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.

Sign up for GitHub

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

expression: implement vectorized evaluation for builtinLocate3ArgsSig #12510

Merged
merged 12 commits into from
Oct 14, 2019
Merged

expression: implement vectorized evaluation for builtinLocate3ArgsSig #12510

Show file tree
Hide file tree
Changes from 11 commits
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
c16e6f7
add builtinLocate3ArgsSig
b41sh Oct 1, 2019
454effc
Merge branch 'master' of https://github.com/pingcap/tidb into vecexpr…
b41sh Oct 3, 2019
d77ac47
update
b41sh Oct 3, 2019
78fcb93
Merge branch 'master' of https://github.com/pingcap/tidb into vecexpr…
b41sh Oct 8, 2019
44179a5
add test case
b41sh Oct 8, 2019
496feaf
optimize
b41sh Oct 8, 2019
9f98ddb
update
b41sh Oct 8, 2019
3ae753a
fix
b41sh Oct 9, 2019
7feddc7
Merge branch 'master' into vecexpr-builtinLocate3ArgsSig
b41sh Oct 9, 2019
c3e2386
merge
b41sh Oct 10, 2019
feac0ce
Merge branch 'vecexpr-builtinLocate3ArgsSig' of https://github.com/b4…
b41sh Oct 10, 2019
9958b6b
Merge branch 'master' into vecexpr-builtinLocate3ArgsSig
sre-bot Oct 14, 2019
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
59 changes: 57 additions & 2 deletions expression/builtin_string_vec.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -334,11 +334,66 @@ func (b *builtinConcatSig) vecEvalString(input *chunk.Chunk, result *chunk.Colum
}

func (b *builtinLocate3ArgsSig) vectorized() bool {
return false
return true
}

// vecEvalInt evals LOCATE(substr,str,pos), non case-sensitive.
// See https://dev.mysql.com/doc/refman/5.7/en/string-functions.html#function_locate
func (b *builtinLocate3ArgsSig) vecEvalInt(input *chunk.Chunk, result *chunk.Column) error {
return errors.Errorf("not implemented")
n := input.NumRows()
buf, err := b.bufAllocator.get(types.ETString, n)
if err != nil {
return err
}
defer b.bufAllocator.put(buf)
if err := b.args[0].VecEvalString(b.ctx, input, buf); err != nil {
return err
}
buf1, err := b.bufAllocator.get(types.ETString, n)
if err != nil {
return err
}
defer b.bufAllocator.put(buf1)
if err := b.args[1].VecEvalString(b.ctx, input, buf1); err != nil {
return err
}
// store positions in result
if err := b.args[2].VecEvalInt(b.ctx, input, result); err != nil {
return err
}

result.MergeNulls(buf, buf1)
i64s := result.Int64s()
for i := 0; i < n; i++ {
if result.IsNull(i) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Consider this case, b.args[2] is null but the result is not null.

mysql> SELECT LOCATE('bar', 'foobarbar', NULL);
+----------------------------------+
| LOCATE('bar', 'foobarbar', NULL) |
+----------------------------------+
|                                0 |
+----------------------------------+
1 row in set (0.00 sec)

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

My test result is correct

mysql> SELECT LOCATE('bar', 'foobarbar', NULL);
+----------------------------------+
| LOCATE('bar', 'foobarbar', NULL) |
+----------------------------------+
|                             NULL |
+----------------------------------+
1 row in set (0.00 sec)

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can args[0] and args[1] be null?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yes, any args is null will result in null

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Consider this case, b.args[2] is null but the result is not null.

mysql> SELECT LOCATE('bar', 'foobarbar', NULL);
+----------------------------------+
| LOCATE('bar', 'foobarbar', NULL) |
+----------------------------------+
|                                0 |
+----------------------------------+
1 row in set (0.00 sec)

In MySQL 8.x, this SQL returns NULL.

continue
}
subStr := buf.GetString(i)
str := buf1.GetString(i)
pos := i64s[i]

// Transfer the argument which starts from 1 to real index which starts from 0.
pos--
strLen := int64(len([]rune(str)))
subStrLen := int64(len([]rune(subStr)))
if pos < 0 || pos > strLen-subStrLen {
i64s[i] = 0
continue
} else if subStrLen == 0 {
i64s[i] = pos + 1
continue
}
slice := string([]rune(str)[pos:])
subStr = strings.ToLower(subStr)
slice = strings.ToLower(slice)
idx := strings.Index(slice, subStr)
if idx != -1 {
i64s[i] = pos + int64(utf8.RuneCountInString(slice[:idx])) + 1
continue
}
i64s[i] = 0
}
return nil
}

func (b *builtinHexStrArgSig) vectorized() bool {
Expand Down
10 changes: 10 additions & 0 deletions expression/builtin_string_vec_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,6 +40,16 @@ var vecBuiltinStringCases = map[string][]vecExprBenchCase{
childrenTypes: []types.EvalType{types.ETString, types.ETString},
geners: []dataGenerator{&randLenStrGener{1, 2}, &randLenStrGener{0, 20}},
},
{
retEvalType: types.ETInt,
childrenTypes: []types.EvalType{types.ETString, types.ETString, types.ETInt},
geners: []dataGenerator{&randLenStrGener{0, 10}, &randLenStrGener{0, 20}, &rangeInt64Gener{-10, 20}},
},
{
retEvalType: types.ETInt,
childrenTypes: []types.EvalType{types.ETString, types.ETString, types.ETInt},
geners: []dataGenerator{&randLenStrGener{1, 2}, &randLenStrGener{0, 10}, &rangeInt64Gener{0, 8}},
},
},
ast.Hex: {},
ast.Unhex: {
Expand Down