*: restrict the length of enum/set value (pingcap#19109)

config/config.go

@@ -162,6 +162,12 @@ type Config struct {
 	EnableGlobalIndex bool `toml:"enable-global-index" json:"enable-global-index"`
 	// DeprecateIntegerDisplayWidth indicates whether deprecating the max display length for integer.
 	DeprecateIntegerDisplayWidth bool `toml:"deprecate-integer-display-length" json:"deprecate-integer-display-length"`
+	// EnableEnumLengthLimit indicates whether the enum/set element length is limited.
+	// According to MySQL 8.0 Refman:
+	// The maximum supported length of an individual SET element is M <= 255 and (M x w) <= 1020,
+	// where M is the element literal length and w is the number of bytes required for the maximum-length character in the character set.
+	// See https://dev.mysql.com/doc/refman/8.0/en/string-type-syntax.html for more details.
+	EnableEnumLengthLimit bool `toml:"enable-enum-length-limit" json:"enable-enum-length-limit"`
 }
 
 // UpdateTempStoragePath is to update the `TempStoragePath` if port/statusPort was changed
@@ -770,6 +776,7 @@ var defaultConf = Config{
 		SpilledFileEncryptionMethod: SpilledFileEncryptionMethodPlaintext,
 	},
 	DeprecateIntegerDisplayWidth: false,
+	EnableEnumLengthLimit:        true,
 }
 
 var (

config/config.toml.example

@@ -127,6 +127,13 @@ enable-telemetry = true
 # a warning like `Integer display width is deprecated and will be removed in a future release`.
 deprecate-integer-display-length = false
 
+# enable-enum-length-limit is used to deal with compatibility issues. When true, the enum/set element length is limited.
+# According to MySQL 8.0 Refman:
+# The maximum supported length of an individual SET element is M <= 255 and (M x w) <= 1020,
+# where M is the element literal length and w is the number of bytes required for the maximum-length character in the character set.
+# See https://dev.mysql.com/doc/refman/8.0/en/string-type-syntax.html for more details.
+enable-enum-length-limit = true
+
 [log]
 # Log level: debug, info, warn, error, fatal.
 level = "info"

config/config_test.go

@@ -193,6 +193,7 @@ nested-loop-join-cache-capacity = 100
 max-index-length = 3080
 skip-register-to-dashboard = true
 deprecate-integer-display-length = true
+enable-enum-length-limit = false
 [performance]
 txn-total-size-limit=2000
 [tikv-client]
@@ -271,6 +272,7 @@ spilled-file-encryption-method = "plaintext"
 	c.Assert(conf.Labels["group"], Equals, "abc")
 	c.Assert(conf.Security.SpilledFileEncryptionMethod, Equals, SpilledFileEncryptionMethodPlaintext)
 	c.Assert(conf.DeprecateIntegerDisplayWidth, Equals, true)
+	c.Assert(conf.EnableEnumLengthLimit, Equals, false)
 
 	_, err = f.WriteString(`
 [log.file]

ddl/ddl_api.go

@@ -962,8 +962,20 @@ func checkColumnValueConstraint(col *table.Column, collation string) error {
 	}
 	valueMap := make(map[string]bool, len(col.Elems))
 	ctor := collate.GetCollator(collation)
+	enumLengthLimit := config.GetGlobalConfig().EnableEnumLengthLimit
+	desc, err := charset.GetCharsetDesc(col.Charset)
+	if err != nil {
+		return errors.Trace(err)
+	}
 	for i := range col.Elems {
 		val := string(ctor.Key(col.Elems[i]))
+		// According to MySQL 8.0 Refman:
+		// The maximum supported length of an individual ENUM element is M <= 255 and (M x w) <= 1020,
+		// where M is the element literal length and w is the number of bytes required for the maximum-length character in the character set.
+		// See https://dev.mysql.com/doc/refman/8.0/en/string-type-syntax.html for more details.
+		if enumLengthLimit && (len(val) > 255 || len(val)*desc.Maxlen > 1020) {
+			return ErrTooLongValueForType.GenWithStackByArgs(col.Name)
+		}
 		if _, ok := valueMap[val]; ok {
 			tpStr := "ENUM"
 			if col.Tp == mysql.TypeSet {

ddl/error.go

@@ -259,4 +259,7 @@ var (
 	// ErrWarnDataOutOfRange is returned when the value in a numeric column that is outside the permissible range of the column data type.
 	// See https://dev.mysql.com/doc/refman/5.5/en/out-of-range-and-overflow.html for details
 	ErrWarnDataOutOfRange = dbterror.ClassDDL.NewStd(mysql.ErrWarnDataOutOfRange)
+
+	// ErrTooLongValueForType is returned when the individual enum element length is too long.
+	ErrTooLongValueForType = dbterror.ClassDDL.NewStd(mysql.ErrTooLongValueForType)
 )

ddl/serial_test.go

@@ -1480,3 +1480,29 @@ func (s *testSerialSuite) TestCreateTableNoBlock(c *C) {
 	_, err := tk.Exec("create table t(a int)")
 	c.Assert(err, NotNil)
 }
+
+func (s *testSerialSuite) TestCheckEnumLength(c *C) {
+	tk := testkit.NewTestKitWithInit(c, s.store)
+	tk.MustExec("drop table if exists t1,t2,t3,t4,t5")
+	tk.MustGetErrCode("create table t1 (a enum('aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'))", errno.ErrTooLongValueForType)
+	tk.MustGetErrCode("create table t1 (a set('aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'))", errno.ErrTooLongValueForType)
+	tk.MustExec("create table t2 (id int primary key)")
+	tk.MustGetErrCode("alter table t2 add a enum('aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa')", errno.ErrTooLongValueForType)
+	tk.MustGetErrCode("alter table t2 add a set('aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa')", errno.ErrTooLongValueForType)
+	config.UpdateGlobal(func(conf *config.Config) {
+		conf.EnableEnumLengthLimit = false
+	})
+	_, err := tk.Exec("create table t3 (a enum('aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'))")
+	c.Assert(err, IsNil)
+	tk.MustExec("insert into t3 values(1)")
+	tk.MustQuery("select a from t3").Check(testkit.Rows("aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"))
+	_, err = tk.Exec("create table t4 (a set('aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'))")
+	c.Assert(err, IsNil)
+
+	config.UpdateGlobal(func(conf *config.Config) {
+		conf.EnableEnumLengthLimit = true
+	})
+	tk.MustGetErrCode("create table t5 (a enum('aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'))", errno.ErrTooLongValueForType)
+	tk.MustGetErrCode("create table t5 (a set('aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'))", errno.ErrTooLongValueForType)
+	tk.MustExec("drop table if exists t1,t2,t3,t4,t5")
+}

errno/errcode.go

@@ -847,6 +847,7 @@ const (
 	ErrUserAlreadyExists                                            = 3163
 	ErrInvalidJSONPathArrayCell                                     = 3165
 	ErrInvalidEncryptionOption                                      = 3184
+	ErrTooLongValueForType                                          = 3505
 	ErrPKIndexCantBeInvisible                                       = 3522
 	ErrRoleNotGranted                                               = 3530
 	ErrLockAcquireFailAndNoWaitSet                                  = 3572

errno/errname.go

@@ -849,6 +849,7 @@ var MySQLErrName = map[uint16]*mysql.ErrMessage{
 	ErrUserAlreadyExists:                                     mysql.Message("User %s already exists.", nil),
 	ErrInvalidJSONPathArrayCell:                              mysql.Message("A path expression is not a path to a cell in an array.", nil),
 	ErrInvalidEncryptionOption:                               mysql.Message("Invalid encryption option.", nil),
+	ErrTooLongValueForType:                                   mysql.Message("Too long enumeration/set value for column %s.", nil),
 	ErrPKIndexCantBeInvisible:                                mysql.Message("A primary key index cannot be invisible", nil),
 	ErrWindowNoSuchWindow:                                    mysql.Message("Window name '%s' is not defined.", nil),
 	ErrWindowCircularityInWindowGraph:                        mysql.Message("There is a circularity in the window dependency graph.", nil),

errors.toml

@@ -311,6 +311,11 @@ error = '''
 Generated column '%s' cannot refer to auto-increment column.
 '''
 
+["ddl:3505"]
+error = '''
+Too long enumeration/set value for column %s.
+'''
+
 ["ddl:3522"]
 error = '''
 A primary key index cannot be invisible

go.mod

@@ -1,6 +1,7 @@
 module github.com/pingcap/tidb
 
 require (
+	cloud.google.com/go v0.51.0 // indirect
 	github.com/BurntSushi/toml v0.3.1
 	github.com/HdrHistogram/hdrhistogram-go v0.9.0 // indirect
 	github.com/Jeffail/gabs/v2 v2.5.1

go.sum

@@ -7,6 +7,8 @@ cloud.google.com/go v0.45.1/go.mod h1:RpBamKRgapWJb87xiFSdk4g1CME7QZg3uwTez+TSTj
 cloud.google.com/go v0.46.3/go.mod h1:a6bKKbmY7er1mI7TEI4lsAkts/mkhTSZK8w33B4RAg0=
 cloud.google.com/go v0.50.0 h1:0E3eE8MX426vUOs7aHfI7aN1BrIzzzf4ccKCSfSjGmc=
 cloud.google.com/go v0.50.0/go.mod h1:r9sluTvynVuxRIOHXQEHMFffphuXHOMZMycpNR5e6To=
+cloud.google.com/go v0.51.0 h1:PvKAVQWCtlGUSlZkGW3QLelKaWq7KYv/MW1EboG8bfM=
+cloud.google.com/go v0.51.0/go.mod h1:hWtGJ6gnXH+KgDv+V0zFGDvpi07n3z8ZNj3T1RW0Gcw=
 cloud.google.com/go/bigquery v1.0.1/go.mod h1:i/xbL2UlR5RvWAURpBYZTtm/cXjCha9lbfbpx4poX+o=
 cloud.google.com/go/bigquery v1.3.0 h1:sAbMqjY1PEQKZBWfbu6Y6bsupJ9c4QdHnzg/VvYTLcE=
 cloud.google.com/go/bigquery v1.3.0/go.mod h1:PjpwJnslEMmckchkHFfq+HTD2DmtT67aNFKH1/VBDHE=
@@ -216,6 +218,7 @@ github.com/google/martian v2.1.0+incompatible h1:/CP5g8u/VJHijgedC/Legn3BAbAaWPg
 github.com/google/martian v2.1.0+incompatible/go.mod h1:9I4somxYTbIHy5NJKHRl3wXiIaQGbYVAs8BPL6v8lEs=
 github.com/google/pprof v0.0.0-20181206194817-3ea8567a2e57/go.mod h1:zfwlbNMJ+OItoe0UupaVj+oy1omPYYDuagoSzA8v9mc=
 github.com/google/pprof v0.0.0-20190515194954-54271f7e092f/go.mod h1:zfwlbNMJ+OItoe0UupaVj+oy1omPYYDuagoSzA8v9mc=
+github.com/google/pprof v0.0.0-20191218002539-d4f498aebedc/go.mod h1:ZgVRPoUq/hfqzAqh7sHMqb3I9Rq5C59dIz2SbBwJ4eM=
 github.com/google/pprof v0.0.0-20200407044318-7d83b28da2e9 h1:K+lX49/3eURCE1IjlaZN//u6c+9nfDAMnyQ9E2dsJbY=
 github.com/google/pprof v0.0.0-20200407044318-7d83b28da2e9/go.mod h1:ZgVRPoUq/hfqzAqh7sHMqb3I9Rq5C59dIz2SbBwJ4eM=
 github.com/google/renameio v0.1.0/go.mod h1:KWCgfxg9yswjAJkECMjeO8J8rahYeXnNhOm40UhjYkI=