yy_parser.go

// Copyright 2015 PingCAP, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// See the License for the specific language governing permissions and
// limitations under the License.

package parser

import (
	"math"
	"regexp"
	"strconv"
	"unicode"

	"github.com/juju/errors"
	"github.com/pingcap/tidb/ast"
	"github.com/pingcap/tidb/mysql"
	"github.com/pingcap/tidb/terror"
	"github.com/pingcap/tidb/util/hack"
	"github.com/pingcap/tidb/util/types"
)

// Error instances.
var (
	ErrSyntax = terror.ClassParser.New(CodeSyntaxErr, "syntax error")
)

// Error codes.
const (
	CodeSyntaxErr terror.ErrCode = 1
)

var (
	specCodePattern = regexp.MustCompile(`\/\*!(M?[0-9]{5,6})?([^*]|\*+[^*/])*\*+\/`)
	specCodeStart   = regexp.MustCompile(`^\/\*!(M?[0-9]{5,6} )?[ \t]*`)
	specCodeEnd     = regexp.MustCompile(`[ \t]*\*\/$`)
)

func trimComment(txt string) string {
	txt = specCodeStart.ReplaceAllString(txt, "")
	return specCodeEnd.ReplaceAllString(txt, "")
}

// Parser represents a parser instance. Some temporary objects are stored in it to reduce object allocation during Parse function.
type Parser struct {
	charset   string
	collation string
	result    []ast.StmtNode
	src       string
	lexer     Scanner

	// the following fields are used by yyParse to reduce allocation.
	cache  []yySymType
	yylval yySymType
	yyVAL  yySymType
}

type stmtTexter interface {
	stmtText() string
}

// New returns a Parser object.
func New() *Parser {
	return &Parser{
		cache: make([]yySymType, 200),
	}
}

// Parse parses a query string to raw ast.StmtNode.
// If charset or collation is "", default charset and collation will be used.
func (parser *Parser) Parse(sql, charset, collation string) ([]ast.StmtNode, error) {
	if charset == "" {
		charset = mysql.DefaultCharset
	}
	if collation == "" {
		collation = mysql.DefaultCollationName
	}
	parser.charset = charset
	parser.collation = collation
	parser.src = sql
	parser.result = parser.result[:0]

	var l yyLexer
	parser.lexer.reset(sql)
	l = &parser.lexer
	yyParse(l, parser)

	if len(l.Errors()) != 0 {
		return nil, errors.Trace(l.Errors()[0])
	}
	for _, stmt := range parser.result {
		ast.SetFlag(stmt)
	}
	return parser.result, nil
}

// ParseOneStmt parses a query and returns an ast.StmtNode.
// The query must have one statement, otherwise ErrSyntax is returned.
func (parser *Parser) ParseOneStmt(sql, charset, collation string) (ast.StmtNode, error) {
	stmts, err := parser.Parse(sql, charset, collation)
	if err != nil {
		return nil, errors.Trace(err)
	}
	if len(stmts) != 1 {
		return nil, ErrSyntax
	}
	ast.SetFlag(stmts[0])
	return stmts[0], nil
}

// SetSQLMode sets the SQL mode for parser.
func (parser *Parser) SetSQLMode(mode mysql.SQLMode) {
	parser.lexer.SetSQLMode(mode)
}

// The select statement is not at the end of the whole statement, if the last
// field text was set from its offset to the end of the src string, update
// the last field text.
func (parser *Parser) setLastSelectFieldText(st *ast.SelectStmt, lastEnd int) {
	lastField := st.Fields.Fields[len(st.Fields.Fields)-1]
	if lastField.Offset+len(lastField.Text()) >= len(parser.src)-1 {
		lastField.SetText(parser.src[lastField.Offset:lastEnd])
	}
}

func (parser *Parser) startOffset(v *yySymType) int {
	return v.offset
}

func (parser *Parser) endOffset(v *yySymType) int {
	offset := v.offset
	for offset > 0 && unicode.IsSpace(rune(parser.src[offset-1])) {
		offset--
	}
	return offset
}

func toInt(l yyLexer, lval *yySymType, str string) int {
	n, err := strconv.ParseUint(str, 10, 64)
	if err != nil {
		e := err.(*strconv.NumError)
		if e.Err == strconv.ErrRange {
			// TODO: toDecimal maybe out of range still.
			// This kind of error should be throw to higher level, because truncated data maybe legal.
			// For example, this SQL returns error:
			// create table test (id decimal(30, 0));
			// insert into test values(123456789012345678901234567890123094839045793405723406801943850);
			// While this SQL:
			// select 1234567890123456789012345678901230948390457934057234068019438509023041874359081325875128590860234789847359871045943057;
			// get value 99999999999999999999999999999999999999999999999999999999999999999
			return toDecimal(l, lval, str)
		}
		l.Errorf("integer literal: %v", err)
		return int(unicode.ReplacementChar)
	}

	switch {
	case n < math.MaxInt64:
		lval.item = int64(n)
	default:
		lval.item = uint64(n)
	}
	return intLit
}

func toDecimal(l yyLexer, lval *yySymType, str string) int {
	dec := new(types.MyDecimal)
	err := dec.FromString(hack.Slice(str))
	if err != nil {
		l.Errorf("decimal literal: %v", err)
	}
	lval.item = dec
	return decLit
}

func toFloat(l yyLexer, lval *yySymType, str string) int {
	n, err := strconv.ParseFloat(str, 64)
	if err != nil {
		l.Errorf("float literal: %v", err)
		return int(unicode.ReplacementChar)
	}

	lval.item = float64(n)
	return floatLit
}

// See https://dev.mysql.com/doc/refman/5.7/en/hexadecimal-literals.html
func toHex(l yyLexer, lval *yySymType, str string) int {
	h, err := types.ParseHex(str)
	if err != nil {
		// If parse hexadecimal literal to numerical value error, we should treat it as a string.
		hexStr, err1 := types.ParseHexStr(str)
		if err1 != nil {
			l.Errorf("hex literal: %v", err)
			return int(unicode.ReplacementChar)
		}
		lval.item = hexStr
		return hexLit
	}
	lval.item = h
	return hexLit
}

// See https://dev.mysql.com/doc/refman/5.7/en/bit-type.html
func toBit(l yyLexer, lval *yySymType, str string) int {
	b, err := types.ParseBit(str, -1)
	if err != nil {
		l.Errorf("bit literal: %v", err)
		return int(unicode.ReplacementChar)
	}
	lval.item = b
	return bitLit
}

func getUint64FromNUM(num interface{}) uint64 {
	switch v := num.(type) {
	case int64:
		return uint64(v)
	case uint64:
		return uint64(v)
	}
	return 0
}