miller/pkg/input/record_reader_dkvpx.go
John Kerl d2acbfeea3 Batch-allocate per-record objects; reuse CSV writer field buffer
After batch-arena field allocation, profiling cat over 1M-record CSV showed
the remaining ~5M allocations were almost entirely per-record (one each):
the Mlrmap struct, the RecordAndContext wrapper, the CSV writer's []string,
and the go-csv parser's own buffers.

Address the first three:

- mlrval.RecordArena gains NewRecord(), vending the Mlrmap struct itself from
  a per-batch slab (respecting --no-hash-records). Rolled out to every
  line-based reader (CSV, CSV-lite, TSV, DKVP, NIDX, PPRINT, XTAB, DKVPX) in
  place of NewMlrmapAsRecord.

- The CSV reader batch-allocates RecordAndContext wrappers from a per-batch
  slab instead of one heap object per record (comment/output-string entries
  still allocate individually, but they are rare).

- RecordWriterCSV reuses a single fieldsBuffer []string across records instead
  of allocating one per Write; WriteCSVRecordMaybeColorized consumes it
  synchronously and the writer is single-goroutine, so this is safe.

Effect (big.*, 1M records, cat, best of 5):
  csv   0.26 -> 0.22
  dkvp  0.51 -> 0.45  (Mlrmap slab)

For CSV, cat's allocation-object count drops ~5.0M -> ~2.1M. The remaining
~2M are the go-csv parser's per-record backing string and field slice, which
are intrinsic to parsing and would require a zero-copy/batch-slab parser
rework. A CPU profile of cat now shows it is I/O-bound (syscall ~56%, bufio
read+flush), with allocation/GC down to ~10% -- i.e. further allocation
trimming no longer moves cat's wall-clock. GOGC=off confirms (no change).

Verified: go test ./pkg/... and full regression suite pass; output is
byte-identical across all formats including record-retaining verbs (tac),
hashed and --no-hash-records.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-19 12:27:22 -04:00

185 lines
4.6 KiB
Go

// RecordReaderDKVPX reads DKVPX format: comma-delimited key=value pairs with
// CSV-style quoting. It uses the dkvpx package for parsing.
package input
import (
"fmt"
"io"
"github.com/johnkerl/miller/v6/pkg/cli"
"github.com/johnkerl/miller/v6/pkg/dkvpx"
"github.com/johnkerl/miller/v6/pkg/lib"
"github.com/johnkerl/miller/v6/pkg/mlrval"
"github.com/johnkerl/miller/v6/pkg/types"
)
type RecordReaderDKVPX struct {
readerOptions *cli.TReaderOptions
recordsPerBatch int64
}
func NewRecordReaderDKVPX(
readerOptions *cli.TReaderOptions,
recordsPerBatch int64,
) (*RecordReaderDKVPX, error) {
if readerOptions.IRS != "\n" && readerOptions.IRS != "\r\n" {
return nil, fmt.Errorf("for DKVPX, IRS cannot be altered; LF vs CR/LF is autodetected")
}
return &RecordReaderDKVPX{
readerOptions: readerOptions,
recordsPerBatch: recordsPerBatch,
}, nil
}
func (reader *RecordReaderDKVPX) Read(
filenames []string,
context types.Context,
readerChannel chan<- []*types.RecordAndContext,
errorChannel chan error,
downstreamDoneChannel <-chan bool,
) {
if filenames != nil {
if len(filenames) == 0 {
handle, err := lib.OpenStdin(
reader.readerOptions.Prepipe,
reader.readerOptions.PrepipeIsRaw,
reader.readerOptions.FileInputEncoding,
)
if err != nil {
errorChannel <- err
} else {
reader.processHandle(handle, "(stdin)", &context, readerChannel, errorChannel, downstreamDoneChannel)
}
} else {
for _, filename := range filenames {
handle, err := lib.OpenFileForRead(
filename,
reader.readerOptions.Prepipe,
reader.readerOptions.PrepipeIsRaw,
reader.readerOptions.FileInputEncoding,
)
if err != nil {
errorChannel <- err
} else {
reader.processHandle(handle, filename, &context, readerChannel, errorChannel, downstreamDoneChannel)
handle.Close()
}
}
}
}
readerChannel <- types.NewEndOfStreamMarkerList(&context)
}
func (reader *RecordReaderDKVPX) processHandle(
handle io.Reader,
filename string,
context *types.Context,
readerChannel chan<- []*types.RecordAndContext,
errorChannel chan<- error,
downstreamDoneChannel <-chan bool,
) {
context.UpdateForStartOfFile(filename)
recordsPerBatch := reader.recordsPerBatch
dkvpxReader := dkvpx.NewReader(NewBOMStrippingReader(handle))
dkvpxReader.Comma = ','
if reader.readerOptions.CommentHandling != cli.CommentsAreData &&
len(reader.readerOptions.CommentString) == 1 {
dkvpxReader.Comment = rune(reader.readerOptions.CommentString[0])
}
dkvpxRecordsChannel := make(chan []*lib.OrderedMap[string], recordsPerBatch)
go channelizedDKVPXRecordScanner(dkvpxReader, dkvpxRecordsChannel, downstreamDoneChannel, errorChannel, recordsPerBatch)
for {
recordsAndContexts, eof := reader.getRecordBatch(dkvpxRecordsChannel, errorChannel, context)
if len(recordsAndContexts) > 0 {
readerChannel <- recordsAndContexts
}
if eof {
break
}
}
}
func channelizedDKVPXRecordScanner(
dkvpxReader *dkvpx.Reader,
dkvpxRecordsChannel chan<- []*lib.OrderedMap[string],
downstreamDoneChannel <-chan bool,
errorChannel chan<- error,
recordsPerBatch int64,
) {
i := int64(0)
done := false
dkvpxRecords := make([]*lib.OrderedMap[string], 0, recordsPerBatch)
for {
i++
dkvpxRecord, err := dkvpxReader.Read()
if lib.IsEOF(err) {
break
}
if err != nil {
errorChannel <- err
break
}
dkvpxRecords = append(dkvpxRecords, dkvpxRecord)
if i%recordsPerBatch == 0 {
select {
case <-downstreamDoneChannel:
done = true
break
default:
break
}
if done {
break
}
dkvpxRecordsChannel <- dkvpxRecords
dkvpxRecords = make([]*lib.OrderedMap[string], 0, recordsPerBatch)
}
if done {
break
}
}
dkvpxRecordsChannel <- dkvpxRecords
close(dkvpxRecordsChannel)
}
func (reader *RecordReaderDKVPX) getRecordBatch(
dkvpxRecordsChannel <-chan []*lib.OrderedMap[string],
errorChannel chan<- error,
context *types.Context,
) ([]*types.RecordAndContext, bool) {
recordsAndContexts := []*types.RecordAndContext{}
dedupeFieldNames := reader.readerOptions.DedupeFieldNames
dkvpxRecords, more := <-dkvpxRecordsChannel
if !more {
return recordsAndContexts, true
}
nfields := 0
for _, omap := range dkvpxRecords {
nfields += int(omap.FieldCount)
}
arena := mlrval.NewRecordArena(nfields)
for _, omap := range dkvpxRecords {
record := arena.NewRecord()
for pe := omap.Head; pe != nil; pe = pe.Next {
arena.PutDeferred(record, pe.Key, pe.Value, dedupeFieldNames)
}
context.UpdateForInputRecord()
recordsAndContexts = append(recordsAndContexts, types.NewRecordAndContext(record, context))
}
return recordsAndContexts, false
}