Enum to varchar (#963)

* v1.1.0

* WIP: copying AUTO_INCREMENT value to ghost table
Initial commit: towards setting up a test suite

Signed-off-by: Shlomi Noach <[email protected]>

* greping for 'expect_table_structure' content

* Adding simple test for 'expect_table_structure' scenario

* adding tests for AUTO_INCREMENT value after row deletes. Should initially fail

* clear event beforehand

* parsing AUTO_INCREMENT from alter query, reading AUTO_INCREMENT from original table, applying AUTO_INCREMENT value onto ghost table if applicable and user has not specified AUTO_INCREMENT in alter statement

* support GetUint64

Signed-off-by: Shlomi Noach <[email protected]>

* minor update to test

Signed-off-by: Shlomi Noach <[email protected]>

* adding test for user defined AUTO_INCREMENT statement

* Generated column as part of UNIQUE (or PRIMARY) KEY

Signed-off-by: Shlomi Noach <[email protected]>

* skip analysis of generated column data type in unique key

Signed-off-by: Shlomi Noach <[email protected]>

* All MySQL DBs limited to max 3 concurrent/idle connections

Signed-off-by: Shlomi Noach <[email protected]>

* hooks: reporting GH_OST_ETA_SECONDS. ETA stored as part of migration context

Signed-off-by: Shlomi Noach <[email protected]>

* GH_OST_ETA_NANOSECONDS

Signed-off-by: Shlomi Noach <[email protected]>

* N/A denoted by negative value

Signed-off-by: Shlomi Noach <[email protected]>

* ETAUnknown constant

Signed-off-by: Shlomi Noach <[email protected]>

* Convering enum to varchar

Signed-off-by: Shlomi Noach <[email protected]>

* test: not null

Signed-off-by: Shlomi Noach <[email protected]>

* first attempt at setting enum-to-string right

Signed-off-by: Shlomi Noach <[email protected]>

* fix insert query

Signed-off-by: Shlomi Noach <[email protected]>

* store enum values, use when populating

Signed-off-by: Shlomi Noach <[email protected]>

* apply EnumValues to mapped column

Signed-off-by: Shlomi Noach <[email protected]>

* fix compilation error

Signed-off-by: Shlomi Noach <[email protected]>

* gofmt

Signed-off-by: Shlomi Noach <[email protected]>

Author: shlomi-noach

go/logic/inspect.go

@@ -187,6 +187,10 @@ func (this *Inspector) inspectOriginalAndGhostTables() (err error) {
 		if column.Name == mappedColumn.Name && column.Type == sql.DateTimeColumnType && mappedColumn.Type == sql.TimestampColumnType {
 			this.migrationContext.MappedSharedColumns.SetConvertDatetimeToTimestamp(column.Name, this.migrationContext.ApplierTimeZone)
 		}
+		if column.Name == mappedColumn.Name && column.Type == sql.EnumColumnType && mappedColumn.Charset != "" {
+			this.migrationContext.MappedSharedColumns.SetEnumToTextConversion(column.Name)
+			this.migrationContext.MappedSharedColumns.SetEnumValues(column.Name, column.EnumValues)
+		}
 	}
 
 	for _, column := range this.migrationContext.UniqueKey.Columns.Columns() {
@@ -590,6 +594,7 @@ func (this *Inspector) applyColumnTypes(databaseName, tableName string, columnsL
 			}
 			if strings.HasPrefix(columnType, "enum") {
 				column.Type = sql.EnumColumnType
+				column.EnumValues = sql.ParseEnumValues(m.GetString("COLUMN_TYPE"))
 			}
 			if strings.HasPrefix(columnType, "binary") {
 				column.Type = sql.BinaryColumnType

go/sql/builder.go

@@ -38,6 +38,8 @@ func buildColumnsPreparedValues(columns *ColumnList) []string {
 		var token string
 		if column.timezoneConversion != nil {
 			token = fmt.Sprintf("convert_tz(?, '%s', '%s')", column.timezoneConversion.ToTimezone, "+00:00")
+		} else if column.enumToTextConversion {
+			token = fmt.Sprintf("ELT(?, %s)", column.EnumValues)
 		} else if column.Type == JSONColumnType {
 			token = "convert(? using utf8mb4)"
 		} else {
@@ -108,6 +110,8 @@ func BuildSetPreparedClause(columns *ColumnList) (result string, err error) {
 		var setToken string
 		if column.timezoneConversion != nil {
 			setToken = fmt.Sprintf("%s=convert_tz(?, '%s', '%s')", EscapeName(column.Name), column.timezoneConversion.ToTimezone, "+00:00")
+		} else if column.enumToTextConversion {
+			setToken = fmt.Sprintf("%s=ELT(?, %s)", EscapeName(column.Name), column.EnumValues)
 		} else if column.Type == JSONColumnType {
 			setToken = fmt.Sprintf("%s=convert(? using utf8mb4)", EscapeName(column.Name))
 		} else {

go/sql/parser.go

@@ -33,6 +33,7 @@ var (
 		// ALTER TABLE tbl something
 		regexp.MustCompile(`(?i)\balter\s+table\s+([\S]+)\s+(.*$)`),
 	}
+	enumValuesRegexp = regexp.MustCompile("^enum[(](.*)[)]$")
 )
 
 type AlterTableParser struct {
@@ -205,3 +206,10 @@ func (this *AlterTableParser) HasExplicitTable() bool {
 func (this *AlterTableParser) GetAlterStatementOptions() string {
 	return this.alterStatementOptions
 }
+
+func ParseEnumValues(enumColumnType string) string {
+	if submatch := enumValuesRegexp.FindStringSubmatch(enumColumnType); len(submatch) > 0 {
+		return submatch[1]
+	}
+	return enumColumnType
+}

go/sql/parser_test.go

@@ -322,3 +322,21 @@ func TestParseAlterStatementExplicitTable(t *testing.T) {
 		test.S(t).ExpectTrue(reflect.DeepEqual(parser.alterTokens, []string{"drop column b", "add index idx(i)"}))
 	}
 }
+
+func TestParseEnumValues(t *testing.T) {
+	{
+		s := "enum('red','green','blue','orange')"
+		values := ParseEnumValues(s)
+		test.S(t).ExpectEquals(values, "'red','green','blue','orange'")
+	}
+	{
+		s := "('red','green','blue','orange')"
+		values := ParseEnumValues(s)
+		test.S(t).ExpectEquals(values, "('red','green','blue','orange')")
+	}
+	{
+		s := "zzz"
+		values := ParseEnumValues(s)
+		test.S(t).ExpectEquals(values, "zzz")
+	}
+}

go/sql/types.go

@@ -33,15 +33,16 @@ type TimezoneConversion struct {
 }
 
 type Column struct {
-	Name       string
-	IsUnsigned bool
-	Charset    string
-	Type       ColumnType
-
+	Name                 string
+	IsUnsigned           bool
+	Charset              string
+	Type                 ColumnType
+	EnumValues           string
+	timezoneConversion   *TimezoneConversion
+	enumToTextConversion bool
 	// add Octet length for binary type, fix bytes with suffix "00" get clipped in mysql binlog.
 	// https://github.com/github/gh-ost/issues/909
-	BinaryOctetLength  uint
-	timezoneConversion *TimezoneConversion
+	BinaryOctetLength uint
 }
 
 func (this *Column) convertArg(arg interface{}, isUniqueKeyColumn bool) interface{} {
@@ -198,6 +199,18 @@ func (this *ColumnList) HasTimezoneConversion(columnName string) bool {
 	return this.GetColumn(columnName).timezoneConversion != nil
 }
 
+func (this *ColumnList) SetEnumToTextConversion(columnName string) {
+	this.GetColumn(columnName).enumToTextConversion = true
+}
+
+func (this *ColumnList) IsEnumToTextConversion(columnName string) bool {
+	return this.GetColumn(columnName).enumToTextConversion
+}
+
+func (this *ColumnList) SetEnumValues(columnName string, enumValues string) {
+	this.GetColumn(columnName).EnumValues = enumValues
+}
+
 func (this *ColumnList) String() string {
 	return strings.Join(this.Names(), ",")
 }

localtests/enum-to-varchar/create.sql

@@ -0,0 +1,26 @@
+drop table if exists gh_ost_test;
+create table gh_ost_test (
+  id int auto_increment,
+  i int not null,
+  e enum('red', 'green', 'blue', 'orange') null default null collate 'utf8_bin',
+  primary key(id)
+) auto_increment=1;
+
+insert into gh_ost_test values (null, 7, 'red');
+
+drop event if exists gh_ost_test;
+delimiter ;;
+create event gh_ost_test
+  on schedule every 1 second
+  starts current_timestamp
+  ends current_timestamp + interval 60 second
+  on completion not preserve
+  enable
+  do
+begin
+  insert into gh_ost_test values (null, 11, 'red');
+  insert into gh_ost_test values (null, 13, 'green');
+  insert into gh_ost_test values (null, 17, 'blue');
+  set @last_insert_id := last_insert_id();
+  update gh_ost_test set e='orange' where id = @last_insert_id;
+end ;;

localtests/enum-to-varchar/extra_args

@@ -0,0 +1 @@
+--alter="change e e varchar(32) not null default ''"