feat: Lucene query parser with sql and dynamodb support

go.mod

@@ -11,6 +11,7 @@ require (
 	github.com/aws/aws-sdk-go-v2/credentials v1.19.12
 	github.com/aws/aws-sdk-go-v2/service/dynamodb v1.56.2
 	github.com/aws/aws-sdk-go-v2/service/sns v1.39.11
+	github.com/grindlemire/go-lucene v0.0.26
 	gopkg.in/yaml.v3 v3.0.1
 	gorm.io/gorm v1.31.1
 )

go.sum

@@ -75,6 +75,8 @@ github.com/google/go-cmp v0.7.0 h1:wk8382ETsv4JYUZwIsn6YpYiWiBsYLSJiTsyBybVuN8=
 github.com/google/go-cmp v0.7.0/go.mod h1:pXiqmnSA92OHEEa9HXL2W4E7lf9JzCmGVUdgjX3N/iU=
 github.com/google/uuid v1.6.0 h1:NIvaJDMOsjHA8n1jAhLSgzrAzy1Hgr+hNrb57e+94F0=
 github.com/google/uuid v1.6.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
+github.com/grindlemire/go-lucene v0.0.26 h1:81ttZkMvU3rFD0TfmjdIZT2U0Fd4TT7buDy+xq1x5EQ=
+github.com/grindlemire/go-lucene v0.0.26/go.mod h1:INRJBdhkLjS4jc7XgkGPfzC5wuFg3BHDukXMTc+OTbc=
 github.com/jackc/pgpassfile v1.0.0 h1:/6Hmqy13Ss2zCq62VdNG8tM1wchn8zjSGOBJ6icpsIM=
 github.com/jackc/pgpassfile v1.0.0/go.mod h1:CEx0iS5ambNFdcRtxPj5JhEz+xB6uRky5eyVu/W2HEg=
 github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 h1:iCEnooe7UlwOQYpKFhBabPMi4aNAfoODPEFNiAnClxo=

storage/dynamodb.go

@@ -241,13 +241,19 @@ func (s *DynamoDBAdapter) Search(dest any, sortKey string, query string, limit i
 	if err := validateSortKey(sortKey); err != nil {
 		return "", err
 	}
-	return s.executePaginatedQuery(dest, limit, cursor, func(input *dynamodb.ExecuteStatementInput) *dynamodb.ExecuteStatementInput {
-		// Parse Lucene query
-		destType := reflect.TypeOf(dest).Elem().Elem()
-		model := reflect.New(destType).Elem().Interface()
-		parser, _ := lucene.NewParserFromType(model)
-		whereClause, params, _ := parser.ParseToDynamoDBPartiQL(query)
 
+	destType := reflect.TypeOf(dest).Elem().Elem()
+	model := reflect.New(destType).Elem().Interface()
+	parser, err := lucene.NewParser(model)
+	if err != nil {
+		return "", err
+	}
+	whereClause, dynamoParams, err := parser.ParseToDynamoDBPartiQL(query)
+	if err != nil {
+		return "", err
+	}
+
+	return s.executePaginatedQuery(dest, limit, cursor, func(input *dynamodb.ExecuteStatementInput) *dynamodb.ExecuteStatementInput {
 		// Build query
 		query := fmt.Sprintf(`SELECT * FROM "%s"`, s.getTableName(dest))
 		if whereClause != "" {
@@ -258,7 +264,7 @@ func (s *DynamoDBAdapter) Search(dest any, sortKey string, query string, limit i
 		}
 
 		input.Statement = aws.String(query)
-		input.Parameters = params
+		input.Parameters = dynamoParams
 		return input
 	})
 }

storage/search/lucene/dynamodb_driver.go

@@ -0,0 +1,137 @@
+package lucene
+
+import (
+	"fmt"
+	"regexp"
+	"strings"
+
+	"github.com/aws/aws-sdk-go-v2/service/dynamodb/types"
+	"github.com/grindlemire/go-lucene/pkg/driver"
+	"github.com/grindlemire/go-lucene/pkg/lucene/expr"
+)
+
+// DynamoDBPartiQLDriver converts Lucene queries to DynamoDB PartiQL.
+type DynamoDBPartiQLDriver struct {
+	driver.Base
+	fields map[string]FieldInfo
+}
+
+func NewDynamoDBDriver(fields []FieldInfo) (*DynamoDBPartiQLDriver, error) {
+	fieldMap, err := buildFieldMap(fields)
+	if err != nil {
+		return nil, err
+	}
+
+	fns := map[expr.Operator]driver.RenderFN{
+		expr.Literal:   driver.Shared[expr.Literal],
+		expr.And:       driver.Shared[expr.And],
+		expr.Or:        driver.Shared[expr.Or],
+		expr.Not:       driver.Shared[expr.Not],
+		expr.Equals:    driver.Shared[expr.Equals],
+		expr.Range:     driver.Shared[expr.Range],
+		expr.Must:      driver.Shared[expr.Must],
+		expr.MustNot:   driver.Shared[expr.MustNot],
+		expr.Wild:      driver.Shared[expr.Wild],
+		expr.Regexp:    driver.Shared[expr.Regexp],
+		expr.Like:      dynamoDBLike, // Custom LIKE for DynamoDB functions
+		expr.Greater:   driver.Shared[expr.Greater],
+		expr.GreaterEq: driver.Shared[expr.GreaterEq],
+		expr.Less:      driver.Shared[expr.Less],
+		expr.LessEq:    driver.Shared[expr.LessEq],
+		expr.In:        driver.Shared[expr.In],
+		expr.List:      driver.Shared[expr.List],
+	}
+
+	return &DynamoDBPartiQLDriver{
+		Base: driver.Base{
+			RenderFNs: fns,
+		},
+		fields: fieldMap,
+	}, nil
+}
+
+// RenderPartiQL renders the expression to DynamoDB PartiQL with AttributeValue parameters.
+func (d *DynamoDBPartiQLDriver) RenderPartiQL(e *expr.Expression) (string, []types.AttributeValue, error) {
+	// Use base rendering with ? placeholders
+	str, params, err := d.RenderParam(e)
+	if err != nil {
+		return "", nil, err
+	}
+
+	// Convert params to DynamoDB AttributeValues
+	attrValues := make([]types.AttributeValue, len(params))
+	for i, param := range params {
+		attrValues[i] = &types.AttributeValueMemberS{Value: fmt.Sprintf("%v", param)}
+	}
+
+	return str, attrValues, nil
+}
+
+// escapePartiQLString escapes a string value for safe use in PartiQL string literals.
+// Escapes single quotes by doubling them (PartiQL standard).
+func escapePartiQLString(s string) string {
+	return strings.ReplaceAll(s, "'", "''")
+}
+
+var (
+	// partiQLIdentifierPattern matches valid PartiQL identifiers (alphanumeric and underscore only)
+	partiQLIdentifierPattern = regexp.MustCompile(`^[a-zA-Z0-9_]+$`)
+)
+
+// escapePartiQLIdentifier escapes a field name for safe use in PartiQL.
+// Validates that the identifier contains only safe characters (alphanumeric, underscore).
+// Returns error if identifier contains potentially dangerous characters.
+func escapePartiQLIdentifier(identifier string) (string, error) {
+	if !partiQLIdentifierPattern.MatchString(identifier) {
+		return "", fmt.Errorf("invalid identifier: contains unsafe characters (only alphanumeric and underscore allowed)")
+	}
+	return identifier, nil
+}
+
+// unquotePartiQLString safely removes surrounding quotes from a PartiQL string literal.
+// Handles already-escaped quotes correctly.
+func unquotePartiQLString(s string) string {
+	if len(s) >= 2 && s[0] == '\'' && s[len(s)-1] == '\'' {
+		return s[1 : len(s)-1]
+	}
+	return s
+}
+
+// dynamoDBLike implements LIKE using DynamoDB's begins_with and contains functions.
+func dynamoDBLike(left, right string) (string, error) {
+	// Validate and escape field name (left)
+	safeLeft, err := escapePartiQLIdentifier(left)
+	if err != nil {
+		return "", fmt.Errorf("invalid field name: %w", err)
+	}
+
+	// Extract the raw value from the right side (remove quotes if present)
+	rawValue := unquotePartiQLString(right)
+
+	// Analyze pattern for wildcards
+	hasPrefix := strings.HasPrefix(rawValue, "%")
+	hasSuffix := strings.HasSuffix(rawValue, "%")
+
+	if hasPrefix && hasSuffix {
+		// %value% -> contains(field, value)
+		value := strings.Trim(rawValue, "%")
+		escapedValue := escapePartiQLString(value)
+		return fmt.Sprintf("contains(%s, '%s')", safeLeft, escapedValue), nil
+	}
+	if !hasPrefix && hasSuffix {
+		// value% -> begins_with(field, value)
+		value := strings.TrimSuffix(rawValue, "%")
+		escapedValue := escapePartiQLString(value)
+		return fmt.Sprintf("begins_with(%s, '%s')", safeLeft, escapedValue), nil
+	}
+	if hasPrefix && !hasSuffix {
+		// %value -> contains(field, value) (DynamoDB doesn't have ends_with)
+		value := strings.TrimPrefix(rawValue, "%")
+		escapedValue := escapePartiQLString(value)
+		return fmt.Sprintf("contains(%s, '%s')", safeLeft, escapedValue), nil
+	}
+
+	// Exact match - escape the value and wrap in quotes
+	escapedValue := escapePartiQLString(rawValue)
+	return fmt.Sprintf("%s = '%s'", safeLeft, escapedValue), nil
+}