Add SKIP and SKIP REGEXP support for JSON type parameters

This adds proper parsing and formatting for SKIP clauses in JSON type
parameters, including:
- SKIP path (e.g., SKIP a.b for dotted paths)
- SKIP REGEXP 'pattern' (for regex-based path matching)

These are used in JSON type casts like:
  json::JSON(SKIP a.b, max_dynamic_paths=2)
  json::JSON(SKIP REGEXP '.*a.*', max_dynamic_paths=2)

Author: claude

internal/explain/format.go

@@ -237,6 +237,19 @@ func FormatDataType(dt *ast.DataType) string {
 		} else if binExpr, ok := p.(*ast.BinaryExpr); ok {
 			// Binary expression (e.g., 'hello' = 1 for Enum types)
 			params = append(params, formatBinaryExprForType(binExpr))
+		} else if fn, ok := p.(*ast.FunctionCall); ok {
+			// Function call (e.g., SKIP for JSON types)
+			if fn.Name == "SKIP" && len(fn.Arguments) > 0 {
+				if ident, ok := fn.Arguments[0].(*ast.Identifier); ok {
+					params = append(params, "SKIP "+ident.Name())
+				}
+			} else if fn.Name == "SKIP REGEXP" && len(fn.Arguments) > 0 {
+				if lit, ok := fn.Arguments[0].(*ast.Literal); ok {
+					params = append(params, fmt.Sprintf("SKIP REGEXP \\\\\\'%s\\\\\\'", lit.Value))
+				}
+			} else {
+				params = append(params, fmt.Sprintf("%v", p))
+			}
 		} else {
 			params = append(params, fmt.Sprintf("%v", p))
 		}

parser/parser.go

@@ -3266,6 +3266,64 @@ func (p *Parser) parseDataType() *ast.DataType {
 
 		// Parse type parameters, but stop on keywords that can't be part of type params
 		for !p.currentIs(token.RPAREN) && !p.currentIs(token.EOF) && !p.currentIs(token.COLLATE) {
+			var param ast.Expression
+
+			// Special handling for SKIP in JSON/OBJECT types: SKIP path or SKIP REGEXP 'pattern'
+			if isObjectType && (p.currentIs(token.IDENT) || p.current.Token.IsKeyword()) && strings.ToUpper(p.current.Value) == "SKIP" {
+				pos := p.current.Pos
+				p.nextToken() // consume SKIP
+
+				// Check for SKIP REGEXP 'pattern'
+				if p.currentIs(token.REGEXP) || (p.currentIs(token.IDENT) && strings.ToUpper(p.current.Value) == "REGEXP") {
+					p.nextToken() // consume REGEXP
+					// Parse the pattern string
+					if p.currentIs(token.STRING) {
+						pattern := p.current.Value
+						p.nextToken()
+						param = &ast.FunctionCall{
+							Position:  pos,
+							Name:      "SKIP REGEXP",
+							Arguments: []ast.Expression{&ast.Literal{Position: pos, Value: pattern, Type: ast.LiteralString}},
+						}
+					}
+				} else {
+					// Parse dotted path: a, a.b, a.b.c, etc.
+					var pathParts []string
+					for {
+						if p.currentIs(token.IDENT) || p.current.Token.IsKeyword() {
+							pathParts = append(pathParts, p.current.Value)
+							p.nextToken()
+						}
+						if p.currentIs(token.DOT) {
+							p.nextToken() // consume dot
+						} else {
+							break
+						}
+					}
+					if len(pathParts) > 0 {
+						param = &ast.FunctionCall{
+							Position:  pos,
+							Name:      "SKIP",
+							Arguments: []ast.Expression{&ast.Identifier{Position: pos, Parts: pathParts}},
+						}
+					}
+				}
+				// Wrap in ObjectTypeArgument
+				if param != nil {
+					param = &ast.ObjectTypeArgument{
+						Position: param.Pos(),
+						Expr:     param,
+					}
+					dt.Parameters = append(dt.Parameters, param)
+				}
+				if p.currentIs(token.COMMA) {
+					p.nextToken()
+				} else {
+					break
+				}
+				continue
+			}
+
 			// Check if this is a named parameter: identifier followed by a type name
 			// e.g., "a UInt32" where "a" is the name and "UInt32" is the type
 			isNamedParam := false
@@ -3291,7 +3349,6 @@ func (p *Parser) parseDataType() *ast.DataType {
 				}
 			}
 
-			var param ast.Expression
 			if isNamedParam {
 				// Parse as name + type pair
 				pos := p.current.Pos

parser/testdata/00564_temporary_table_management/metadata.json

@@ -1,7 +1,6 @@
 {
   "explain_todo": {
     "stmt3": true,
-    "stmt4": true,
     "stmt5": true,
     "stmt7": true
   }

parser/testdata/00688_low_cardinality_nullable_cast/metadata.json

@@ -1,5 +1 @@
-{
-  "explain_todo": {
-    "stmt6": true
-  }
-}
+{}

parser/testdata/01095_tpch_like_smoke/metadata.json

@@ -1,16 +1 @@
-{
-  "explain_todo": {
-    "stmt20": true,
-    "stmt24": true,
-    "stmt26": true,
-    "stmt28": true,
-    "stmt30": true,
-    "stmt32": true,
-    "stmt34": true,
-    "stmt38": true,
-    "stmt42": true,
-    "stmt46": true,
-    "stmt49": true,
-    "stmt61": true
-  }
-}
+{}

parser/testdata/01268_data_numeric_parameters/metadata.json

@@ -1,5 +1 @@
-{
-  "explain_todo": {
-    "stmt4": true
-  }
-}
+{}

parser/testdata/01602_temporary_table_in_system_tables/metadata.json

@@ -1,6 +1 @@
-{
-  "explain_todo": {
-    "stmt7": true,
-    "stmt8": true
-  }
-}
+{}

parser/testdata/02184_default_table_engine/metadata.json

@@ -1,11 +1,9 @@
 {
   "explain_todo": {
     "stmt107": true,
-    "stmt114": true,
     "stmt26": true,
     "stmt56": true,
     "stmt61": true,
-    "stmt69": true,
     "stmt73": true
   }
 }

parser/testdata/02969_analyzer_eliminate_injective_functions/metadata.json

@@ -1,6 +1 @@
-{
-  "explain_todo": {
-    "stmt3": true,
-    "stmt4": true
-  }
-}
+{}

parser/testdata/02997_projections_formatting/metadata.json

@@ -1,6 +1 @@
-{
-  "explain_todo": {
-    "stmt2": true,
-    "stmt4": true
-  }
-}
+{}