p209327/nlnieuws
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

cats en tags

Browse Source
This commit is contained in:
Peter Kleiweg
2026-04-01 19:30:10 +02:00
parent 1fb1867550
commit e550b58889
14 changed files with 61 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
VRT/cmd/vrt/vrt.go
View File

@@ -151,7 +151,9 @@ func main() {
}
}
func doArticle(filename string, url string, title string, tags []string, labels []string, timestamp time.Time, needUpdate bool) bool {
// Nstag -> tag
// Nslabeltag -> cat
func doArticle(filename string, url string, title string, tags []string, cats []string, timestamp time.Time, needUpdate bool) bool {
if exists(filename + ".skip") {
return true
}
@@ -236,18 +238,18 @@ func doArticle(filename string, url string, title string, tags []string, labels
}
}
if len(tags) == 0 {
if len(cats) == 0 {
p(fmt.Fprintln(&buf, "##META text cat ="))
} else {
for _, tag := range tags {
p(fmt.Fprintf(&buf, "##META text cat = %s\n", fixSpace(tag)))
for _, cat := range cats {
p(fmt.Fprintf(&buf, "##META text cat = %s\n", fixSpace(cat)))
}
}
if len(labels) == 0 {
p(fmt.Fprintln(&buf, "##META text label ="))
if len(tags) == 0 {
p(fmt.Fprintln(&buf, "##META text tag ="))
} else {
for _, label := range labels {
p(fmt.Fprintf(&buf, "##META text label = %s\n", fixSpace(label)))
for _, tag := range tags {
p(fmt.Fprintf(&buf, "##META text tag = %s\n", fixSpace(tag)))
}
}

4
VRT/txt2corpus.sh
View File

@@ -56,6 +56,10 @@ 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 9 -s $corpus.data.dz > $corpus.cat.txt
/net/corpora/nlnieuws/namen.sh -x 10 -s $corpus.data.dz > $corpus.tag.txt
cd ../..
rm -fr out