|
def optional: -> Parser[A?]
|
12
13
|
def product: [B](Parser[A]) -> Parser[B]
|
14
|
+
def product_l: [B](Parser[B]) -> Parser[B]
|
15
|
+
def product_r: [B](Parser[B]) -> Parser[A]
|
13
16
|
def run: (String) -> Result[A]
|
14
17
|
def run_with_state: (ParserState) -> Result[A]
|
15
|
-
def repeat: (Integer, Integer)-> Parser[Array[A]]
|
18
|
+
def repeat: [B](Integer, Integer, Parser[B]?)-> Parser[Array[A]]
|
16
19
|
def zero_or_more: -> Parser[Array[A]]
|
17
20
|
def one_or_more: (Integer?) -> Parser[Array[A]]
|
21
|
+
def lazy: -> Parser[A]
|
18
22
|
end
|
19
23
|
end
|
data/sig/dolos/parser_state.rbs
CHANGED
data/sig/dolos/parsers.rbs
CHANGED
@@ -1,6 +1,10 @@
|
|
1
1
|
module Dolos
|
2
2
|
module Parsers
|
3
3
|
def any_char: -> Parser[String]
|
4
|
+
def char_in: -> Parser[String]
|
5
|
+
def char_while : -> Parser[String]
|
6
|
+
def recursive: [A,B,C]() { (Parser[A]) -> Parser[B] } -> Parser[C]
|
7
|
+
|
4
8
|
def regex: (Regexp) -> Parser[String]
|
5
9
|
def string: (String)-> Parser[String]
|
6
10
|
end
|
data/sig/dolos/result.rbs
CHANGED
@@ -15,6 +15,11 @@ module Dolos
|
|
15
15
|
end
|
16
16
|
|
17
17
|
class Failure < Result[bot]
|
18
|
+
@message_proc: ^-> String
|
19
|
+
@message_evaluated: bool
|
20
|
+
@message_value: String
|
21
|
+
@state: ParserState
|
22
|
+
|
18
23
|
attr_reader committed: bool
|
19
24
|
attr_reader error_position: Integer
|
20
25
|
attr_reader message: String
|
@@ -25,6 +30,8 @@ module Dolos
|
|
25
30
|
|
26
31
|
def map: [B](^(bot) -> B) -> Result[B]
|
27
32
|
|
33
|
+
def pretty_print: -> String
|
34
|
+
|
28
35
|
def success?: -> bool
|
29
36
|
end
|
30
37
|
end
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: dolos
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.2.
|
4
|
+
version: 0.2.1
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- benetis
|
8
8
|
autorequire:
|
9
9
|
bindir: exe
|
10
10
|
cert_chain: []
|
11
|
-
date: 2023-08-
|
11
|
+
date: 2023-08-22 00:00:00.000000000 Z
|
12
12
|
dependencies: []
|
13
13
|
description: Parser combinators library for Ruby. In active development, not stable
|
14
14
|
yet.
|
@@ -24,7 +24,9 @@ files:
|
|
24
24
|
- README.md
|
25
25
|
- Rakefile
|
26
26
|
- benchmarks/json/json.rb
|
27
|
-
- benchmarks/json/
|
27
|
+
- benchmarks/json/nested_json_166.json
|
28
|
+
- benchmarks/json/nested_json_1m.json
|
29
|
+
- benchmarks/letter.rb
|
28
30
|
- docs/dolos_stable_diff.png
|
29
31
|
- examples/letter.rb
|
30
32
|
- lib/dolos.rb
|
File without changes
|