miller/pkg/transformers/cut.go

package transformers

import (
	"cmp"
	"fmt"
	"os"
	"regexp"
	"slices"
	"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 verbNameCut = "cut"

var cutOptions = []OptionSpec{
	{Flag: "-f", Arg: "{a,b,c}", Type: "csv-list", Desc: "Comma-separated field names to include or exclude, e.g. a,b,c."},
	{Flag: "-o", Type: "bool", Desc: "Retain fields in the order specified by -f rather than in input-record order."},
	{Flag: "-x", Aliases: []string{"--complement"}, Type: "bool", Desc: "Exclude, rather than include, the field names specified by -f."},
	{Flag: "-r", Type: "bool", Desc: "Treat field names as regular expressions. \"ab\", \"a.*b\" will match any field name containing the substring \"ab\" or matching \"a.*b\", respectively; anchors of the form \"^ab$\", \"^a.*b$\" may be used."},
}

var CutSetup = TransformerSetup{
	Verb:         verbNameCut,
	UsageFunc:    transformerCutUsage,
	ParseCLIFunc: transformerCutParseCLI,
	IgnoresInput: false,
	Options:      cutOptions,
}

func transformerCutUsage(
	o *os.File,
) {
	fmt.Fprintf(o, "Usage: %s %s [options]\n", "mlr", verbNameCut)
	fmt.Fprintf(o, "Passes through input records with specified fields included/excluded.\n")
	WriteVerbOptions(o, cutOptions)
	fmt.Fprintf(o, "Examples:\n")
	fmt.Fprintf(o, "  %s %s -f hostname,status\n", "mlr", verbNameCut)
	fmt.Fprintf(o, "  %s %s -x -f hostname,status\n", "mlr", verbNameCut)
	fmt.Fprintf(o, "  %s %s -r -f '^status$,sda[0-9]'\n", "mlr", verbNameCut)
	fmt.Fprintf(o, "  %s %s -r -f '^status$,\"sda[0-9]\"'\n", "mlr", verbNameCut)
	fmt.Fprintf(o, "  %s %s -r -f '^status$,\"sda[0-9]\"i' (this is case-insensitive)\n", "mlr", verbNameCut)
}

func transformerCutParseCLI(
	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 err error
	var fieldNames []string = nil
	doArgOrder := false
	doComplement := false
	doRegexes := false

	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":
			transformerCutUsage(os.Stdout)
			return nil, cli.ErrHelpRequested

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

		case "-o":
			doArgOrder = true

		case "-x", "--complement":
			doComplement = true

		case "-r":
			doRegexes = true

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

	if fieldNames == nil {
		return nil, cli.VerbErrorf(verb, "-f field names required")
	}

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

	transformer, err := NewTransformerCut(
		fieldNames,
		doArgOrder,
		doComplement,
		doRegexes,
	)
	if err != nil {
		return nil, err
	}

	return transformer, nil
}

type TransformerCut struct {
	fieldNameList []string
	fieldNameSet  map[string]bool

	doComplement bool
	doArgOrder   bool
	regexes      []*regexp.Regexp

	recordTransformerFunc RecordTransformerFunc
}

func NewTransformerCut(
	fieldNames []string,
	doArgOrder bool,
	doComplement bool,
	doRegexes bool,
) (*TransformerCut, error) {

	tr := &TransformerCut{}

	tr.doArgOrder = doArgOrder

	if !doRegexes {
		tr.fieldNameList = fieldNames
		tr.fieldNameSet = lib.StringListToSet(fieldNames)
		if !doComplement {
			if !doArgOrder {
				tr.recordTransformerFunc = tr.includeWithInputOrder
			} else {
				tr.recordTransformerFunc = tr.includeWithArgOrder
			}
		} else {
			tr.recordTransformerFunc = tr.exclude
		}
	} else {
		tr.doComplement = doComplement
		tr.regexes = make([]*regexp.Regexp, len(fieldNames))
		for i, regexString := range fieldNames {
			// Handles "a.*b"i Miller case-insensitive-regex specification
			regex, err := lib.CompileMillerRegex(regexString)
			if err != nil {
				fmt.Fprintf(
					os.Stderr,
					"%s %s: cannot compile regex [%s]\n",
					"mlr", verbNameCut, regexString,
				)
				os.Exit(1)
			}
			tr.regexes[i] = regex
		}
		tr.recordTransformerFunc = tr.processWithRegexes
	}

	return tr, nil
}

func (tr *TransformerCut) Transform(
	inrecAndContext *types.RecordAndContext,
	outputRecordsAndContexts *[]*types.RecordAndContext, // list of *types.RecordAndContext
	inputDownstreamDoneChannel <-chan bool,
	outputDownstreamDoneChannel chan<- bool,
) {
	HandleDefaultDownstreamDone(inputDownstreamDoneChannel, outputDownstreamDoneChannel)
	tr.recordTransformerFunc(inrecAndContext, outputRecordsAndContexts, inputDownstreamDoneChannel, outputDownstreamDoneChannel)
}

// mlr cut -f a,b,c
func (tr *TransformerCut) includeWithInputOrder(
	inrecAndContext *types.RecordAndContext,
	outputRecordsAndContexts *[]*types.RecordAndContext, // list of *types.RecordAndContext
	inputDownstreamDoneChannel <-chan bool,
	outputDownstreamDoneChannel chan<- bool,
) {
	if !inrecAndContext.EndOfStream {
		inrec := inrecAndContext.Record
		outrec := mlrval.NewMlrmap()
		for pe := inrec.Head; pe != nil; pe = pe.Next {
			fieldName := pe.Key
			_, wanted := tr.fieldNameSet[fieldName]
			if wanted {
				outrec.PutReference(fieldName, pe.Value) // inrec will be GC'ed
			}
		}
		outrecAndContext := types.NewRecordAndContext(outrec, &inrecAndContext.Context)
		*outputRecordsAndContexts = append(*outputRecordsAndContexts, outrecAndContext)
	} else {
		*outputRecordsAndContexts = append(*outputRecordsAndContexts, inrecAndContext)
	}
}

// mlr cut -o -f a,b,c
func (tr *TransformerCut) includeWithArgOrder(
	inrecAndContext *types.RecordAndContext,
	outputRecordsAndContexts *[]*types.RecordAndContext, // list of *types.RecordAndContext
	inputDownstreamDoneChannel <-chan bool,
	outputDownstreamDoneChannel chan<- bool,
) {
	if !inrecAndContext.EndOfStream {
		inrec := inrecAndContext.Record
		outrec := mlrval.NewMlrmap()
		for _, fieldName := range tr.fieldNameList {
			value := inrec.Get(fieldName)
			if value != nil {
				outrec.PutReference(fieldName, value) // inrec will be GC'ed
			}
		}
		outrecAndContext := types.NewRecordAndContext(outrec, &inrecAndContext.Context)
		*outputRecordsAndContexts = append(*outputRecordsAndContexts, outrecAndContext)
	} else {
		*outputRecordsAndContexts = append(*outputRecordsAndContexts, inrecAndContext)
	}
}

// mlr cut -x -f a,b,c
func (tr *TransformerCut) exclude(
	inrecAndContext *types.RecordAndContext,
	outputRecordsAndContexts *[]*types.RecordAndContext, // list of *types.RecordAndContext
	inputDownstreamDoneChannel <-chan bool,
	outputDownstreamDoneChannel chan<- bool,
) {
	if !inrecAndContext.EndOfStream {
		inrec := inrecAndContext.Record
		for _, fieldName := range tr.fieldNameList {
			if inrec.Has(fieldName) {
				inrec.Remove(fieldName)
			}
		}
	}
	*outputRecordsAndContexts = append(*outputRecordsAndContexts, inrecAndContext)
}

type entryIndex struct {
	index int
	entry *mlrval.MlrmapEntry
}

func (tr *TransformerCut) processWithRegexes(
	inrecAndContext *types.RecordAndContext,
	outputRecordsAndContexts *[]*types.RecordAndContext, // list of *types.RecordAndContext
	inputDownstreamDoneChannel <-chan bool,
	outputDownstreamDoneChannel chan<- bool,
) {
	if !inrecAndContext.EndOfStream {
		inrec := inrecAndContext.Record
		newrec := mlrval.NewMlrmapAsRecord()
		var entries []entryIndex
		for pe := inrec.Head; pe != nil; pe = pe.Next {
			matchesAny := false
			var index int
			for i, regex := range tr.regexes {
				if regex.MatchString(pe.Key) {
					matchesAny = true
					index = i
					break
				}
			}
			// Boolean XOR is spelt '!=' in Go
			if matchesAny != tr.doComplement {
				// Pointer-motion is OK since the inrec is being hereby discarded.
				// We're simply transferring ownership to the newrec.
				if tr.doArgOrder {
					entries = append(entries, entryIndex{index, pe})
				} else {
					newrec.PutReference(pe.Key, pe.Value)
				}
			}
		}
		if tr.doArgOrder {
			slices.SortStableFunc(entries, func(a, b entryIndex) int {
				return cmp.Compare(a.index, b.index)
			})
			for _, ei := range entries {
				newrec.PutReference(ei.entry.Key, ei.entry.Value)
			}
		}
		*outputRecordsAndContexts = append(*outputRecordsAndContexts, types.NewRecordAndContext(newrec, &inrecAndContext.Context))
	} else {
		*outputRecordsAndContexts = append(*outputRecordsAndContexts, inrecAndContext)
	}
}