Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix Enum column definition parse logic to match ClickHouse spec #1380

Merged
merged 2 commits into from
Aug 22, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
188 changes: 128 additions & 60 deletions lib/column/enum.go
Original file line number Diff line number Diff line change
Expand Up @@ -18,89 +18,157 @@
package column

import (
"bytes"
"errors"
"github.com/ClickHouse/ch-go/proto"
"math"
"strconv"
"strings"

"github.com/ClickHouse/ch-go/proto"
)

func Enum(chType Type, name string) (Interface, error) {
var (
payload string
columnType = string(chType)
)
if len(columnType) < 8 {
return nil, &Error{
ColumnType: string(chType),
Err: errors.New("invalid Enum"),
}
}
switch {
case strings.HasPrefix(columnType, "Enum8"):
payload = columnType[6:]
case strings.HasPrefix(columnType, "Enum16"):
payload = columnType[7:]
default:
enumType, values, indexes, valid := extractEnumNamedValues(chType)
if !valid {
return nil, &Error{
ColumnType: string(chType),
Err: errors.New("invalid Enum"),
}
}
var (
idents []string
indexes []int64
)
for _, block := range strings.Split(payload[:len(payload)-1], ",") {
parts := strings.Split(block, "=")
if len(parts) != 2 {
return nil, &Error{
ColumnType: string(chType),
Err: errors.New("invalid Enum"),
}
}
var (
ident = strings.TrimSpace(parts[0])
index, err = strconv.ParseInt(strings.TrimSpace(parts[1]), 10, 16)
)
if err != nil || len(ident) < 2 {
return nil, &Error{
ColumnType: string(chType),
Err: errors.New("invalid Enum"),
}
}
ident = ident[1 : len(ident)-1]
idents, indexes = append(idents, ident), append(indexes, index)
}
if strings.HasPrefix(columnType, "Enum8") {

if enumType == enum8Type {
enum := Enum8{
iv: make(map[string]proto.Enum8, len(idents)),
vi: make(map[proto.Enum8]string, len(idents)),
iv: make(map[string]proto.Enum8, len(values)),
vi: make(map[proto.Enum8]string, len(values)),
chType: chType,
name: name,
}
for i := range idents {
if indexes[i] > math.MaxUint8 {
return nil, &Error{
ColumnType: string(chType),
Err: errors.New("invalid Enum"),
}
}
for i := range values {
v := int8(indexes[i])

Check failure

Code scanning / CodeQL

Incorrect conversion between integer types High

Incorrect conversion of an integer with architecture-dependent bit size from
strconv.Atoi
to a lower bit size type int8 without an upper bound check.
Incorrect conversion of an integer with architecture-dependent bit size from
strconv.Atoi
to a lower bit size type int8 without an upper bound check.
enum.iv[idents[i]] = proto.Enum8(v)
enum.vi[proto.Enum8(v)] = idents[i]
enum.iv[values[i]] = proto.Enum8(v)
enum.vi[proto.Enum8(v)] = values[i]
}
return &enum, nil
}
enum := Enum16{
iv: make(map[string]proto.Enum16, len(idents)),
vi: make(map[proto.Enum16]string, len(idents)),
iv: make(map[string]proto.Enum16, len(values)),
vi: make(map[proto.Enum16]string, len(values)),
chType: chType,
name: name,
}
for i := range idents {
enum.iv[idents[i]] = proto.Enum16(indexes[i])
enum.vi[proto.Enum16(indexes[i])] = idents[i]

for i := range values {
enum.iv[values[i]] = proto.Enum16(indexes[i])
Dismissed Show dismissed Hide dismissed
enum.vi[proto.Enum16(indexes[i])] = values[i]
Dismissed Show dismissed Hide dismissed
}
return &enum, nil
}

const (
enum8Type = "Enum8"
E
enum16Type = "Enum16"
)

func extractEnumNamedValues(chType Type) (typ string, values []string, indexes []int, valid bool) {
src := []byte(chType)

var bracketOpen, stringOpen bool

var foundValueOffset int
var foundValueLen int
var skippedValueTokens []int
var indexFound bool
var valueIndex = 0

for c := 0; c < len(src); c++ {
token := src[c]

switch {
// open bracket found, capture the type
case token == '(' && !stringOpen:
typ = string(src[:c])

// Ignore everything captured as non-enum type
if typ != enum8Type && typ != enum16Type {
return
}

bracketOpen = true
break
// when inside a bracket, we can start capture value inside single quotes
case bracketOpen && token == '\'' && !stringOpen:
foundValueOffset = c + 1
stringOpen = true
break
// close the string and capture the value
case token == '\'' && stringOpen:
stringOpen = false
foundValueLen = c - foundValueOffset
break
// escape character, skip the next character
case token == '\\' && stringOpen:
skippedValueTokens = append(skippedValueTokens, c-foundValueOffset)
c++
break
// capture optional index. `=` token is followed with an integer index
case token == '=' && !stringOpen:
if foundValueLen == 0 {
return
}

indexStart := c + 1
// find the end of the index, it's either a comma or a closing bracket
for _, token := range src[indexStart:] {
if token == ',' || token == ')' {
break
}
c++
}

idx, err := strconv.Atoi(string(bytes.TrimSpace(src[indexStart : c+1])))
if err != nil {
return
}
valueIndex = idx
indexFound = true
break
// capture the value and index when a comma or closing bracket is found
case (token == ',' || token == ')') && !stringOpen:
if foundValueLen == 0 {
return
}

// if no index was found for current value, increment the value index
// e.g. Enum8('a','b') is equivalent to Enum8('a'=1,'b'=2)
// or Enum8('a'=3,'b') is equivalent to Enum8('a'=3,'b'=4)
// so if no index is provided, we increment the value index
if !indexFound {
valueIndex++
}

// if the index is out of range, return
if (typ == enum8Type && valueIndex > math.MaxUint8) ||
(typ == enum16Type && valueIndex > math.MaxUint16) {
return
}

foundName := src[foundValueOffset : foundValueOffset+foundValueLen]
for _, skipped := range skippedValueTokens {
foundName = append(foundName[:skipped], foundName[skipped+1:]...)
}

indexes = append(indexes, valueIndex)
values = append(values, string(foundName))
indexFound = false
break
}
}

// Enum type must have at least one value
if valueIndex == 0 {
return
}

valid = true
return
}
148 changes: 148 additions & 0 deletions lib/column/enum_test.go
Original file line number Diff line number Diff line change
@@ -0,0 +1,148 @@
package column

import (
"testing"

"github.com/stretchr/testify/assert"
)

func TestExtractEnumNamedValues(t *testing.T) {
tests := []struct {
name string
chType Type
expectedType string
expectedValues map[int]string
isNotValid bool
}{
{
name: "Enum8",
chType: "Enum8('a'=1,'b'=2)",
expectedType: "Enum8",
expectedValues: map[int]string{
1: "a",
2: "b",
},
},
{
name: "Enum16",
chType: "Enum16('a'=1,'b'=2)",
expectedType: "Enum16",
expectedValues: map[int]string{
1: "a",
2: "b",
},
},
{
name: "Enum8 with comma in value",
chType: "Enum8('a'=1,'b'=2,'c,d'=3)",
expectedType: "Enum8",
expectedValues: map[int]string{
1: "a",
2: "b",
3: "c,d",
},
},
{
name: "Enum8 with spaces",
chType: "Enum8('a' = 1, 'b' = 2)",
expectedType: "Enum8",
expectedValues: map[int]string{
1: "a",
2: "b",
},
},
{
name: "Enum8 without indexes",
chType: "Enum8('a','b')",
expectedType: "Enum8",
expectedValues: map[int]string{
1: "a",
2: "b",
},
},
{
name: "Enum8 with a first index only",
chType: "Enum8('a'=1,'b')",
expectedType: "Enum8",
expectedValues: map[int]string{
1: "a",
2: "b",
},
},
{
name: "Enum8 with a last index only",
chType: "Enum8('a','b'=5)",
expectedType: "Enum8",
expectedValues: map[int]string{
1: "a",
5: "b",
},
},
{
name: "Enum8 with a first index only higher than 1",
chType: "Enum8('a'=5,'b')",
expectedType: "Enum8",
expectedValues: map[int]string{
5: "a",
6: "b",
},
},
{
name: "Enum8 with index with spaces",
chType: "Enum8( 'a' , 'b' = 5 )",
expectedType: "Enum8",
expectedValues: map[int]string{
1: "a",
5: "b",
},
},
{
name: "Enum8 with escaped quotes",
chType: `Enum8('a\'b'=1)`,
expectedType: "Enum8",
expectedValues: map[int]string{
1: "a'b",
},
},
{
name: "Enum8 with invalid index",
chType: "Enum8('a'=1,'b'=256)",
isNotValid: true,
},
{
name: "Enum8 with invalid non-integer index",
chType: "Enum8('a'=1,'b'='c')",
isNotValid: true,
},
{
name: "Empty Enum8",
chType: "Enum8()",
isNotValid: true,
},
{
name: "Empty Enum8 without brackets",
chType: "Enum8",
isNotValid: true,
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
actualType, actualValues, actualIndexes, valid := extractEnumNamedValues(tt.chType)

if tt.isNotValid {
assert.False(t, valid, "%s is valid enum", tt.chType)
return
}

actualValuesMap := make(map[int]string)
for i, v := range actualValues {
actualValuesMap[actualIndexes[i]] = v
}

assert.Equal(t, tt.expectedType, actualType)
assert.Equal(t, tt.expectedValues, actualValuesMap)

assert.True(t, valid, "%s is not valid enum", tt.chType)
})
}
}
7 changes: 4 additions & 3 deletions tests/enum_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -21,9 +21,10 @@ import (
"context"
"database/sql/driver"
"fmt"
"github.com/stretchr/testify/require"
"testing"

"github.com/stretchr/testify/require"

"github.com/ClickHouse/clickhouse-go/v2"
"github.com/stretchr/testify/assert"
)
Expand Down Expand Up @@ -110,7 +111,7 @@ func TestEnum(t *testing.T) {
require.NoError(t, err)
const ddl = `
CREATE TABLE test_enum (
Col1 Enum ('hello' = 1, 'world' = 2)
Col1 Enum ('hello,hello' = 1, 'world' = 2)
, Col2 Enum8 ('click' = 5, 'house' = 25)
, Col3 Enum16('house' = 10, 'value' = 50)
, Col4 Array(Enum8 ('click' = 1, 'house' = 2))
Expand All @@ -126,7 +127,7 @@ func TestEnum(t *testing.T) {
batch, err := conn.PrepareBatch(ctx, "INSERT INTO test_enum")
require.NoError(t, err)
var (
col1Data = "hello"
col1Data = "hello,hello"
col2Data = "click"
col3Data = "house"
col4Data = []string{"click", "house"}
Expand Down
Loading
Loading