vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

json: cap length of numbers to 15 digits before/after decimal point

Browse source 
(avoids infinite gen, e.g. "one third" -> `0.333333333333...`)
This commit is contained in:
ochafik 2024-04-08 20:14:36 +01:00
parent 07163fb627
commit dcf5d3283a
5 changed files with 1713 additions and 1654 deletions
Download patch file Download diff file Expand all files Collapse all files

8
common/json-schema-to-grammar.cpp
View file

@ -35,10 +35,14 @@ struct BuiltinRule {
std::vector<std::string> deps;
};
const std::string _up_to_15_digits = build_repetition("[0-9]", 15);
std::unordered_map<std::string, BuiltinRule> PRIMITIVE_RULES = {
{"boolean", {"(\"true\" | \"false\") space", {}}},
{"number", {"(\"-\"? ([0-9] | [1-9] [0-9]*)) (\".\" [0-9]+)? ([eE] [-+]? [0-9]+)? space", {}}},
{"integer", {"(\"-\"? ([0-9] | [1-9] [0-9]*)) space", {}}},
{"decimal-part", {"[0-9] " + _up_to_15_digits, {}}},
{"integral-part", {"[0-9] | [1-9] " + _up_to_15_digits, {}}},
{"number", {"(\"-\"? integral-part) (\".\" decimal-part)? ([eE] [-+]? integral-part)? space", {"integral-part", "decimal-part"}}},
{"integer", {"(\"-\"? integral-part) space", {"integral-part"}}},
{"value", {"object | array | string | number | boolean | null", {"object", "array", "string", "number", "boolean", "null"}}},
{"object", {"\"{\" space ( string \":\" space value (\",\" space string \":\" space value)* )? \"}\" space", {"string", "value"}}},
{"array", {"\"[\" space ( value (\",\" space value)* )? \"]\" space", {"value"}}},

8
examples/json_schema_to_grammar.py
View file

@ -17,14 +17,18 @@ class BuiltinRule:
self.content = content
self.deps = deps or []
_up_to_15_digits = _build_repetition('[0-9]', 15)
# whitespace is constrained to a single space char to prevent model "running away" in
# whitespace. Also maybe improves generation quality?
SPACE_RULE = '" "?'
PRIMITIVE_RULES = {
'boolean' : BuiltinRule('("true" | "false") space', []),
'number' : BuiltinRule('("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space', []),
'integer' : BuiltinRule('("-"? ([0-9] | [1-9] [0-9]*)) space', []),
'decimal-part' : BuiltinRule('[0-9] ' + _up_to_15_digits, []),
'integral-part': BuiltinRule('[0-9] | [1-9] ' + _up_to_15_digits, []),
'number' : BuiltinRule('("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space', ['integral-part', 'decimal-part']),
'integer' : BuiltinRule('("-"? integral-part) space', ['integral-part']),
'value' : BuiltinRule('object | array | string | number | boolean | null', ['object', 'array', 'string', 'number', 'boolean', 'null']),
'object' : BuiltinRule('"{" space ( string ":" space value ("," space string ":" space value)* )? "}" space', ['string', 'value']),
'array' : BuiltinRule('"[" space ( value ("," space value)* )? "]" space', ['value']),

3286
examples/server/json-schema-to-grammar.mjs.hpp
View file

File diff suppressed because it is too large Load diff

8
examples/server/public/json-schema-to-grammar.mjs
View file

@ -15,10 +15,14 @@ class BuiltinRule {
}
}
const UP_TO_15_DIGITS = _buildRepetition('[0-9]', 15);
const PRIMITIVE_RULES = {
boolean : new BuiltinRule('("true" | "false") space', []),
number : new BuiltinRule('("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space', []),
integer : new BuiltinRule('("-"? ([0-9] | [1-9] [0-9]*)) space', []),
'decimal-part' : new BuiltinRule('[0-9] ' + UP_TO_15_DIGITS, []),
'integral-part': new BuiltinRule('[0-9] | [1-9] ' + UP_TO_15_DIGITS, []),
number : new BuiltinRule('("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space', ['integral-part', 'decimal-part']),
integer : new BuiltinRule('("-"? integral-part) space', ['integral-part']),
value : new BuiltinRule('object | array | string | number | boolean | null', ['object', 'array', 'string', 'number', 'boolean', 'null']),
object : new BuiltinRule('"{" space ( string ":" space value ("," space string ":" space value)* )? "}" space', ['string', 'value']),
array : new BuiltinRule('"[" space ( value ("," space value)* )? "]" space', ['value']),

57
tests/test-json-schema-to-grammar.cpp
View file

@ -104,8 +104,10 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
R"""(
array ::= "[" space ( value ("," space value)* )? "]" space
boolean ::= ("true" | "false") space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
null ::= "null" space
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
object ::= "{" space ( string ":" space value ("," space string ":" space value)* )? "}" space
root ::= object
space ::= " "?
@ -178,7 +180,8 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
"type": "integer"
})""",
R"""(
root ::= ("-"? ([0-9] | [1-9] [0-9]*)) space
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
root ::= ("-"? integral-part) space
space ::= " "?
)"""
});
@ -242,7 +245,9 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
"prefixItems": [{ "type": "string" }, { "type": "number" }]
})""",
R"""(
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
root ::= "[" space string "," space number "]" space
space ::= " "?
string ::= "\"" (
@ -259,7 +264,9 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
"type": "number"
})""",
R"""(
root ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
root ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
space ::= " "?
)"""
});
@ -323,9 +330,11 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
"maxItems": 5
})""",
R"""(
integer ::= ("-"? ([0-9] | [1-9] [0-9]*)) space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integer ::= ("-"? integral-part) space
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
item ::= number | integer
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
root ::= "[" space item ( "," space item )( "," space item )(( "," space item ) (( "," space item ))?)? "]" space
space ::= " "?
)"""
@ -503,7 +512,9 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
additional-kv ::= string ":" space additional-value
additional-kvs ::= additional-kv ( "," space additional-kv )*
additional-value ::= "[" space ( number ( "," space number )* )? "]" space
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
root ::= "{" space (additional-kvs )? "}" space
space ::= " "?
string ::= "\"" (
@ -523,8 +534,10 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
R"""(
array ::= "[" space ( value ("," space value)* )? "]" space
boolean ::= ("true" | "false") space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
null ::= "null" space
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
object ::= "{" space ( string ":" space value ("," space string ":" space value)* )? "}" space
root ::= object
space ::= " "?
@ -545,8 +558,10 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
R"""(
array ::= "[" space ( value ("," space value)* )? "]" space
boolean ::= ("true" | "false") space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
null ::= "null" space
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
object ::= "{" space ( string ":" space value ("," space string ":" space value)* )? "}" space
root ::= object
space ::= " "?
@ -586,7 +601,9 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
a-kv ::= "\"a\"" space ":" space number
additional-kv ::= string ":" space string
additional-kvs ::= additional-kv ( "," space additional-kv )*
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
root ::= "{" space a-kv ( "," space ( additional-kvs ) )? "}" space
space ::= " "?
string ::= "\"" (
@ -611,7 +628,9 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
a-rest ::= additional-kvs
additional-kv ::= string ":" space number
additional-kvs ::= additional-kv ( "," space additional-kv )*
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
root ::= "{" space (a-kv a-rest | additional-kvs )? "}" space
space ::= " "?
string ::= "\"" (
@ -639,7 +658,9 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
additional-kvs ::= additional-kv ( "," space additional-kv )*
b-kv ::= "\"b\"" space ":" space number
b-rest ::= additional-kvs
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
root ::= "{" space a-kv ( "," space ( b-kv b-rest | additional-kvs ) )? "}" space
space ::= " "?
string ::= "\"" (
@ -704,9 +725,11 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
alternative-1 ::= bar
bar ::= "{" space (bar-b-kv )? "}" space
bar-b-kv ::= "\"b\"" space ":" space number
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
foo ::= "{" space (foo-a-kv )? "}" space
foo-a-kv ::= "\"a\"" space ":" space number
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
root ::= alternative-0 | alternative-1
space ::= " "?
)"""
@ -748,7 +771,9 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
c-kv ::= "\"c\"" space ":" space number
d-kv ::= "\"d\"" space ":" space number
d-rest ::= ( "," space c-kv )?
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
root ::= "{" space a-kv "," space b-kv ( "," space ( d-kv d-rest | c-kv ) )? "}" space
space ::= " "?
)"""
@ -789,7 +814,9 @@ static void test_all(const std::string & lang, std::function<void(const TestCase
"definitions": {}
})""",
R"""(
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? space
decimal-part ::= [0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
integral-part ::= [0-9] | [1-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9] ([0-9])?)?)?)?)?)?)?)?)?)?)?)?)?)?)?
number ::= ("-"? integral-part) ("." decimal-part)? ([eE] [-+]? integral-part)? space
number- ::= "{" space number-number-kv "}" space
number-kv ::= "\"number\"" space ":" space number-
number-number ::= "{" space number-number-root-kv "}" space