Add a Wikidata language option!

Though I don't think there are any formats labelled with another
language code yet.

Author: ross-spencer

cmd/roy/harvest_wikidata.go

@@ -12,7 +12,7 @@ import (
 )
 
 func harvestWikidata() {
-	log.Println("roy harvesting Wikidata definitions")
+	log.Printf("roy harvesting Wikidata definitions: lang '%s'", config.WikidataLang())
 	err := os.MkdirAll(config.WikidataHome(), os.ModePerm)
 	if err != nil {
 		fmt.Errorf("roy: error harvesting Wikidata definitions: %s", err)

cmd/roy/roy.go

@@ -156,14 +156,15 @@ var (
 	repetition  = build.Int("repetition", config.Repetition(), "define a maximum tolerable repetition in a segment, used in combination with cost to determine segmentation")
 
 	// HARVEST
-	harvest            = flag.NewFlagSet("harvest", flag.ExitOnError)
-	harvestHome        = harvest.String("home", config.Home(), "override the default home directory")
-	harvestDroid       = harvest.String("droid", config.Droid(), "set name/path for DROID signature file")
-	harvestChanges     = harvest.Bool("changes", false, "harvest the latest PRONOM release-notes.xml file")
-	_, htimeout, _, _  = config.HarvestOptions()
-	timeout            = harvest.Duration("timeout", htimeout, "set duration before timing-out harvesting requests e.g. 120s")
-	throttlef          = harvest.Duration("throttle", 0, "set a time to wait HTTP requests e.g. 50ms")
-	harvestWikidataSig = harvest.Bool("wikidata", false, "harvest a static Wikidata report")
+	harvest             = flag.NewFlagSet("harvest", flag.ExitOnError)
+	harvestHome         = harvest.String("home", config.Home(), "override the default home directory")
+	harvestDroid        = harvest.String("droid", config.Droid(), "set name/path for DROID signature file")
+	harvestChanges      = harvest.Bool("changes", false, "harvest the latest PRONOM release-notes.xml file")
+	_, htimeout, _, _   = config.HarvestOptions()
+	timeout             = harvest.Duration("timeout", htimeout, "set duration before timing-out harvesting requests e.g. 120s")
+	throttlef           = harvest.Duration("throttle", 0, "set a time to wait HTTP requests e.g. 50ms")
+	harvestWikidataSig  = harvest.Bool("wikidata", false, "harvest a static Wikidata report")
+	harvestWikidataLang = harvest.String("lang", config.WikidataLang(), "two-letter language-code to download Wikidata strings, e.g. DE (default EN)")
 
 	// INSPECT (roy inspect | roy inspect fmt/121 | roy inspect usr/local/mysig.sig | roy inspect 10)
 	inspect        = flag.NewFlagSet("inspect", flag.ExitOnError)
@@ -482,6 +483,9 @@ func setHarvestOptions() {
 	if *throttlef > 0 {
 		config.SetHarvestThrottle(*throttlef)
 	}
+	if *harvestWikidataLang != "" {
+		config.SetWikidataLang(*harvestWikidataLang)
+	}
 }
 
 func setSetsOptions() {

pkg/config/wikidata.go

@@ -3,6 +3,7 @@ package config
 import (
 	"os"
 	"path/filepath"
+	"strings"
 )
 
 var wikidata = struct {
@@ -11,6 +12,9 @@ var wikidata = struct {
 	nopronom    bool
 	filemode    os.FileMode
 
+	wikidataLang     string
+	languageTemplate string
+
 	endpoint string
 	sparql   string
 }{
@@ -20,6 +24,9 @@ var wikidata = struct {
 	name:     "wikidata",
 	filemode: 0644,
 
+	wikidataLang:     "en",
+	languageTemplate: "<<lang>>",
+
 	endpoint: "https://query.wikidata.org/sparql",
 	sparql: `
 		SELECT DISTINCT ?format ?formatLabel ?puid ?ldd ?extension ?mimetype ?sig ?referenceLabel ?date ?encodingLabel ?offset ?relativityLabel WHERE
@@ -44,7 +51,7 @@ var wikidata = struct {
 		     ?format p:P4152 ?object.
 		     ?object pq:P2210 ?relativity.
 		  }
-		  SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
+		  SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],<<lang>>". }
 		}
 		order by ?format`,
 }
@@ -84,5 +91,15 @@ func WikidataEndpoint() string {
 
 // WikidataSPARQL ...
 func WikidataSPARQL() string {
-	return wikidata.sparql
+	return strings.Replace(wikidata.sparql, wikidata.languageTemplate, wikidata.wikidataLang, 1)
+}
+
+// WikidataLang ...
+func WikidataLang() string {
+	return wikidata.wikidataLang
+}
+
+// SetWikidataLang ...
+func SetWikidataLang(lang string) {
+	wikidata.wikidataLang = lang
 }