Skip to content

⚡ Avoid fmt.Sprintf for Map Keys in temporal Mode functions #158

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

Open
suyashkumar wants to merge 1 commit into
base: main
Choose a base branch
from
Open

⚡ Avoid fmt.Sprintf for Map Keys in temporal Mode functions #158

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
61 changes: 41 additions & 20 deletions interpreter/operator_aggregate.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1584,8 +1584,11 @@ func (i *interpreter) evalModeDate(_ model.IUnaryExpression, operand result.Valu
}

// Count occurrences of each value
counts := make(map[string]int)
dateMap := make(map[string]result.Date)
type dateKey struct {
year, month, day int
}
counts := make(map[dateKey]int)
dateMap := make(map[dateKey]result.Date)

for _, elem := range l {
if result.IsNull(elem) {
Expand All @@ -1596,8 +1599,12 @@ func (i *interpreter) evalModeDate(_ model.IUnaryExpression, operand result.Valu
return result.Value{}, err
}

// Format date as string to use as map key
key := fmt.Sprintf("%d-%02d-%02d", v.Date.Year(), v.Date.Month(), v.Date.Day())
// Use a struct key to avoid fmt.Sprintf overhead
key := dateKey{
year: v.Date.Year(),
month: int(v.Date.Month()),
day: v.Date.Day(),
}
counts[key]++
dateMap[key] = v
}
Expand All @@ -1607,7 +1614,7 @@ func (i *interpreter) evalModeDate(_ model.IUnaryExpression, operand result.Valu
}

// Find the most frequent value
var modeKey string
var modeKey dateKey
maxCount := 0
for key, count := range counts {
if count > maxCount {
Expand All @@ -1631,8 +1638,11 @@ func (i *interpreter) evalModeDateTime(_ model.IUnaryExpression, operand result.
}

// Count occurrences of each value
counts := make(map[string]int)
dtMap := make(map[string]result.DateTime)
type dateTimeKey struct {
year, month, day, hour, minute, second, millisecond int
}
counts := make(map[dateTimeKey]int)
dtMap := make(map[dateTimeKey]result.DateTime)

for _, elem := range l {
if result.IsNull(elem) {
Expand All @@ -1643,11 +1653,16 @@ func (i *interpreter) evalModeDateTime(_ model.IUnaryExpression, operand result.
return result.Value{}, err
}

// Format datetime as string to use as map key
key := fmt.Sprintf("%d-%02d-%02dT%02d:%02d:%02d.%03d",
v.Date.Year(), v.Date.Month(), v.Date.Day(),
v.Date.Hour(), v.Date.Minute(), v.Date.Second(),
v.Date.Nanosecond()/1000000)
// Use a struct key to avoid fmt.Sprintf overhead
key := dateTimeKey{
year: v.Date.Year(),
month: int(v.Date.Month()),
day: v.Date.Day(),
hour: v.Date.Hour(),
minute: v.Date.Minute(),
second: v.Date.Second(),
millisecond: v.Date.Nanosecond() / 1000000,
}
counts[key]++
dtMap[key] = v
}
Expand All @@ -1657,7 +1672,7 @@ func (i *interpreter) evalModeDateTime(_ model.IUnaryExpression, operand result.
}

// Find the most frequent value
var modeKey string
var modeKey dateTimeKey
maxCount := 0
for key, count := range counts {
if count > maxCount {
Expand All @@ -1681,8 +1696,11 @@ func (i *interpreter) evalModeTime(_ model.IUnaryExpression, operand result.Valu
}

// Count occurrences of each value
counts := make(map[string]int)
timeMap := make(map[string]result.Time)
type timeKey struct {
hour, minute, second, millisecond int
}
counts := make(map[timeKey]int)
timeMap := make(map[timeKey]result.Time)

for _, elem := range l {
if result.IsNull(elem) {
Expand All @@ -1693,10 +1711,13 @@ func (i *interpreter) evalModeTime(_ model.IUnaryExpression, operand result.Valu
return result.Value{}, err
}

// Format time as string to use as map key
key := fmt.Sprintf("%02d:%02d:%02d.%03d",
v.Date.Hour(), v.Date.Minute(), v.Date.Second(),
v.Date.Nanosecond()/1000000)
// Use a struct key to avoid fmt.Sprintf overhead
key := timeKey{
hour: v.Date.Hour(),
minute: v.Date.Minute(),
second: v.Date.Second(),
millisecond: v.Date.Nanosecond() / 1000000,
}
counts[key]++
timeMap[key] = v
}
Expand All @@ -1706,7 +1727,7 @@ func (i *interpreter) evalModeTime(_ model.IUnaryExpression, operand result.Valu
}

// Find the most frequent value
var modeKey string
var modeKey timeKey
maxCount := 0
for key, count := range counts {
if count > maxCount {
Expand Down
15 changes: 15 additions & 0 deletions tests/enginetests/operator_aggregate_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1364,6 +1364,21 @@ func TestMode(t *testing.T) {
cql: "Mode({'a', 'b', 'c', 'b', 'a', 'b'})",
wantResult: newOrFatal(t, "b"),
},
{
name: "Mode({@T01:01, @T01:02, @T01:01})",
cql: "Mode({@T01:01, @T01:02, @T01:01})",
wantResult: newOrFatal(t, result.Time{Date: time.Date(0, 1, 1, 1, 1, 0, 0, time.UTC), Precision: model.MINUTE}),
},
{
name: "Mode({@2012-01-01, @2012-01-02, @2012-01-01})",
cql: "Mode({@2012-01-01, @2012-01-02, @2012-01-01})",
wantResult: newOrFatal(t, result.Date{Date: time.Date(2012, 1, 1, 0, 0, 0, 0, defaultEvalTimestamp.Location()), Precision: model.DAY}),
},
{
name: "Mode({@2012-01-01T01:01, @2012-01-01T01:02, @2012-01-01T01:01})",
cql: "Mode({@2012-01-01T01:01, @2012-01-01T01:02, @2012-01-01T01:01})",
wantResult: newOrFatal(t, result.DateTime{Date: time.Date(2012, 1, 1, 1, 1, 0, 0, time.UTC), Precision: model.MINUTE}),
},
}

for _, tc := range tests {
Expand Down
Loading