cats en tags

NU/cmd/nu/nu.go

@@ -211,11 +211,11 @@ func doArticle(filename string, url string, timestamp time.Time, needUpdate bool
 		return false
 	}
 
-	cats := make([]string, 0)
+	tags := make([]string, 0)
 	var buffer bytes.Buffer
 	for _, i := range doc.Graph {
 		p(buffer.WriteString(html.UnescapeString(i.ArticleBody)))
-		cats = append(cats, i.ArticleSection...)
+		tags = append(tags, i.ArticleSection...)
 	}
 	text := buffer.String()
 
@@ -245,11 +245,11 @@ func doArticle(filename string, url string, timestamp time.Time, needUpdate bool
 
 		fp, err := os.Create(filename + ".txt")
 		p(err)
-		if len(cats) == 0 {
-			p(fmt.Fprintln(fp, "##META text cat ="))
+		if len(tags) == 0 {
+			p(fmt.Fprintln(fp, "##META text tag ="))
 		} else {
-			for _, cat := range cats {
-				p(fmt.Fprintf(fp, "##META text cat = %s\n", fixSpace(cat)))
+			for _, tag := range tags {
+				p(fmt.Fprintf(fp, "##META text tag = %s\n", fixSpace(tag)))
 			}
 		}
 		p(fp.WriteString(text))

NU/txt2corpus.sh

@@ -56,6 +56,9 @@ Alpino -flag treebank xml debug=1 end_hook=xml user_max=900000 -parse < $corpus.
 cd xml
 alto -o $corpus.data.dz *.xml 2> /dev/null
 
+# telling per bericht, niet per zin
+/net/corpora/nlnieuws/namen.sh -x 10 -s $corpus.data.dz > $corpus.tag.txt
+
 cd ../..
 rm -fr out