minimal CLI app for comparison (close #8)

config/config.go

@@ -1,12 +1,23 @@
 package config
 
+import "github.com/gnames/gnfmt"
+
 type Config struct {
+	Format gnfmt.Format
 }
 
 type Option func(*Config)
 
+func OptFormat(f gnfmt.Format) Option {
+	return func(cfg *Config) {
+		cfg.Format = f
+	}
+}
+
 func New(opts ...Option) Config {
-	res := Config{}
+	res := Config{
+		Format: gnfmt.CSV,
+	}
 	for _, opt := range opts {
 		opt(&res)
 	}

ent/output/output.go

@@ -40,7 +40,10 @@ func MatchOutput(o Output, f gnfmt.Format) string {
 
 // CSVHeader returns the header string for CSV output format.
 func CSVHeader(f gnfmt.Format) string {
-	header := []string{"Source", "SourceRow", "Id", "Name", "MatchType", "Reference", "ReferenceRow", "Id", "Name", "EditDistance"}
+	header := []string{
+		"Source", "SourceRow", "Id",
+		"Name", "Reference", "MatchType",
+		"ReferenceRow", "Id", "Name", "EditDistance"}
 	switch f {
 	case gnfmt.CSV:
 		return gnfmt.ToCSV(header, ',')
@@ -54,15 +57,15 @@ func CSVHeader(f gnfmt.Format) string {
 func csvOutput(o Output, sep rune) string {
 	var res []string
 	for i := range o.Matches {
-		rows := csvRow(o.Matches[i], i, sep)
+		rows := csvRow(o.Matches[i], sep)
 		for i := range rows {
 			res = append(res, rows[i])
 		}
 	}
 	return strings.Join(res, "\n")
 }
 
-func csvRow(m Match, i int, sep rune) []string {
+func csvRow(m Match, sep rune) []string {
 	var res []string
 	s := m.SourceRecord
 	r := m.ReferenceRecords
@@ -72,8 +75,8 @@ func csvRow(m Match, i int, sep rune) []string {
 			strconv.Itoa(s.Index),
 			s.ID,
 			s.Name,
-			s.MatchType.String(),
 			r[i].DataSet,
+			r[i].MatchType.String(),
 			strconv.Itoa(r[i].Index),
 			r[i].ID,
 			r[i].Name,

gndiff/cmd/root.go

@@ -23,16 +23,22 @@ package cmd
 
 import (
 	"fmt"
-	"log"
 	"os"
+	"path/filepath"
 
 	"github.com/gnames/gndiff"
+	"github.com/gnames/gndiff/config"
+	"github.com/gnames/gndiff/ent/output"
+	"github.com/gnames/gndiff/ent/record"
+	"github.com/gnames/gndiff/io/ingestio"
+	"github.com/gnames/gnfmt"
+	log "github.com/sirupsen/logrus"
 	"github.com/spf13/cobra"
 
 	"github.com/spf13/viper"
 )
 
-var cfgFile string
+var quiet bool
 
 // rootCmd represents the base command when called without any subcommands
 var rootCmd = &cobra.Command{
@@ -45,10 +51,31 @@ prints out a match of a reference data to the source data.
 	// Uncomment the following line if your bare application
 	// has an action associated with it:
 	Run: func(cmd *cobra.Command, args []string) {
+		var opts []config.Option
 		if versionFlag(cmd) {
 			os.Exit(0)
 		}
 
+		quiet, _ = cmd.Flags().GetBool("quiet")
+
+		opts = append(opts, fmtFlag(cmd))
+		cfg := config.New(opts...)
+		gnd := gndiff.New(cfg)
+		if len(args) != 2 {
+			log.Warn("Supply paths to two CSV/TSV files for comparison")
+			_ = cmd.Help()
+			os.Exit(1)
+		}
+		src, ref, err := readFiles(args[0], args[1])
+		if err != nil {
+			log.Fatal(err)
+		}
+		res, err := gnd.Compare(src, ref)
+		if err != nil {
+			log.Fatal(err)
+		}
+
+		fmt.Print(output.MatchOutput(res, cfg.Format))
 	},
 }
 
@@ -68,6 +95,8 @@ func init() {
 		"'csv', 'tsv', 'compact', 'pretty'\n" +
 		"default is 'csv'."
 	rootCmd.Flags().StringP("format", "f", "", formatHelp)
+
+	rootCmd.Flags().BoolP("quiet", "q", false, "Do not output info and warning logs.")
 }
 
 // initConfig reads in config file and ENV variables if set.
@@ -92,3 +121,45 @@ func versionFlag(cmd *cobra.Command) bool {
 	}
 	return false
 }
+
+func fmtFlag(cmd *cobra.Command) config.Option {
+	f, err := cmd.Flags().GetString("format")
+	if f == "" || err != nil {
+		return config.OptFormat(gnfmt.CSV)
+	}
+
+	switch f {
+	case "csv":
+		return config.OptFormat(gnfmt.CSV)
+	case "tsv":
+		return config.OptFormat(gnfmt.TSV)
+	case "compact":
+		return config.OptFormat(gnfmt.CompactJSON)
+	case "pretty":
+		return config.OptFormat(gnfmt.PrettyJSON)
+	default:
+		if !quiet {
+			log.Warnf("Cannot recognize format '%s', keeping default 'CSV' format.", f)
+		}
+		return config.OptFormat(gnfmt.CSV)
+	}
+}
+
+func readFiles(srcPath, refPath string) ([]record.Record, []record.Record, error) {
+	cfg := config.New()
+	ing := ingestio.New(cfg)
+
+	src := filepath.Join(srcPath)
+	recSrc, err := ing.Records(src)
+	if err != nil {
+		return nil, nil, err
+	}
+
+	ref := filepath.Join(refPath)
+	recRef, err := ing.Records(ref)
+	if err != nil {
+		return nil, nil, err
+	}
+
+	return recSrc, recRef, nil
+}

go.mod

@@ -11,7 +11,9 @@ require (
 	github.com/gnames/gnparser v1.5.7
 	github.com/gnames/gnsys v0.2.2
 	github.com/gnames/levenshtein v0.2.1
+	github.com/sirupsen/logrus v1.7.0
 	github.com/spf13/cobra v1.3.0
+	github.com/spf13/viper v1.10.0
 	github.com/stretchr/testify v1.7.0
 	golang.org/x/perf v0.0.0-20211012211434-03971e389cd3
 	golang.org/x/tools v0.1.8
@@ -46,7 +48,6 @@ require (
 	github.com/spf13/cast v1.4.1 // indirect
 	github.com/spf13/jwalterweatherman v1.1.0 // indirect
 	github.com/spf13/pflag v1.0.5 // indirect
-	github.com/spf13/viper v1.10.0 // indirect
 	github.com/subosito/gotenv v1.2.0 // indirect
 	github.com/tmthrgd/atomics v0.0.0-20180217065130-6910de195248 // indirect
 	github.com/tmthrgd/go-bitset v0.0.0-20180828125936-62ad9ed7ff29 // indirect

go.sum

@@ -555,6 +555,7 @@ github.com/shurcooL/sanitized_anchor_name v1.0.0/go.mod h1:1NzhyTcUVG4SuEtjjoZeV
 github.com/sirupsen/logrus v1.0.6/go.mod h1:pMByvHTf9Beacp5x1UXfOR9xyW/9antXMhjMPG0dEzc=
 github.com/sirupsen/logrus v1.2.0/go.mod h1:LxeOpSwHxABJmUn/MG1IvRgCAasNZTLOkJPxbbu5VWo=
 github.com/sirupsen/logrus v1.4.2/go.mod h1:tLMulIdttU9McNUspp0xgXVQah82FyeX6MwdIuYE2rE=
+github.com/sirupsen/logrus v1.7.0 h1:ShrD1U9pZB12TX0cVy0DtePoCH97K8EtX+mg7ZARUtM=
 github.com/sirupsen/logrus v1.7.0/go.mod h1:yWOB1SBYBC5VeMP7gHvWumXLIWorT60ONWic61uBYv0=
 github.com/smartystreets/assertions v0.0.0-20180820201707-7c9eb446e3cf/go.mod h1:OnSkiWE9lh6wB0YB77sQom3nweQdgAjqCqsofrRNTgc=
 github.com/smartystreets/assertions v0.0.0-20180927180507-b2de0cb4f26d/go.mod h1:OnSkiWE9lh6wB0YB77sQom3nweQdgAjqCqsofrRNTgc=