pattern-ruby 0.1.1 → 0.2.0
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 +4 -4
- data/lib/pattern_ruby/conversation.rb +15 -3
- data/lib/pattern_ruby/pattern_compiler.rb +60 -8
- data/lib/pattern_ruby/pipeline.rb +12 -0
- data/lib/pattern_ruby/version.rb +1 -1
- metadata +1 -1
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 3cf30e8485bb83eb27f583e82c6e54c6556ded68150104ee2a35f95032af655d
|
|
4
|
+
data.tar.gz: 27e26d1dd70b4f89457281e7234b95c766a883f71ce76bbbef4fe6d0c5838a1b
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 26f94c3c765863235125bbedf9a1c6f460a2691744665001b2e91245f324f1976fd6c8a20bd0e2342651c2d8c7122d8ebdc2f4a46ca2d2e35a2ac07a9005f526
|
|
7
|
+
data.tar.gz: fb5b161865d3c210c3e5b8692bb7472fc0995e26f45553c20b039d5db8cbbfbe5cba45be99508b3eaf86db64a8474cbf1dc531b299fb32cd98741611d2f52c33
|
|
@@ -75,12 +75,24 @@ module PatternRuby
|
|
|
75
75
|
return false unless intent
|
|
76
76
|
|
|
77
77
|
# Find entity definitions that don't have a slot value yet
|
|
78
|
-
unfilled = intent.entity_definitions.select { |
|
|
78
|
+
unfilled = intent.entity_definitions.select { |k, _| !@slots.key?(k) || @slots[k].nil? }
|
|
79
79
|
return false if unfilled.empty?
|
|
80
80
|
|
|
81
81
|
# Try the input as a value for the first unfilled slot
|
|
82
|
-
|
|
83
|
-
|
|
82
|
+
slot_name, defn = unfilled.first
|
|
83
|
+
value = input.strip
|
|
84
|
+
|
|
85
|
+
# Validate against entity type constraints if defined
|
|
86
|
+
if defn
|
|
87
|
+
if defn[:values] && !defn[:values].include?(value)
|
|
88
|
+
return false
|
|
89
|
+
end
|
|
90
|
+
if defn[:pattern] && !value.match?(defn[:pattern])
|
|
91
|
+
return false
|
|
92
|
+
end
|
|
93
|
+
end
|
|
94
|
+
|
|
95
|
+
@slots[slot_name] = value
|
|
84
96
|
true
|
|
85
97
|
end
|
|
86
98
|
end
|
|
@@ -24,49 +24,88 @@ module PatternRuby
|
|
|
24
24
|
@entity_registry = entity_registry
|
|
25
25
|
end
|
|
26
26
|
|
|
27
|
-
def
|
|
27
|
+
def self.validate!(pattern_string)
|
|
28
|
+
new.validate!(pattern_string)
|
|
29
|
+
end
|
|
30
|
+
|
|
31
|
+
def validate!(pattern_string)
|
|
28
32
|
raise ArgumentError, "pattern must be a String, got #{pattern_string.class}" unless pattern_string.is_a?(String)
|
|
29
33
|
raise ArgumentError, "pattern cannot be nil or empty" if pattern_string.nil? || pattern_string.strip.empty?
|
|
30
34
|
if pattern_string.length > MAX_PATTERN_LENGTH
|
|
31
35
|
raise ArgumentError, "pattern exceeds maximum length of #{MAX_PATTERN_LENGTH} characters"
|
|
32
36
|
end
|
|
33
37
|
|
|
38
|
+
# Check for unbalanced brackets
|
|
39
|
+
check_balanced(pattern_string, "[", "]", "square brackets")
|
|
40
|
+
check_balanced(pattern_string, "(", ")", "parentheses")
|
|
41
|
+
check_balanced(pattern_string, "{", "}", "curly braces")
|
|
42
|
+
|
|
43
|
+
# Check for empty entity names
|
|
44
|
+
if pattern_string.match?(/\{\s*\}/)
|
|
45
|
+
raise ArgumentError, "empty entity name {} in pattern"
|
|
46
|
+
end
|
|
47
|
+
if pattern_string.match?(/\{\s*:/)
|
|
48
|
+
raise ArgumentError, "entity name cannot start with ':' in pattern"
|
|
49
|
+
end
|
|
50
|
+
|
|
51
|
+
true
|
|
52
|
+
end
|
|
53
|
+
|
|
54
|
+
def compile(pattern_string)
|
|
55
|
+
validate!(pattern_string)
|
|
56
|
+
|
|
34
57
|
tokens = tokenize(pattern_string)
|
|
35
58
|
entity_names = []
|
|
36
59
|
literal_count = 0
|
|
37
60
|
regex_parts = []
|
|
61
|
+
optional_flags = []
|
|
38
62
|
|
|
39
63
|
tokens.each do |token|
|
|
40
64
|
case token
|
|
41
65
|
when EntityToken
|
|
42
66
|
entity_names << token.name.to_sym
|
|
43
67
|
regex_parts << build_entity_regex(token)
|
|
68
|
+
optional_flags << false
|
|
44
69
|
when OptionalToken
|
|
45
70
|
inner = compile_inner(token.content)
|
|
46
71
|
entity_names.concat(inner[:entity_names])
|
|
47
|
-
regex_parts <<
|
|
72
|
+
regex_parts << inner[:regex]
|
|
73
|
+
optional_flags << true
|
|
48
74
|
when AlternationToken
|
|
49
75
|
alts = token.alternatives.map { |a| Regexp.escape(a) }
|
|
50
76
|
regex_parts << "(?:#{alts.join('|')})"
|
|
51
77
|
literal_count += 1
|
|
78
|
+
optional_flags << false
|
|
52
79
|
when WildcardToken
|
|
53
80
|
regex_parts << "(.+)"
|
|
81
|
+
optional_flags << false
|
|
54
82
|
when LiteralToken
|
|
55
83
|
regex_parts << Regexp.escape(token.text)
|
|
56
84
|
literal_count += 1
|
|
85
|
+
optional_flags << false
|
|
57
86
|
end
|
|
58
87
|
end
|
|
59
88
|
|
|
60
89
|
token_count = tokens.size
|
|
61
90
|
entity_count = entity_names.size
|
|
62
91
|
|
|
63
|
-
# Join with \s+
|
|
92
|
+
# Join parts with \s+ separators, wrapping optional parts in (?:...)?
|
|
64
93
|
regex_str = +""
|
|
94
|
+
need_sep = false
|
|
65
95
|
regex_parts.each_with_index do |part, i|
|
|
66
|
-
if i
|
|
67
|
-
|
|
96
|
+
if optional_flags[i]
|
|
97
|
+
if i == 0
|
|
98
|
+
regex_str << "(?:#{part}\\s+)?"
|
|
99
|
+
need_sep = false
|
|
100
|
+
else
|
|
101
|
+
regex_str << "(?:\\s+#{part})?"
|
|
102
|
+
need_sep = true
|
|
103
|
+
end
|
|
104
|
+
else
|
|
105
|
+
regex_str << "\\s+" if need_sep
|
|
106
|
+
regex_str << part
|
|
107
|
+
need_sep = true
|
|
68
108
|
end
|
|
69
|
-
regex_str << part
|
|
70
109
|
end
|
|
71
110
|
|
|
72
111
|
regex = Regexp.new("\\A\\s*#{regex_str}\\s*\\z", Regexp::IGNORECASE)
|
|
@@ -97,11 +136,14 @@ module PatternRuby
|
|
|
97
136
|
scanner.skip(/\s+/)
|
|
98
137
|
break if scanner.eos?
|
|
99
138
|
|
|
100
|
-
if scanner.scan(/\{([^}]
|
|
139
|
+
if scanner.scan(/\{([^}]*)\}/)
|
|
101
140
|
# Entity: {name} or {name:constraint}
|
|
102
141
|
content = scanner[1]
|
|
142
|
+
raise ArgumentError, "empty entity name {} in pattern" if content.strip.empty?
|
|
103
143
|
name, constraint = content.split(":", 2)
|
|
104
|
-
|
|
144
|
+
name = name.strip
|
|
145
|
+
raise ArgumentError, "empty entity name in pattern" if name.empty?
|
|
146
|
+
tokens << EntityToken.new(name, constraint&.strip)
|
|
105
147
|
elsif scanner.scan(/\[/)
|
|
106
148
|
# Optional: [content]
|
|
107
149
|
depth = 1
|
|
@@ -159,6 +201,16 @@ module PatternRuby
|
|
|
159
201
|
end
|
|
160
202
|
end
|
|
161
203
|
|
|
204
|
+
def check_balanced(str, open_char, close_char, name)
|
|
205
|
+
depth = 0
|
|
206
|
+
str.each_char do |c|
|
|
207
|
+
depth += 1 if c == open_char
|
|
208
|
+
depth -= 1 if c == close_char
|
|
209
|
+
raise ArgumentError, "unbalanced #{name} in pattern: #{str.inspect}" if depth < 0
|
|
210
|
+
end
|
|
211
|
+
raise ArgumentError, "unbalanced #{name} in pattern: #{str.inspect}" if depth != 0
|
|
212
|
+
end
|
|
213
|
+
|
|
162
214
|
def compile_inner(content)
|
|
163
215
|
inner_tokens = tokenize(content)
|
|
164
216
|
entity_names = []
|
|
@@ -2,6 +2,7 @@ module PatternRuby
|
|
|
2
2
|
class Pipeline
|
|
3
3
|
def initialize(&block)
|
|
4
4
|
@steps = []
|
|
5
|
+
@error_handler = nil
|
|
5
6
|
instance_eval(&block) if block
|
|
6
7
|
end
|
|
7
8
|
|
|
@@ -9,10 +10,21 @@ module PatternRuby
|
|
|
9
10
|
@steps << { name: name, handler: block }
|
|
10
11
|
end
|
|
11
12
|
|
|
13
|
+
def on_error(&block)
|
|
14
|
+
@error_handler = block
|
|
15
|
+
end
|
|
16
|
+
|
|
12
17
|
def process(input)
|
|
13
18
|
result = input
|
|
14
19
|
@steps.each do |s|
|
|
15
20
|
result = s[:handler].call(result)
|
|
21
|
+
rescue => e
|
|
22
|
+
if @error_handler
|
|
23
|
+
result = @error_handler.call(e, s[:name], result)
|
|
24
|
+
break if result.nil?
|
|
25
|
+
else
|
|
26
|
+
raise
|
|
27
|
+
end
|
|
16
28
|
end
|
|
17
29
|
result
|
|
18
30
|
end
|
data/lib/pattern_ruby/version.rb
CHANGED