4 files changed, 65 insertions, 25 deletions
diff --git a/.gitignore b/.gitignore
index c5c3e26..d9c799e 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,3 +1,5 @@
 *.o
 sorter
 *.tar.gz
+*.diff
+data/*
diff --git a/generate_test_data.sh b/generate_test_data.sh
new file mode 100755
index 0000000..1cc6c9f
--- /dev/null
+++ b/generate_test_data.sh
@@ -0,0 +1,22 @@
+#!/bin/sh
+# Note generating test data
+# may take a while
+
+mkdir -p data
+sizes=(10000 50000 100000 500000 1000000 5000000)
+
+for size in "${sizes[@]}"; do
+    for run in {1..3}; do
+        outfile="data/random_${size}_run${run}.tsv"     # '.tsv' stands for
+        echo "Generating $outfile ..."                  # tab seperaed values
+        {
+            for ((i=1; i<=size; i++)); do
+                # using 'printf' instead of 'echo'
+                # due to performance issues
+                printf "%d\t%d\n" $((1 + RANDOM % 32767)) $((1 + RANDOM % 10000))
+            done
+        } > "$outfile"
+    done
+done
+
+echo "All random data files gernated in ./data/"
diff --git a/test.sh b/test.sh
new file mode 100755
index 0000000..0a1f6b8
--- /dev/null
+++ b/test.sh
@@ -0,0 +1,41 @@
+#!/bin/sh
+
+TEST_DIR="test"
+DATA_DIR="$TEST_DIR/data"
+
+SRC_DIR="src"
+SORTER="$SRC_DIR/sorter"
+REF="sort -s -n -k 2"       # '-s' for using a stable algo
+
+total=0
+passed=0
+failed=0
+
+for TEST_FILE in "$DATA_DIR"/*; do
+    ((total++))
+    FILE_NAME=$(basename "$TEST_FILE")
+    DIFF_FILE="$TEST_DIR/$FILE_NAME.diff"
+
+   diff -u <($REF "$TEST_FILE") <($SORTER "$TEST_FILE") > "$DIFF_FILE"
+
+    if [ ! -s "$DIFF_FILE" ]; then
+        echo "Passed: $TEST_FILE"
+        ((passed++))
+        rm -f "$DIFF_FILE"
+    else
+        echo "Failed: $TEST_FILE"
+        ((failed++))
+    fi
+done
+
+# Summary overview
+echo "=============================="
+echo "Test Summary:"
+echo "Total tests: $total"
+echo "Passed:      $passed"
+echo "Failed:      $failed"
+echo "=============================="
+
+if [ $failed -ne 0 ]; then
+    echo "Diff files for failed tests are saved in $TEST_DIR/"
+fi
diff --git a/test/test.sh b/test/test.sh
deleted file mode 100755
index 855d6cf..0000000
--- a/test/test.sh
+++ /dev/null
@@ -1,25 +0,0 @@
-#!/bin/sh
-
-TEST_INPUT="numbers.txt"
-
-SORTER="./sorter"
-REF="sort -s -n -k 2"
-
-SORTER_OUT="sorter_out.txt"
-REF_OUT="ref_output.txt"
-DIFF_FILE="diff.txt"
-
-$SORTER "$TEST_INPUT" > "$SORTER_OUT"
-$REF "$TEST_INPUT" > "$REF_OUT"
-
-# Compare outputs
-if diff -q "$SORTER_OUT" "$REF_OUT" > /dev/null; then
-    echo "Passed: $TEST_INPUT"
-else
-    echo "Failed: $TEST_INPUT"
-    echo "Differences:"
-    diff -u "$SORTER_OUT" "$REF_OUT" > "$DIFF_FILE"
-    diff -u "$SORTER_OUT" "$REF_OUT" | head -n 20
-    echo
-fi
-