src: improve buffer.transcode performance

nodejs · Aug 1, 2024 · 5f9f633 · 5f9f633
1 parent 20aff2b
commit 5f9f633
Showing 1 changed file with 30 additions and 36 deletions.
diff --git a/src/node_i18n.cc b/src/node_i18n.cc
@@ -42,6 +42,7 @@
 
 #include "node_i18n.h"
 #include "node_external_reference.h"
+#include "simdutf.h"
 
 #if defined(NODE_HAVE_I18N_SUPPORT)
 
@@ -222,25 +223,21 @@ MaybeLocal<Object> TranscodeUcs2FromUtf8(Environment* env,
                                          const size_t source_length,
                                          UErrorCode* status) {
   *status = U_ZERO_ERROR;
-  MaybeStackBuffer<UChar> destbuf;
-  int32_t result_length;
-  u_strFromUTF8(*destbuf, destbuf.capacity(), &result_length,
-                source, source_length, status);
   MaybeLocal<Object> ret;
-  if (U_SUCCESS(*status)) {
-    destbuf.SetLength(result_length);
-    ret = ToBufferEndian(env, &destbuf);
-  } else if (*status == U_BUFFER_OVERFLOW_ERROR) {
-    *status = U_ZERO_ERROR;
-    destbuf.AllocateSufficientStorage(result_length);
-    u_strFromUTF8(*destbuf, result_length, &result_length,
-                  source, source_length, status);
-    if (U_SUCCESS(*status)) {
-      destbuf.SetLength(result_length);
-      ret = ToBufferEndian(env, &destbuf);
-    }
+  size_t expected_utf16_length =
+      simdutf::utf16_length_from_utf8(source, source_length);
+  MaybeStackBuffer<UChar> destbuf(expected_utf16_length);
+  auto actual_length =
+      simdutf::convert_utf8_to_utf16(source, source_length, destbuf.out());
+
+  if (actual_length == 0) {
+    *status = U_INVALID_CHAR_FOUND;
+    return ret;
   }
-  return ret;
+
+  CHECK_EQ(actual_length, expected_utf16_length);
+
+  return ToBufferEndian(env, &destbuf);
 }
 
 MaybeLocal<Object> TranscodeUtf8FromUcs2(Environment* env,
@@ -252,26 +249,23 @@ MaybeLocal<Object> TranscodeUtf8FromUcs2(Environment* env,
   *status = U_ZERO_ERROR;
   MaybeLocal<Object> ret;
   const size_t length_in_chars = source_length / sizeof(UChar);
-  int32_t result_length;
-  MaybeStackBuffer<UChar> sourcebuf;
-  MaybeStackBuffer<char> destbuf;
-  CopySourceBuffer(&sourcebuf, source, source_length, length_in_chars);
-  u_strToUTF8(*destbuf, destbuf.capacity(), &result_length,
-              *sourcebuf, length_in_chars, status);
-  if (U_SUCCESS(*status)) {
-    destbuf.SetLength(result_length);
-    ret = ToBufferEndian(env, &destbuf);
-  } else if (*status == U_BUFFER_OVERFLOW_ERROR) {
-    *status = U_ZERO_ERROR;
-    destbuf.AllocateSufficientStorage(result_length);
-    u_strToUTF8(*destbuf, result_length, &result_length, *sourcebuf,
-                length_in_chars, status);
-    if (U_SUCCESS(*status)) {
-      destbuf.SetLength(result_length);
-      ret = ToBufferEndian(env, &destbuf);
-    }
+  size_t expected_utf8_length = simdutf::utf8_length_from_utf16(
+      reinterpret_cast<const char16_t*>(source), length_in_chars);
+
+  MaybeStackBuffer<char> destbuf(expected_utf8_length);
+  auto actual_length = simdutf::convert_utf16_to_utf8(
+      reinterpret_cast<const char16_t*>(source),
+      length_in_chars,
+      destbuf.out());
+
+  if (actual_length == 0) {
+    *status = U_INVALID_CHAR_FOUND;
+    return ret;
   }
-  return ret;
+
+  CHECK_EQ(actual_length, expected_utf8_length);
+
+  return ToBufferEndian(env, &destbuf);
 }
 
 const char* EncodingName(const enum encoding encoding) {