[Hunspell Fuzzer] Restrict fuzzer data to valid UTF8

Hunspell cannot handle invalid UTF8. Chrome makes sure that all
data is in valid UTF8 - this CL enforces the same for the fuzzer.

BUG=none
R=mmoroz@chromium.org

Review-Url: https://codereview.chromium.org/2223603002
Cr-Commit-Position: refs/heads/master@{#414878}

third_party/hunspell/BUILD.gn

@@ -119,6 +119,7 @@ fuzzer_test("hunspell_fuzzer") {
   ]
   deps = [
     ":hunspell",
+    "//base:base",
   ]
 
   # This is a dictionary for the fuzzer, not a spellcheck dictionary.

third_party/hunspell/fuzz/DEPS

+include_rules = [
+  "+base",
+]

third_party/hunspell/fuzz/hunspell_fuzzer.cc

@@ -6,6 +6,8 @@
 #include <stdint.h>
 #include <string>
 
+#include "base/strings/string16.h"
+#include "base/strings/utf_string_conversions.h"
 #include "third_party/hunspell/src/hunspell/hunspell.hxx"
 #include "third_party/hunspell/fuzz/hunspell_fuzzer_hunspell_dictionary.h"
 
@@ -18,11 +20,17 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) {
                                            sizeof(kHunspellDictionary));
 
   std::string data_string(reinterpret_cast<const char*>(data), size);
+
+  // hunspell is not handling invalid UTF8. To avoid that, do the same thing
+  // Chromium does - convert to UTF16, and back to UTF8. Valid UTF8 guaranteed.
+  base::string16 utf16_string = base::UTF8ToUTF16(data_string);
+  data_string = base::UTF16ToUTF8(utf16_string);
+
   hunspell->spell(data_string.c_str());
 
   char** suggestions = nullptr;
-  int suggetion_length = hunspell->suggest(&suggestions, data_string.c_str());
-  hunspell->free_list(&suggestions, suggetion_length);
+  int suggestion_length = hunspell->suggest(&suggestions, data_string.c_str());
+  hunspell->free_list(&suggestions, suggestion_length);
 
   return 0;
 }