miller/pkg/transformers/count.go

package transformers

import (
	"fmt"
	"os"
	"strings"

	"github.com/johnkerl/miller/v6/pkg/cli"
	"github.com/johnkerl/miller/v6/pkg/lib"
	"github.com/johnkerl/miller/v6/pkg/mlrval"
	"github.com/johnkerl/miller/v6/pkg/types"
)

const verbNameCount = "count"

var countOptions = []OptionSpec{
	{Flag: "-g", Arg: "{a,b,c}", Type: "csv-list", Desc: "Optional group-by-field names for counts, e.g. a,b,c"},
	{Flag: "-n", Type: "bool", Desc: "Show only the number of distinct values. Not interesting without -g."},
	{Flag: "-o", Arg: "{name}", Type: "string", Desc: "Field name for output-count. Default \"count\"."},
}

var CountSetup = TransformerSetup{
	Verb:         verbNameCount,
	UsageFunc:    transformerCountUsage,
	ParseCLIFunc: transformerCountParseCLI,
	IgnoresInput: false,
	Options:      countOptions,
}

func transformerCountUsage(
	o *os.File,
) {
	fmt.Fprintf(o, "Usage: %s %s [options]\n", "mlr", verbNameCount)
	fmt.Fprint(o,
		`Prints number of records, optionally grouped by distinct values for specified field names.
`)
	WriteVerbOptions(o, countOptions)
}

func transformerCountParseCLI(
	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++

	var groupByFieldNames []string = nil
	showCountsOnly := false
	outputFieldName := "count"

	var err error
	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":
			transformerCountUsage(os.Stdout)
			return nil, cli.ErrHelpRequested

		case "-g":
			groupByFieldNames, err = cli.VerbGetStringArrayArg(verb, opt, args, &argi, argc)
			if err != nil {
				return nil, err
			}

		case "-n":
			showCountsOnly = true

		case "-o":
			outputFieldName, err = cli.VerbGetStringArg(verb, opt, args, &argi, argc)
			if err != nil {
				return nil, err
			}

		default:
			return nil, cli.VerbErrorf(verb, "option \"%s\" not recognized", opt)
		}
	}

	*pargi = argi
	if !doConstruct { // All transformers must do this for main command-line parsing
		return nil, nil
	}

	transformer, err := NewTransformerCount(
		groupByFieldNames,
		showCountsOnly,
		outputFieldName,
	)
	if err != nil {
		return nil, err
	}

	return transformer, nil
}

type TransformerCount struct {
	// input
	groupByFieldNames []string
	showCountsOnly    bool
	outputFieldName   string

	// state
	recordTransformerFunc RecordTransformerFunc
	ungroupedCount        int64
	// Example:
	// * Suppose group-by fields are a,b.
	// * One record has a=foo,b=bar
	// * Another record has a=baz,b=quux
	// * Map keys are strings "foo,bar" and "baz,quux".
	// * groupedCounts maps "foo,bar" to 1 and "baz,quux" to 1.
	// * groupByValues maps "foo,bar" to ["foo", "bar"] and "baz,quux" to ["baz", "quux"].
	groupedCounts  *lib.OrderedMap[int64]
	groupingValues *lib.OrderedMap[[]*mlrval.Mlrval]
}

func NewTransformerCount(
	groupByFieldNames []string,
	showCountsOnly bool,
	outputFieldName string,
) (*TransformerCount, error) {

	tr := &TransformerCount{
		groupByFieldNames: groupByFieldNames,
		showCountsOnly:    showCountsOnly,
		outputFieldName:   outputFieldName,

		ungroupedCount: 0,
		groupedCounts:  lib.NewOrderedMap[int64](),
		groupingValues: lib.NewOrderedMap[[]*mlrval.Mlrval](),
	}

	if groupByFieldNames == nil {
		tr.recordTransformerFunc = tr.countUngrouped
	} else {
		tr.recordTransformerFunc = tr.countGrouped
	}

	return tr, nil
}

func (tr *TransformerCount) 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 *TransformerCount) countUngrouped(
	inrecAndContext *types.RecordAndContext,
	outputRecordsAndContexts *[]*types.RecordAndContext, // list of *types.RecordAndContext
	inputDownstreamDoneChannel <-chan bool,
	outputDownstreamDoneChannel chan<- bool,
) error {
	if !inrecAndContext.EndOfStream {
		tr.ungroupedCount++
	} else {
		newrec := mlrval.NewMlrmapAsRecord()
		newrec.PutCopy(tr.outputFieldName, mlrval.FromInt(tr.ungroupedCount))
		*outputRecordsAndContexts = append(*outputRecordsAndContexts, types.NewRecordAndContext(newrec, &inrecAndContext.Context))

		*outputRecordsAndContexts = append(*outputRecordsAndContexts, inrecAndContext) // end-of-stream marker
	}
	return nil
}

func (tr *TransformerCount) countGrouped(
	inrecAndContext *types.RecordAndContext,
	outputRecordsAndContexts *[]*types.RecordAndContext, // list of *types.RecordAndContext
	inputDownstreamDoneChannel <-chan bool,
	outputDownstreamDoneChannel chan<- bool,
) error {
	if !inrecAndContext.EndOfStream {
		inrec := inrecAndContext.Record

		groupingKey, selectedValues, ok := inrec.GetSelectedValuesAndJoined(
			tr.groupByFieldNames,
		)
		if !ok { // Current record does not have specified fields; ignore
			return nil
		}

		if !tr.groupedCounts.Has(groupingKey) {
			var count int64 = 1
			tr.groupedCounts.Put(groupingKey, count)
			tr.groupingValues.Put(groupingKey, selectedValues)
		} else {
			tr.groupedCounts.Put(
				groupingKey,
				tr.groupedCounts.Get(groupingKey)+1,
			)
		}

	} else {
		if tr.showCountsOnly {
			newrec := mlrval.NewMlrmapAsRecord()
			newrec.PutCopy(tr.outputFieldName, mlrval.FromInt(tr.groupedCounts.FieldCount))

			outrecAndContext := types.NewRecordAndContext(newrec, &inrecAndContext.Context)
			*outputRecordsAndContexts = append(*outputRecordsAndContexts, outrecAndContext)

		} else {
			for outer := tr.groupedCounts.Head; outer != nil; outer = outer.Next {
				groupingKey := outer.Key
				newrec := mlrval.NewMlrmapAsRecord()

				// Example:
				// * Suppose group-by fields are a,b.
				// * Record has a=foo,b=bar
				// * Grouping key is "foo,bar"
				// * Grouping values for key is ["foo", "bar"]
				// Here we populate a record with "a=foo,b=bar".

				groupingValuesForKey := tr.groupingValues.Get(groupingKey)
				i := 0
				for _, groupingValueForKey := range groupingValuesForKey {
					newrec.PutCopy(tr.groupByFieldNames[i], groupingValueForKey)
					i++
				}

				countForGroup := outer.Value
				newrec.PutCopy(tr.outputFieldName, mlrval.FromInt(countForGroup))

				outrecAndContext := types.NewRecordAndContext(newrec, &inrecAndContext.Context)
				*outputRecordsAndContexts = append(*outputRecordsAndContexts, outrecAndContext)
			}
		}

		*outputRecordsAndContexts = append(*outputRecordsAndContexts, inrecAndContext) // end-of-stream marker
	}
	return nil
}