diff --git a/README.md b/README.md
index 2ab0dd8..2d80ff4 100644
--- a/README.md
+++ b/README.md
@@ -21,9 +21,9 @@ in the results match the search terms).
 
 Wikipedia publishes [dumps](https://meta.wikimedia.org/wiki/Data_dumps) of their databases once per month.
 
-To run one build you need 150GB of disc space (of which 90GB is Postgresql database). The scripts process
-39 languages and output one file. Runtime is approximately 9 hours on a 4 core, 4GB RAM machine with SSD
-discs.
+To run one build you need 150GB of disc space (of which 90GB Postgresql database). The scripts process
+39 languages and output 4 files. Runtime is approximately 9 hours on a 4 core, 4GB RAM machine with NVMe
+drives.
 
 ```
 334M wikimedia_importance.tsv.gz
diff --git a/bin/filter_page.py b/bin/filter_page.py
index ff6a58e..435cc80 100755
--- a/bin/filter_page.py
+++ b/bin/filter_page.py
@@ -6,7 +6,6 @@
     #   `page_id`            int(8) unsigned     NOT NULL AUTO_INCREMENT,
     #   `page_namespace`     int(11)             NOT NULL DEFAULT 0,
     #   `page_title`         varbinary(255)      NOT NULL DEFAULT '',
-    #   `page_restrictions`  tinyblob                     DEFAULT NULL,
     #   `page_is_redirect`   tinyint(1) unsigned NOT NULL DEFAULT 0,
     #   `page_is_new`        tinyint(1) unsigned NOT NULL DEFAULT 0,
     #   `page_random`        double unsigned     NOT NULL DEFAULT 0,
diff --git a/bin/filter_pagelinks.py b/bin/filter_pagelinks.py
index 21123f8..3188d4f 100755
--- a/bin/filter_pagelinks.py
+++ b/bin/filter_pagelinks.py
@@ -3,16 +3,28 @@
 '''
 Input from STDIN
     # CREATE TABLE `pagelinks` (
-    #   `pl_from`            int(8) unsigned    NOT NULL DEFAULT 0,
-    #   `pl_namespace`       int(11)            NOT NULL DEFAULT 0,
-    #   `pl_title`           varbinary(255)     NOT NULL DEFAULT '',
-    #   `pl_from_namespace`  int(11)            NOT NULL DEFAULT 0,
+    #   `pl_from`            int(8) unsigned     NOT NULL DEFAULT 0,
+    #   `pl_namespace`       int(11)             NOT NULL DEFAULT 0,
+    #   `pl_target_id`       bigint(20) unsigned NOT NULL,
 
 Output to STDOUT: pl_title, count
 '''
 
 import sys
 import csv
+import gzip
+
+if len(sys.argv) < 2:
+    print("Usage: filter_pagelinks.py linktarget.csv.gz")
+    exit(1)
+
+linktarget_filename = sys.argv[1]
+linktarget_id_to_title = dict()
+
+with gzip.open(linktarget_filename, 'rt') as gzfile:
+    reader = csv.reader(gzfile)
+    for row in reader:
+        linktarget_id_to_title[row[0]] = row[1]
 
 reader = csv.reader(sys.stdin)
 writer = csv.writer(sys.stdout, dialect='unix', quoting=csv.QUOTE_MINIMAL)
@@ -23,8 +35,8 @@
     if (row[1] != '0'):
         continue
 
-    title = row[2].replace('\r', '')
-    if len(title) == 0:
+    title = linktarget_id_to_title.get(row[2])
+    if title is None:
         continue
 
     if title not in counts:
diff --git a/bin/filter_redirect.py b/bin/filter_redirect.py
index 83ae381..bbcdb48 100755
--- a/bin/filter_redirect.py
+++ b/bin/filter_redirect.py
@@ -10,6 +10,12 @@
     #   `rd_fragment`     varbinary(255)             DEFAULT NULL,
 
 Output to STDOUT: rd_from_page_id, rd_title
+
+Same for linktarget table
+    # CREATE TABLE `linktarget` (
+    #   `lt_id`          bigint(20) unsigned  NOT NULL AUTO_INCREMENT,
+    #   `lt_namespace`   int(11)              NOT NULL,
+    #   `lt_title`       varbinary(255)       NOT NULL,
 '''
 
 import sys
diff --git a/steps/latest_available_data.sh b/steps/latest_available_data.sh
index 698e3e0..75ad1e7 100755
--- a/steps/latest_available_data.sh
+++ b/steps/latest_available_data.sh
@@ -61,7 +61,7 @@ check_all_files_ready() {
     ## usually the last to be dumped
     ##
     # from wikipedia_download.sh
-    WIKIPEDIA_REQUIRED_FILES="page pagelinks langlinks redirect"
+    WIKIPEDIA_REQUIRED_FILES="page pagelinks langlinks linktarget redirect"
     DUMP_RUN_INFO_URL="https://mirror.clarkson.edu/wikimedia/zhwiki/$CHECK_DATE/dumpruninfo.json"
     debug $DUMP_RUN_INFO_URL
     DUMP_RUN_INFO=$(curl -s --fail "$DUMP_RUN_INFO_URL")
diff --git a/steps/wikipedia_download.sh b/steps/wikipedia_download.sh
index 3b5ce7e..32f7311 100755
--- a/steps/wikipedia_download.sh
+++ b/steps/wikipedia_download.sh
@@ -42,19 +42,21 @@ do
     mkdir -p "$DOWNLOADED_PATH/$LANG"
 
     # English is the largest
-    # 1.7G  downloaded/en/page.sql.gz
-    # 6.2G  downloaded/en/pagelinks.sql.gz
-    # 355M  downloaded/en/langlinks.sql.gz
-    # 128M  downloaded/en/redirect.sql.gz
+    # 2.1G  downloaded/en/page.sql.gz
+    # 6.4G  downloaded/en/pagelinks.sql.gz
+    # 492M  downloaded/en/langlinks.sql.gz
+    # 992M  downloaded/en/linktarget.sql.gz
+    # 160M  downloaded/en/redirect.sql.gz
 
     # Smaller language Turkish
-    #  53M  downloaded/tr/page.sql.gz
-    # 176M  downloaded/tr/pagelinks.sql.gz
-    # 106M  downloaded/tr/langlinks.sql.gz
-    # 3.2M  downloaded/tr/redirect.sql.gz
+    #  90M  downloaded/tr/page.sql.gz
+    # 255M  downloaded/tr/pagelinks.sql.gz
+    # 166M  downloaded/tr/langlinks.sql.gz
+    #  62M  downloaded/tr/linktarget.sql.gz
+    # 4.2M  downloaded/tr/redirect.sql.gz
 
   
-    for FN in page.sql.gz pagelinks.sql.gz langlinks.sql.gz redirect.sql.gz; do
+    for FN in page.sql.gz pagelinks.sql.gz langlinks.sql.gz linktarget.sql.gz redirect.sql.gz; do
 
         download https://$WIKIMEDIA_HOST/${LANG}wiki/$WIKIPEDIA_DATE/${LANG}wiki-$WIKIPEDIA_DATE-$FN             "$DOWNLOADED_PATH/$LANG/$FN"
         download https://$WIKIMEDIA_HOST/${LANG}wiki/$WIKIPEDIA_DATE/md5sums-${LANG}wiki-$WIKIPEDIA_DATE-$FN.txt "$DOWNLOADED_PATH/$LANG/$FN.md5"
diff --git a/steps/wikipedia_sql2csv.sh b/steps/wikipedia_sql2csv.sh
index 1cb7e04..76a6b98 100755
--- a/steps/wikipedia_sql2csv.sh
+++ b/steps/wikipedia_sql2csv.sh
@@ -17,11 +17,12 @@ do
     mkdir -p "$CONVERTED_PATH/$LANG/"
 
     echo "[language $LANG] Page table SQL => CSV"
+    # https://www.mediawiki.org/wiki/Manual:Page_table
+    #
     # CREATE TABLE `page` (
     #   `page_id`            int(8) unsigned     NOT NULL AUTO_INCREMENT,
     #   `page_namespace`     int(11)             NOT NULL DEFAULT 0,
     #   `page_title`         varbinary(255)      NOT NULL DEFAULT '',
-    #   `page_restrictions`  tinyblob                     DEFAULT NULL,
     #   `page_is_redirect`   tinyint(1) unsigned NOT NULL DEFAULT 0,
     #   `page_is_new`        tinyint(1) unsigned NOT NULL DEFAULT 0,
     #   `page_random`        double unsigned     NOT NULL DEFAULT 0,
@@ -44,26 +45,50 @@ do
     pigz -9 > $CONVERTED_PATH/$LANG/pages.csv.gz
 
 
+    echo "[language $LANG] linktarget table SQL => CSV"
+    # https://www.mediawiki.org/wiki/Manual:Linktarget_table
+    #
+    # CREATE TABLE `linktarget` (
+    #   `lt_id`          bigint(20) unsigned  NOT NULL AUTO_INCREMENT,
+    #   `lt_namespace`   int(11)              NOT NULL,
+    #   `lt_title`       varbinary(255)       NOT NULL,
+    #
+    # Only interested in lt_namespace == 0 (articles)
+    # English wikipedia:
+    #   input 964MB compressed (100m rows)
+    #   output 322MB compressed (30m rows)
+    # Output columns: lt_id, lt_title
+
+    unpigz -c $DOWNLOADED_PATH/${LANG}/linktarget.sql.gz | \
+    bin/mysqldump_to_csv.py | \
+    bin/filter_redirect.py  | \
+    pigz -9 > $CONVERTED_PATH/$LANG/linktarget.csv.gz
+
+
+
     echo "[language $LANG] Pagelinks table SQL => CSV"
+    # https://www.mediawiki.org/wiki/Manual:Pagelinks_table
+    #
     # CREATE TABLE `pagelinks` (
-    #   `pl_from`            int(8) unsigned    NOT NULL DEFAULT 0,
-    #   `pl_namespace`       int(11)            NOT NULL DEFAULT 0,
-    #   `pl_title`           varbinary(255)     NOT NULL DEFAULT '',
-    #   `pl_from_namespace`  int(11)            NOT NULL DEFAULT 0,
+    #   `pl_from`            int(8) unsigned     NOT NULL DEFAULT 0,
+    #   `pl_namespace`       int(11)             NOT NULL DEFAULT 0,
+    #   `pl_target_id`       bigint(20) unsigned NOT NULL,
     #
-    # Only interested in pl_namespace == 0 (articles)
+    # Only interested in target_ids that point to  == 0 (articles)
     # English wikipedia:
     #   input 6.8GB compressed
     #   output 200MB compressed
-    # Output columns: pl_title, count
+    # Output columns: lt_title (from linktarget file), count (unique pl_from)
 
     unpigz -c $DOWNLOADED_PATH/$LANG/pagelinks.sql.gz | \
     bin/mysqldump_to_csv.py | \
-    bin/filter_pagelinks.py | \
+    bin/filter_pagelinks.py $CONVERTED_PATH/$LANG/linktarget.csv.gz | \
     pigz -9 > $CONVERTED_PATH/$LANG/pagelinks.csv.gz
 
 
     echo "[language $LANG] langlinks table SQL => CSV"
+    # https://www.mediawiki.org/wiki/Manual:Langlinks_table
+    #
     # CREATE TABLE `langlinks` (
     #   `ll_from`         int(8) unsigned   NOT NULL DEFAULT 0,
     #   `ll_lang`         varbinary(35)     NOT NULL DEFAULT '',
@@ -81,7 +106,11 @@ do
     pigz -9 > $CONVERTED_PATH/$LANG/langlinks.csv.gz
 
 
+
+
     echo "[language $LANG] redirect table SQL => CSV"
+    # https://www.mediawiki.org/wiki/Manual:Redirect_table
+    #
     # CREATE TABLE `redirect` (
     #   `rd_from`         int(8) unsigned   NOT NULL DEFAULT 0,
     #   `rd_namespace`    int(11)           NOT NULL DEFAULT 0,