refactor: break down json grammar parser in different files (#3004)

* refactor: break down json grammar parser in different files

Signed-off-by: Ettore Di Giacinto <[email protected]>

* fix: patch to `refactor_grammars` - propagate errors (#3006)

propagate errors around

Signed-off-by: Dave Lee <[email protected]>

---------

Signed-off-by: Ettore Di Giacinto <[email protected]>
Signed-off-by: Dave Lee <[email protected]>
Co-authored-by: Dave <[email protected]>

core/http/endpoints/openai/chat.go

@@ -226,9 +226,15 @@ func ChatEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, startup
 
 			// Update input grammar
 			jsStruct := funcs.ToJSONStructure(config.FunctionsConfig.FunctionNameKey, config.FunctionsConfig.FunctionNameKey)
-			config.Grammar = jsStruct.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
+			g, err := jsStruct.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
+			if err == nil {
+				config.Grammar = g
+			}
 		case input.JSONFunctionGrammarObject != nil:
-			config.Grammar = input.JSONFunctionGrammarObject.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
+			g, err := input.JSONFunctionGrammarObject.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
+			if err == nil {
+				config.Grammar = g
+			}
 		default:
 			// Force picking one of the functions by the request
 			if config.FunctionToCall() != "" {

pkg/functions/bnf_rules.go

@@ -0,0 +1,47 @@
+package functions
+
+import "regexp"
+
+var (
+	PRIMITIVE_RULES = map[string]string{
+		"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`,
+		"string": `"\"" (
+			[^"\\] |
+			"\\" (["\\/bfnrt] | "u" [0-9a-fA-F] [0-9a-fA-F] [0-9a-fA-F] [0-9a-fA-F])
+		  )* "\"" space`,
+		// TODO: we shouldn't forbid \" and \\ or all unicode and have this branch here,
+		// however, if we don't have it, the grammar will be ambiguous and
+		// empirically results are way worse.
+		"freestring": `(
+			[^\x00] |
+			"\\" (["\\/bfnrt] | "u" [0-9a-fA-F] [0-9a-fA-F] [0-9a-fA-F] [0-9a-fA-F])
+		  )* space`,
+		"null": `"null" space`,
+	}
+
+	INVALID_RULE_CHARS_RE     = regexp.MustCompile(`[^a-zA-Z0-9-]+`)
+	GRAMMAR_LITERAL_ESCAPE_RE = regexp.MustCompile(`[\r\n"]`)
+	GRAMMAR_LITERAL_ESCAPES   = map[string]string{
+		"\r": `\r`,
+		"\n": `\n`,
+		`"`:  `\"`,
+	}
+)
+
+const (
+	SPACE_RULE = `" "?`
+
+	arrayNewLines = `arr  ::=
+  "[\n"  (
+		realvalue
+    (",\n"  realvalue)*
+  )? "]"`
+
+	array = `arr  ::=
+  "["  (
+		realvalue
+    (","  realvalue)*
+  )? "]"`
+)

pkg/functions/function_structure.go

@@ -0,0 +1,25 @@
+package functions
+
+import "encoding/json"
+
+type Item struct {
+	Type       string                 `json:"type"`
+	Properties map[string]interface{} `json:"properties"`
+}
+
+type JSONFunctionStructure struct {
+	OneOf []Item                 `json:"oneOf,omitempty"`
+	AnyOf []Item                 `json:"anyOf,omitempty"`
+	Defs  map[string]interface{} `json:"$defs,omitempty"`
+}
+
+func (j JSONFunctionStructure) Grammar(options ...func(*GrammarOption)) (string, error) {
+	grammarOpts := &GrammarOption{}
+	grammarOpts.Apply(options...)
+
+	dat, err := json.Marshal(j)
+	if err != nil {
+		return "", err
+	}
+	return NewJSONSchemaConverter(grammarOpts.PropOrder).GrammarFromBytes(dat, options...)
+}

pkg/functions/functions.go

@@ -18,6 +18,15 @@ type Function struct {
 }
 type Functions []Function
 
+type FunctionName struct {
+	Const string `json:"const"`
+}
+
+type Argument struct {
+	Type       string                 `json:"type"`
+	Properties map[string]interface{} `json:"properties"`
+}
+
 type Tool struct {
 	Type     string   `json:"type"`
 	Function Function `json:"function,omitempty"`
@@ -86,3 +95,11 @@ func (f Functions) Select(name string) Functions {
 
 	return funcs
 }
+
+func jsonString(v interface{}) (string, error) {
+	b, err := json.Marshal(v)
+	if err != nil {
+		return "", err
+	}
+	return string(b), nil
+}

pkg/functions/functions_suite_test.go

@@ -1,8 +1,10 @@
-package functions
+package functions_test
 
 import (
 	"testing"
 
+	. "github.com/mudler/LocalAI/pkg/functions"
+
 	. "github.com/onsi/ginkgo/v2"
 	. "github.com/onsi/gomega"
 )
@@ -11,3 +13,13 @@ func TestGrammar(t *testing.T) {
 	RegisterFailHandler(Fail)
 	RunSpecs(t, "Grammar test suite")
 }
+
+func createFunction(field1 string, field2 string, name string, properties map[string]interface{}) map[string]interface{} {
+	property := map[string]interface{}{}
+	property[field1] = FunctionName{Const: name}
+	property[field2] = Argument{
+		Type:       "object",
+		Properties: properties,
+	}
+	return property
+}