mirror of
https://github.com/johnkerl/miller.git
synced 2026-07-17 16:38:54 +00:00
Phase 3 of plans/exit.md: the streaming-interface change, the load-bearing piece for #341 (DSL exit statement) and #440 (strict mode). - RecordTransformer.Transform and RecordTransformerFunc now return error. All 69 Transform implementations and their dispatch helpers updated (mechanical rewrite, compiler- and errcheck-verified). - runSingleTransformerBatch, on a Transform error, forwards any output produced before the failure plus an end-of-stream marker downstream, so the rest of the chain and the record-writer drain and finish cleanly; runSingleTransformer then surfaces the error to stream.Stream's select loop (non-blocking send; first error wins) and signals upstream-done so the record-reader stops. This is exactly the flush-then-exit sequencing a future DSL 'exit N' needs. - dataProcessingErrorChannel and FileOutputHandler.recordErroredChannel are now chan error instead of chan bool; ChannelWriter still prints write- error details at the site and sends the 'exiting due to data error' sentinel, preserving the exact stderr shape pinned by regression cases. - Mid-stream os.Exit sites converted to returned errors: put/filter DSL begin/main/end-block errors and the non-boolean filter-expression case, tee write/close failures, split write/open/close failures, join left-file ingest failures (both half-streaming and sorted paths, with full error plumbing through JoinBucketKeeper), histogram/stats2 ingest errors, surv fit errors, and step stepper allocation (tStepperAllocator now returns (tStepper, error); bad EWMA coefficients propagate; negative slwin parameters are reported by the CLI parser via the existing bad-stepper-name pattern). - The two genuinely internal join-bucket-keeper states now use lib.InternalCodingErrorWithMessageIf instead of hand-rolled print+exit. - pkg/transformers is now os.Exit-free. Behavior notes: the non-boolean filter message gains the standard 'mlr: ' prefix and a newline (it previously printed with neither); tee errors now include the underlying cause. All 4779 regression cases pass unchanged; mlr head early-out latency is unaffected (0.02s over 50M records). Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
247 lines
6.9 KiB
Go
247 lines
6.9 KiB
Go
package transformers
|
|
|
|
import (
|
|
"fmt"
|
|
"os"
|
|
"strings"
|
|
|
|
"github.com/johnkerl/miller/v6/pkg/cli"
|
|
"github.com/johnkerl/miller/v6/pkg/types"
|
|
)
|
|
|
|
const verbNameHead = "head"
|
|
|
|
var headOptions = []OptionSpec{
|
|
{Flag: "-g", Arg: "{a,b,c}", Type: "csv-list", Desc: "Optional group-by-field names for head counts, e.g. a,b,c."},
|
|
{Flag: "-n", Arg: "{n}", Type: "int", Desc: "Head-count to print. Default 10. A negative count, e.g. -n -2, passes through all but the last n records, optionally by category."},
|
|
}
|
|
|
|
var HeadSetup = TransformerSetup{
|
|
Verb: verbNameHead,
|
|
UsageFunc: transformerHeadUsage,
|
|
ParseCLIFunc: transformerHeadParseCLI,
|
|
IgnoresInput: false,
|
|
Options: headOptions,
|
|
}
|
|
|
|
func transformerHeadUsage(
|
|
o *os.File,
|
|
) {
|
|
fmt.Fprintf(o, "Usage: %s %s [options]\n", "mlr", verbNameHead)
|
|
fmt.Fprintf(o, "Passes through the first n records, optionally by category.\n")
|
|
fmt.Fprintf(o, "Without -g, ceases consuming more input (i.e. is fast) when n records have been read.\n")
|
|
WriteVerbOptions(o, headOptions)
|
|
}
|
|
|
|
func transformerHeadParseCLI(
|
|
pargi *int,
|
|
argc int,
|
|
args []string,
|
|
_ *cli.TOptions,
|
|
doConstruct bool, // false for first pass of CLI-parse, true for second pass
|
|
) (RecordTransformer, error) {
|
|
|
|
// Skip the verb name from the current spot in the mlr command line
|
|
argi := *pargi
|
|
verb := args[argi]
|
|
argi++
|
|
|
|
headCount := int64(10)
|
|
var groupByFieldNames []string = nil
|
|
|
|
for argi < argc /* variable increment: 1 or 2 depending on flag */ {
|
|
opt := args[argi]
|
|
if !strings.HasPrefix(opt, "-") {
|
|
break // No more flag options to process
|
|
}
|
|
if args[argi] == "--" {
|
|
break // All transformers must do this so main-flags can follow verb-flags
|
|
}
|
|
argi++
|
|
|
|
switch opt {
|
|
case "-h", "--help":
|
|
transformerHeadUsage(os.Stdout)
|
|
return nil, cli.ErrHelpRequested
|
|
|
|
case "-n":
|
|
n, err := cli.VerbGetIntArg(verb, opt, args, &argi, argc)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
headCount = n
|
|
|
|
case "-g":
|
|
names, err := cli.VerbGetStringArrayArg(verb, opt, args, &argi, argc)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
groupByFieldNames = names
|
|
|
|
default:
|
|
transformerHeadUsage(os.Stderr)
|
|
return nil, fmt.Errorf("%s %s: option \"%s\" not recognized", "mlr", verb, opt)
|
|
}
|
|
}
|
|
|
|
*pargi = argi
|
|
if !doConstruct { // All transformers must do this for main command-line parsing
|
|
return nil, nil
|
|
}
|
|
|
|
transformer, err := NewTransformerHead(
|
|
headCount,
|
|
groupByFieldNames,
|
|
)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
return transformer, nil
|
|
}
|
|
|
|
type TransformerHead struct {
|
|
// input
|
|
headCount int64
|
|
groupByFieldNames []string
|
|
|
|
// state
|
|
recordTransformerFunc RecordTransformerFunc
|
|
unkeyedRecordCount int64
|
|
keyedRecordCounts map[string]int64
|
|
recordListsByGroup map[string]*[]*types.RecordAndContext
|
|
|
|
// See ChainTransformer
|
|
wroteDownstreamDone bool
|
|
}
|
|
|
|
func NewTransformerHead(
|
|
headCount int64,
|
|
groupByFieldNames []string,
|
|
) (*TransformerHead, error) {
|
|
|
|
allButLast := headCount < 0
|
|
if allButLast {
|
|
headCount = -headCount
|
|
}
|
|
|
|
tr := &TransformerHead{
|
|
headCount: headCount,
|
|
groupByFieldNames: groupByFieldNames,
|
|
unkeyedRecordCount: 0,
|
|
keyedRecordCounts: make(map[string]int64),
|
|
recordListsByGroup: make(map[string]*[]*types.RecordAndContext),
|
|
wroteDownstreamDone: false,
|
|
}
|
|
|
|
if allButLast {
|
|
tr.recordTransformerFunc = tr.transformAllButLast
|
|
} else if groupByFieldNames == nil {
|
|
tr.recordTransformerFunc = tr.transformUnkeyed
|
|
} else {
|
|
tr.recordTransformerFunc = tr.transformKeyed
|
|
}
|
|
|
|
return tr, nil
|
|
}
|
|
|
|
func (tr *TransformerHead) Transform(
|
|
inrecAndContext *types.RecordAndContext,
|
|
outputRecordsAndContexts *[]*types.RecordAndContext, // list of *types.RecordAndContext
|
|
inputDownstreamDoneChannel <-chan bool,
|
|
outputDownstreamDoneChannel chan<- bool,
|
|
) error {
|
|
HandleDefaultDownstreamDone(inputDownstreamDoneChannel, outputDownstreamDoneChannel)
|
|
return tr.recordTransformerFunc(inrecAndContext, outputRecordsAndContexts, inputDownstreamDoneChannel, outputDownstreamDoneChannel)
|
|
}
|
|
|
|
func (tr *TransformerHead) transformUnkeyed(
|
|
inrecAndContext *types.RecordAndContext,
|
|
outputRecordsAndContexts *[]*types.RecordAndContext, // list of *types.RecordAndContext
|
|
inputDownstreamDoneChannel <-chan bool,
|
|
outputDownstreamDoneChannel chan<- bool,
|
|
) error {
|
|
if !inrecAndContext.EndOfStream {
|
|
tr.unkeyedRecordCount++
|
|
if tr.unkeyedRecordCount <= tr.headCount {
|
|
*outputRecordsAndContexts = append(*outputRecordsAndContexts, inrecAndContext)
|
|
} else if !tr.wroteDownstreamDone {
|
|
// Signify to data producers upstream that we'll ignore further
|
|
// data, so as far as we're concerned they can stop sending it. See
|
|
// ChainTransformer.
|
|
//TODO: maybe remove: *outputRecordsAndContexts = append(*outputRecordsAndContexts, types.NewEndOfStreamMarker(&inrecAndContext.Context))
|
|
outputDownstreamDoneChannel <- true
|
|
tr.wroteDownstreamDone = true
|
|
}
|
|
} else {
|
|
*outputRecordsAndContexts = append(*outputRecordsAndContexts, inrecAndContext)
|
|
}
|
|
return nil
|
|
}
|
|
|
|
func (tr *TransformerHead) transformKeyed(
|
|
inrecAndContext *types.RecordAndContext,
|
|
outputRecordsAndContexts *[]*types.RecordAndContext, // list of *types.RecordAndContext
|
|
inputDownstreamDoneChannel <-chan bool,
|
|
outputDownstreamDoneChannel chan<- bool,
|
|
) error {
|
|
if !inrecAndContext.EndOfStream {
|
|
inrec := inrecAndContext.Record
|
|
|
|
groupingKey, ok := inrec.GetSelectedValuesJoined(tr.groupByFieldNames)
|
|
if !ok {
|
|
return nil
|
|
}
|
|
|
|
count, present := tr.keyedRecordCounts[groupingKey]
|
|
if !present { // first time
|
|
tr.keyedRecordCounts[groupingKey] = 1
|
|
count = 1
|
|
} else {
|
|
tr.keyedRecordCounts[groupingKey] += 1
|
|
count += 1
|
|
}
|
|
|
|
if count <= tr.headCount {
|
|
*outputRecordsAndContexts = append(*outputRecordsAndContexts, inrecAndContext)
|
|
}
|
|
|
|
} else {
|
|
*outputRecordsAndContexts = append(*outputRecordsAndContexts, inrecAndContext)
|
|
}
|
|
return nil
|
|
}
|
|
|
|
func (tr *TransformerHead) transformAllButLast(
|
|
inrecAndContext *types.RecordAndContext,
|
|
outputRecordsAndContexts *[]*types.RecordAndContext, // list of *types.RecordAndContext
|
|
inputDownstreamDoneChannel <-chan bool,
|
|
outputDownstreamDoneChannel chan<- bool,
|
|
) error {
|
|
if !inrecAndContext.EndOfStream {
|
|
inrec := inrecAndContext.Record
|
|
|
|
groupingKey, ok := inrec.GetSelectedValuesJoined(tr.groupByFieldNames)
|
|
if !ok {
|
|
return nil
|
|
}
|
|
|
|
recordListForGroup := tr.recordListsByGroup[groupingKey]
|
|
if recordListForGroup == nil { // first time
|
|
recordListForGroup = &[]*types.RecordAndContext{}
|
|
tr.recordListsByGroup[groupingKey] = recordListForGroup
|
|
}
|
|
|
|
*recordListForGroup = append(*recordListForGroup, inrecAndContext)
|
|
for int64(len(*recordListForGroup)) > tr.headCount {
|
|
// Emit records that have fallen out of the window.
|
|
*outputRecordsAndContexts = append(*outputRecordsAndContexts, (*recordListForGroup)[0])
|
|
(*recordListForGroup)[0] = nil // release the backing-array slot's reference
|
|
*recordListForGroup = (*recordListForGroup)[1:]
|
|
}
|
|
|
|
} else {
|
|
*outputRecordsAndContexts = append(*outputRecordsAndContexts, inrecAndContext)
|
|
}
|
|
return nil
|
|
}
|