geht jetzt

1 files changed, 30 insertions, 0 deletions

diff --git a/generate-random-word-from-wikipedia.sh b/generate-random-word-from-wikipedia.sh
new file mode 100755
index 0000000..fbc720f
--- /dev/null
+++ b/generate-random-word-from-wikipedia.sh
@@ -0,0 +1,30 @@
+#!/bin/bash
+
+base_dir=$(readlink -f "$(dirname "$0")")
+
+depth=$1
+shift
+
+if [ ! -s "${base_dir}/.words.${depth}" ]; then
+
+  urls=()
+
+  for i in $(seq "${depth}"); do
+    urls=(
+      $(
+        "${base_dir}/dive-into-wikipedia.sh" "${urls[@]}" \
+          | sort -u
+      )
+    )
+    echo $i >&2
+  done
+
+  printf '%s\n' "${urls[@]}" \
+    | parallel -j0 -n1 "${base_dir}/remove-tags.sh" \
+    | sort -u \
+    | grep -vxF '' \
+    > "${base_dir}/.words.${depth}"
+
+fi
+
+"${base_dir}/markov" "${base_dir}/.words.${depth}" "$@"