Support ICU (#603)

Co-authored-by: Kunihiro Ishiguro <kunihiro@zebra.dev>

.gitmodules

@@ -22,3 +22,6 @@
 [submodule "third_party/zlib"]
 	path = third_party/zlib
 	url = https://chromium.googlesource.com/chromium/src/third_party/zlib.git
+[submodule "third_party/icu"]
+	path = third_party/icu
+	url = https://chromium.googlesource.com/chromium/deps/icu.git

.gn

@@ -28,7 +28,6 @@ default_args = {
 
   # https://cs.chromium.org/chromium/src/docs/ccache_mac.md
   clang_use_chrome_plugins = false
-  v8_enable_i18n_support = false
   v8_monolithic = false
   v8_use_external_startup_data = false
   v8_use_snapshot = true

src/icu.rs

@@ -0,0 +1,46 @@
+extern "C" {
+  fn udata_setCommonData_68(this: *const u8, error_code: *mut i32);
+}
+
+/// This function bypasses the normal ICU data loading process and allows you to force ICU's system
+/// data to come out of a user-specified area in memory.
+///
+/// ICU data must be at least 8-aligned, and should be 16-aligned. See
+/// https://unicode-org.github.io/icu/userguide/icudata
+///
+/// The format of this data is that of the icu common data file, as is generated by the pkgdata
+/// tool with mode=common or mode=dll. You can read in a whole common mode file and pass the
+/// address to the start of the data, or (with the appropriate link options) pass in the pointer to
+/// the data that has been loaded from a dll by the operating system, as shown in this code:
+///
+///       extern const char U_IMPORT U_ICUDATA_ENTRY_POINT [];
+///        // U_ICUDATA_ENTRY_POINT is same as entry point specified to pkgdata tool
+///       UErrorCode  status = U_ZERO_ERROR;
+///
+///       udata_setCommonData(&U_ICUDATA_ENTRY_POINT, &status);
+///
+/// It is important that the declaration be as above. The entry point must not be declared as an
+/// extern void*.
+///
+/// Starting with ICU 4.4, it is possible to set several data packages, one per call to this
+/// function. udata_open() will look for data in the multiple data packages in the order in which
+/// they were set. The position of the linked-in or default-name ICU .data package in the search
+/// list depends on when the first data item is loaded that is not contained in the already
+/// explicitly set packages. If data was loaded implicitly before the first call to this function
+/// (for example, via opening a converter, constructing a UnicodeString from default-codepage data,
+/// using formatting or collation APIs, etc.), then the default data will be first in the list.
+///
+/// This function has no effect on application (non ICU) data. See udata_setAppData() for similar
+/// functionality for application data.
+// TODO(ry) Map error code to something useful.
+pub fn set_common_data(data: &'static [u8]) -> Result<(), i32> {
+  let mut error_code = 0i32;
+  unsafe {
+    udata_setCommonData_68(data.as_ptr(), &mut error_code);
+  }
+  if error_code == 0 {
+    Ok(())
+  } else {
+    Err(error_code)
+  }
+}

src/lib.rs

@@ -45,6 +45,7 @@ mod external_references;
 mod fixed_array;
 mod function;
 mod handle;
+pub mod icu;
 mod isolate;
 mod isolate_create_params;
 mod module;

tests/test_api.rs

@@ -4715,3 +4715,59 @@ fn prepare_stack_trace_callback() {
     v8::Integer::new(scope, 42).into()
   }
 }
+
+const ICU_DATA: &[u8; 10413584] =
+  include_bytes!("../third_party/icu/common/icudtl.dat");
+
+#[test]
+fn icu_date() {
+  assert!(v8::icu::set_common_data(ICU_DATA).is_ok());
+
+  let _setup_guard = setup();
+  let isolate = &mut v8::Isolate::new(Default::default());
+  {
+    let scope = &mut v8::HandleScope::new(isolate);
+    let context = v8::Context::new(scope);
+    let scope = &mut v8::ContextScope::new(scope, context);
+    let source = r#"
+      (new Date(Date.UTC(2020, 5, 26, 7, 0, 0))).toLocaleString("de-DE", {
+        weekday: "long",
+        year: "numeric",
+        month: "long",
+        day: "numeric",
+      });
+    "#;
+    let value = eval(scope, source).unwrap();
+    let date_de_val = v8::String::new(scope, "Freitag, 26. Juni 2020").unwrap();
+    assert!(value.is_string());
+    assert!(value.strict_equals(date_de_val.into()));
+  }
+}
+
+#[test]
+fn icu_set_common_data_fail() {
+  const BAD_DATA: &[u8; 3] = &[1, 2, 3];
+  assert!(v8::icu::set_common_data(BAD_DATA).is_err());
+}
+
+#[test]
+fn icu_format() {
+  assert!(v8::icu::set_common_data(ICU_DATA).is_ok());
+
+  let _setup_guard = setup();
+  let isolate = &mut v8::Isolate::new(Default::default());
+  {
+    let scope = &mut v8::HandleScope::new(isolate);
+    let context = v8::Context::new(scope);
+    let scope = &mut v8::ContextScope::new(scope, context);
+    let source = r#"
+      new Intl.NumberFormat("ja-JP", { style: "currency", currency: "JPY" }).format(
+        1230000,
+      );
+    "#;
+    let value = eval(scope, source).unwrap();
+    let currency_jpy_val = v8::String::new(scope, "￥1,230,000").unwrap();
+    assert!(value.is_string());
+    assert!(value.strict_equals(currency_jpy_val.into()));
+  }
+}

third_party/icu

@@ -0,0 +1 @@
+Subproject commit f4147b227f174e03f273053bb49fde33426d05bc