You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@arrow.apache.org by GitBox <gi...@apache.org> on 2023/01/18 14:21:42 UTC
[GitHub] [arrow] lidavidm commented on a diff in pull request #14114: ARROW-17709: [Go] Array Builder for REE arrays
lidavidm commented on code in PR #14114:
URL: https://github.com/apache/arrow/pull/14114#discussion_r1073589805
##########
go/arrow/array/encoded.go:
##########
@@ -156,3 +158,160 @@ func arrayRunEndEncodedApproxEqual(l, r *RunEndEncoded, opt equalOption) bool {
}
return true
}
+
+type RunEndEncodedBuilder struct {
+ builder
+
+ dt arrow.DataType
+ runEnds Builder
+ values Builder
+ maxRunEnd uint64
+}
+
+func NewRunEndEncodedBuilder(mem memory.Allocator, runEnds, encoded arrow.DataType) *RunEndEncodedBuilder {
+ dt := arrow.RunEndEncodedOf(runEnds, encoded)
+ if !dt.ValidRunEndsType(runEnds) {
+ panic("arrow/ree: invalid runEnds type for run length encoded array")
+ }
+
+ var maxEnd uint64
+ switch runEnds.ID() {
+ case arrow.INT16:
+ maxEnd = math.MaxInt16
+ case arrow.INT32:
+ maxEnd = math.MaxInt32
+ case arrow.INT64:
+ maxEnd = math.MaxInt64
+ }
+ return &RunEndEncodedBuilder{
+ builder: builder{refCount: 1, mem: mem},
+ dt: dt,
+ runEnds: NewBuilder(mem, runEnds),
+ values: NewBuilder(mem, encoded),
+ maxRunEnd: maxEnd,
+ }
+}
+
+func (b *RunEndEncodedBuilder) Type() arrow.DataType {
+ return b.dt
+}
+
+func (b *RunEndEncodedBuilder) Release() {
+ debug.Assert(atomic.LoadInt64(&b.refCount) > 0, "too many releases")
+
+ if atomic.AddInt64(&b.refCount, -1) == 0 {
+ b.values.Release()
+ b.runEnds.Release()
+ }
+}
+
+func (b *RunEndEncodedBuilder) addLength(n uint64) {
+ if uint64(b.length)+n > b.maxRunEnd {
+ panic(fmt.Errorf("%w: %s array length must fit be less than %d", arrow.ErrInvalid, b.dt, b.maxRunEnd))
+ }
+
+ b.length += int(n)
+}
+
+func (b *RunEndEncodedBuilder) finishRun() {
+ if b.length == 0 {
+ return
+ }
+
+ switch bldr := b.runEnds.(type) {
+ case *Int16Builder:
+ bldr.Append(int16(b.length))
+ case *Int32Builder:
+ bldr.Append(int32(b.length))
+ case *Int64Builder:
+ bldr.Append(int64(b.length))
+ }
+}
+
+func (b *RunEndEncodedBuilder) ValueBuilder() Builder { return b.values }
+func (b *RunEndEncodedBuilder) Append(n uint64) {
+ b.finishRun()
+ b.addLength(n)
+}
+func (b *RunEndEncodedBuilder) ContinueRun(n uint64) {
+ b.addLength(n)
+}
+func (b *RunEndEncodedBuilder) AppendNull() {
Review Comment:
Would it be useful to have an AppendNulls as well?
##########
go/arrow/array/encoded.go:
##########
@@ -156,3 +158,160 @@ func arrayRunEndEncodedApproxEqual(l, r *RunEndEncoded, opt equalOption) bool {
}
return true
}
+
+type RunEndEncodedBuilder struct {
+ builder
+
+ dt arrow.DataType
+ runEnds Builder
+ values Builder
+ maxRunEnd uint64
+}
+
+func NewRunEndEncodedBuilder(mem memory.Allocator, runEnds, encoded arrow.DataType) *RunEndEncodedBuilder {
+ dt := arrow.RunEndEncodedOf(runEnds, encoded)
+ if !dt.ValidRunEndsType(runEnds) {
+ panic("arrow/ree: invalid runEnds type for run length encoded array")
+ }
+
+ var maxEnd uint64
+ switch runEnds.ID() {
+ case arrow.INT16:
+ maxEnd = math.MaxInt16
+ case arrow.INT32:
+ maxEnd = math.MaxInt32
+ case arrow.INT64:
+ maxEnd = math.MaxInt64
+ }
+ return &RunEndEncodedBuilder{
+ builder: builder{refCount: 1, mem: mem},
+ dt: dt,
+ runEnds: NewBuilder(mem, runEnds),
+ values: NewBuilder(mem, encoded),
+ maxRunEnd: maxEnd,
+ }
+}
+
+func (b *RunEndEncodedBuilder) Type() arrow.DataType {
+ return b.dt
+}
+
+func (b *RunEndEncodedBuilder) Release() {
+ debug.Assert(atomic.LoadInt64(&b.refCount) > 0, "too many releases")
+
+ if atomic.AddInt64(&b.refCount, -1) == 0 {
+ b.values.Release()
+ b.runEnds.Release()
+ }
+}
+
+func (b *RunEndEncodedBuilder) addLength(n uint64) {
+ if uint64(b.length)+n > b.maxRunEnd {
+ panic(fmt.Errorf("%w: %s array length must fit be less than %d", arrow.ErrInvalid, b.dt, b.maxRunEnd))
+ }
+
+ b.length += int(n)
+}
+
+func (b *RunEndEncodedBuilder) finishRun() {
+ if b.length == 0 {
+ return
+ }
+
+ switch bldr := b.runEnds.(type) {
+ case *Int16Builder:
+ bldr.Append(int16(b.length))
+ case *Int32Builder:
+ bldr.Append(int32(b.length))
+ case *Int64Builder:
+ bldr.Append(int64(b.length))
+ }
+}
+
+func (b *RunEndEncodedBuilder) ValueBuilder() Builder { return b.values }
+func (b *RunEndEncodedBuilder) Append(n uint64) {
+ b.finishRun()
+ b.addLength(n)
+}
+func (b *RunEndEncodedBuilder) ContinueRun(n uint64) {
+ b.addLength(n)
+}
+func (b *RunEndEncodedBuilder) AppendNull() {
Review Comment:
I suppose you can always do that in terms of ValueBuilder and Append though
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org