Add functions for marshalling wchar_t (which is fixed 32-bit UTF32LE on Unix) strings to and from JS, and add a test.

3 files changed, 56 insertions, 0 deletions

diff --git a/src/preamble.js b/src/preamble.js
index 227b3043..465e47a3 100644
--- a/src/preamble.js
+++ b/src/preamble.js
@@ -569,6 +569,33 @@ function Pointer_stringify(ptr, /* optional */ length) {
 }
 Module['Pointer_stringify'] = Pointer_stringify;
 
+// Given a pointer 'ptr' to a null-terminated UTF32LE-encoded string in the emscripten HEAP, returns
+// a copy of that string as a Javascript String object.
+function utf32_to_jsstring(ptr) {
+  var i = 0;
+
+  var str = '';
+  while (1) {
+    var utf32 = {{{ makeGetValue('ptr', 'i*4', 'i32') }}};
+    if (utf32 == 0)
+      return str;
+    ++i;
+    str += String.fromCharCode(utf32);
+  }
+}
+Module['utf32_to_jsstring'] = utf32_to_jsstring;
+
+// Copies the given Javascript String object 'str' to the emscripten HEAP at address 'outPtr', 
+// null-terminated and encoded in UTF32LE form. The copy will require (str.length+1)*4 bytes of space in the HEAP.
+function jsstring_to_utf32(str, outPtr) {
+  for(var i = 0; i < str.length; ++i) {
+    var utf32 = str.charCodeAt(i);
+    {{{ makeSetValue('outPtr', 'i*4', 'utf32', 'i32') }}}
+  }
+  {{{ makeSetValue('outPtr', 'str.length*4', 0, 'i32') }}}
+}
+Module['jsstring_to_utf32'] = jsstring_to_utf32;
+
 // Memory management
 
 var PAGE_SIZE = 4096;
diff --git a/tests/test_core.py b/tests/test_core.py
index 5d4f35e8..7ceeebd2 100644
--- a/tests/test_core.py
+++ b/tests/test_core.py
@@ -7525,6 +7525,9 @@ def process(filename):
     '''
     self.do_run(src, '206 188 226 128 μ†ℱ ╋ℯ╳╋ 😇\nμ†ℱ ╋ℯ╳╋ 😇,206,188,226,128\n');
 
+  def test_utf32(self):
+    self.do_run(open(path_from_root('tests', 'utf32.cpp')).read(), 'OK.')
+
   def test_direct_string_constant_usage(self):
     if self.emcc_args is None: return self.skip('requires libcxx')
 
diff --git a/tests/utf32.cpp b/tests/utf32.cpp
new file mode 100644
index 00000000..a3e660ee
--- /dev/null
+++ b/tests/utf32.cpp
@@ -0,0 +1,26 @@
+#include <stdio.h>
+#include <string>
+#include <emscripten.h>
+#include <cassert>
+#include <wchar.h>
+
+// This code tests that utf32-encoded std::wstrings can be marshalled between C++ and JS.
+int main() {
+	std::wstring wstr = L"abc\u2603\u20AC123"; // U+2603 is snowman, U+20AC is the Euro sign.
+	const int len = (wstr.length()+1)*4;
+	char *memory = new char[len];
+
+	asm("var str = Module.utf32_to_jsstring(%0);"
+		"Module.print(str);"
+		"Module.jsstring_to_utf32(str, %1);"
+	:
+	: "r"(wstr.c_str()), "r"(memory));
+
+	// Compare memory to confirm that the string is intact after taking a route through JS side.
+	const char *srcPtr = reinterpret_cast<const char *>(wstr.c_str());
+	for(int i = 0; i < len; ++i) {
+		assert(memory[i] == srcPtr[i]);
+	}
+	printf("OK.\n");
+	delete[] memory;
+}