From 4a498fb9c35fa61a0c07f873fb891e11b92804a1 Mon Sep 17 00:00:00 2001
From: Sen Huang <senhuang96@fb.com>
Date: Thu, 9 Sep 2021 08:55:43 -0700
Subject: [PATCH] Add a dictionary training large corpus test

---
 tests/playTests.sh | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/tests/playTests.sh b/tests/playTests.sh
index 774655aa9..45d1d3fe8 100755
--- a/tests/playTests.sh
+++ b/tests/playTests.sh
@@ -936,8 +936,13 @@ cat tmp | zstd -14 -f --size-hint=5500  | zstd -t  # considerably too low
 
 
 println "\n===>  dictionary tests "
-
-println "- test with raw dict (content only) "
+println "- Test high/low compressibility corpus training"
+datagen -g12M -P90 > tmpCorpusHighCompress
+datagen -g12M -P5 > tmpCorpusLowCompress
+zstd --train -B2K tmpCorpusHighCompress -o tmpDictHighCompress
+zstd --train -B2K tmpCorpusLowCompress -o tmpDictLowCompress
+rm -f tmpCorpusHighCompress tmpCorpusLowCompress tmpDictHighCompress tmpDictLowCompress
+println "- Test with raw dict (content only) "
 datagen > tmpDict
 datagen -g1M | $MD5SUM > tmp1
 datagen -g1M | zstd -D tmpDict | zstd -D tmpDict -dvq | $MD5SUM > tmp2
-- 
2.47.2