[subset] Have the subset fuzzer pull the text string to subset to from the end...

[subset] Have the subset fuzzer pull the text string to subset to from the end of the fuzzer provided data.

test/fuzzing/hb-subset-fuzzer.cc

@@ -6,41 +6,55 @@
 
 #include "hb-subset.h"
 
-extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size)
+
+void trySubset (hb_face_t *face,
+                const hb_codepoint_t text[],
+                int text_length,
+                bool drop_hints)
 {
-  hb_blob_t *blob = hb_blob_create ((const char *)data, size,
-                                    HB_MEMORY_MODE_READONLY, NULL, NULL);
-  hb_face_t *face = hb_face_create (blob, 0);
   hb_subset_profile_t *profile = hb_subset_profile_create ();
 
   hb_subset_input_t *input = hb_subset_input_create_or_fail ();
-  *hb_subset_input_drop_hints(input) = false;
+  *hb_subset_input_drop_hints(input) = drop_hints;
   hb_set_t *codepoints = hb_subset_input_unicode_set (input);
 
-  hb_subset_input_t *input_drop_hints = hb_subset_input_create_or_fail ();
-  *hb_subset_input_drop_hints(input_drop_hints) = true;
-  hb_set_t *codepoints_drop_hints = hb_subset_input_unicode_set (input);
-
-  const hb_codepoint_t text[] =
-      {
-        'A', 'B', 'C', 'D', 'E', 'X', 'Y', 'Z', '1', '2',
-        '3', '@', '_', '%', '&', ')', '*', '$', '!'
-      };
-  for (int i = 0; i < sizeof (text) / sizeof (hb_codepoint_t); i++)
+  for (int i = 0; i < text_length; i++)
   {
     hb_set_add (codepoints, text[i]);
-    hb_set_add (codepoints_drop_hints, text[i]);
   }
 
   hb_face_t *result = hb_subset (face, profile, input);
   hb_face_destroy (result);
 
-  result = hb_subset (face, profile, input_drop_hints);
-  hb_face_destroy (result);
-
   hb_subset_input_destroy (input);
-  hb_subset_input_destroy (input_drop_hints);
   hb_subset_profile_destroy (profile);
+}
+
+extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size)
+{
+  hb_blob_t *blob = hb_blob_create ((const char *)data, size,
+                                    HB_MEMORY_MODE_READONLY, NULL, NULL);
+  hb_face_t *face = hb_face_create (blob, 0);
+
+  const hb_codepoint_t text[] =
+      {
+        'A', 'B', 'C', 'D', 'E', 'X', 'Y', 'Z', '1', '2',
+        '3', '@', '_', '%', '&', ')', '*', '$', '!'
+      };
+
+  trySubset (face, text, sizeof (text) / sizeof (hb_codepoint_t), true);
+  trySubset (face, text, sizeof (text) / sizeof (hb_codepoint_t), false);
+
+  hb_codepoint_t text_from_data[16];
+  if (size > sizeof(text_from_data)) {
+    memcpy(text_from_data,
+           data + size - sizeof(text_from_data),
+           sizeof(text_from_data));
+    unsigned int text_size = sizeof (text_from_data) / sizeof (hb_codepoint_t);
+    trySubset (face, text_from_data, text_size, true);
+    trySubset (face, text_from_data, text_size, false);
+  }
+
   hb_face_destroy (face);
   hb_blob_destroy (blob);