Use UTF-8 strings to avoid duplicate caching, part 1

StringBlock instances containing UTF-8 strings use a cache to convert
into UTF-16, but using that cache and then using a JNI call to NewString
causes the UTF-8 string as well as two copies of the UTF-16 string to
be held in memory. Getting the UTF-8 string directly from the StringPool
eliminates one copy of the UTF-16 string being held in memory.

This is part 1. Part 2 will include ResXMLParser optimizations.

Change-Id: Ibd4509a485db746d59cd4b9501f544877139276c

1 files changed, 6 insertions, 2 deletions

diff --git a/tools/aapt/StringPool.cpp b/tools/aapt/StringPool.cpp
index 51afc0a..a09cec0 100644
--- a/tools/aapt/StringPool.cpp
+++ b/tools/aapt/StringPool.cpp
@@ -25,8 +25,12 @@ void printStringPool(const ResStringPool* pool)
     const size_t NS = pool->size();
     for (size_t s=0; s<NS; s++) {
         size_t len;
-        printf("String #%ld: %s\n", s,
-                String8(pool->stringAt(s, &len)).string());
+        const char *str = (const char*)pool->string8At(s, &len);
+        if (str == NULL) {
+            str = String8(pool->stringAt(s, &len)).string();
+        }
+
+        printf("String #%ld: %s\n", s, str);
     }
 }