grote reorganisatie:

- HLN, NOS, NU, VRT: per week -> per dag
- yyyy-ww -> yyyy.ww
- yyyy*  -> yyyy/yyyy*
etc

NU/cmd/nu/nu.go

@@ -98,8 +98,7 @@ func main() {
 			t, err = time.Parse(time.RFC1123, item.PubDate)
 		}
 		p(err)
-		year, week := t.ISOWeek()
-		dirname := fmt.Sprintf("/net/corpora/nlnieuws/NU/%d/%02d", year, week)
+		dirname := fmt.Sprintf("/net/corpora/nlnieuws/NU/%d/%02d/%02d", t.Year(), int(t.Month()), t.Day())
 		if exists(dirname + "/lock") {
 			continue
 		}

NU/txt2corpus.sh

@@ -2,17 +2,20 @@
 
 set -e
 
+BASE=/net/corpora/nlnieuws
+PART=$BASE/NU
+
 unset CDPATH
-PATH=/net/corpora/nlnieuws/bin:/net/aps/bin:$PATH
+PATH=$PART:$BASE/bin:$BASE:/net/aps/bin:$PATH
 export TZ=Europe/Amsterdam
 . /net/aps/etc/alpino-activate.sh > /dev/null
 
 if [ "$1" = "" ]
 then
-    ds=`date -d -7days +%G-%V`
+    ds=`date -d -2days +%Y-%m-%d`
 else
     case "$1" in
-        2[0-9][0-9][0-9]-[0-5][0-9])
+        2[0-9][0-9][0-9]-[01][0-9]-[0-3][0-9])
             ds=$1
             ;;
         *)
@@ -23,10 +26,11 @@ else
 fi
 
 dp=${ds//-//}
+year=${ds%%-*}
+corpus=$PART/corpus/$year/$ds
+mkdir -p $PART/corpus/$year
 
-corpus=/net/corpora/nlnieuws/NU/corpus/$ds
-
-cd /net/corpora/nlnieuws/NU/$dp
+cd $PART/$dp
 
 ln -s lock.$$ lock
 if [ "`readlink lock`" != lock.$$ ]
@@ -51,14 +55,14 @@ cd out
 mkdir xml
 Alpino -flag treebank xml debug=1 end_hook=xml user_max=900000 -parse < $corpus.lines 2> $corpus.log
 
-../../../metadata
+metadata
 
 cd xml
 rm -f $corpus.data.dz $corpus.index
 alto -q -o $corpus.data.dz *.xml
 
 # telling per bericht, niet per zin
-/net/corpora/nlnieuws/namen.sh -x T -s $corpus.data.dz > $corpus.tag.txt
+query.sh -x T -s $corpus.data.dz > $corpus.tag.txt
 
 cd ../..
 rm -fr out