Snap for 5061196 from d8f8dfa2a0263e2881529073c486a18f51764faa to master-cuttlefish-testing-releasemaster-cuttlefish-testing-release

Change-Id: I9b1de50e0f9a02b58450afa7a9f1c17f674aa6c8

4 files changed, 59 insertions, 17 deletions

diff --git a/include/minikin/IcuUtils.h b/include/minikin/IcuUtils.h
new file mode 100644
index 0000000..f8d214d
--- /dev/null
+++ b/include/minikin/IcuUtils.h
@@ -0,0 +1,33 @@
+/*
+ * Copyright (C) 2018 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef MINIKIN_ICUUTILS_H
+#define MINIKIN_ICUUTILS_H
+
+#include <unicode/ubrk.h>
+#include <memory>
+
+namespace minikin {
+
+struct IcuUbrkDeleter {
+    void operator()(UBreakIterator* v) { ubrk_close(v); }
+};
+
+using IcuUbrkUniquePtr = std::unique_ptr<UBreakIterator, IcuUbrkDeleter>;
+
+}  // namespace minikin
+
+#endif  // MINIKIN_ICUUTILS_H
diff --git a/libs/minikin/WordBreaker.cpp b/libs/minikin/WordBreaker.cpp
index dae5cab..af9873c 100644
--- a/libs/minikin/WordBreaker.cpp
+++ b/libs/minikin/WordBreaker.cpp
@@ -19,6 +19,7 @@
 #include <list>
 #include <map>
 
+#include <unicode/ubrk.h>
 #include <unicode/uchar.h>
 #include <unicode/utf16.h>
 
@@ -31,13 +32,21 @@
 namespace minikin {
 
 namespace {
-static icu::BreakIterator* createNewIterator(const Locale& locale) {
+static UBreakIterator* createNewIterator(const Locale& locale) {
     // TODO: handle failure status
     UErrorCode status = U_ZERO_ERROR;
-    return icu::BreakIterator::createLineInstance(
-            locale.isUnsupported() ? icu::Locale::getRoot()
-                                   : icu::Locale::createFromName(locale.getString().c_str()),
-            status);
+    const char* localeID;
+
+    if (locale.isUnsupported()) {
+        localeID = uloc_getDefault();
+    } else {
+        localeID = locale.getString().c_str();
+    }
+
+    char buf[ULOC_FULLNAME_CAPACITY] = {};
+    uloc_getName(localeID, buf, ULOC_FULLNAME_CAPACITY, &status);
+
+    return ubrk_open(UBreakIteratorType::UBRK_LINE, buf, nullptr, 0, &status);
 }
 }  // namespace
 
@@ -53,7 +62,7 @@ ICULineBreakerPool::Slot ICULineBreakerPoolImpl::acquire(const Locale& locale) {
     }
 
     // Not found in pool. Create new one.
-    return {id, std::unique_ptr<icu::BreakIterator>(createNewIterator(locale))};
+    return {id, IcuUbrkUniquePtr(createNewIterator(locale))};
 }
 
 void ICULineBreakerPoolImpl::release(ICULineBreakerPool::Slot&& slot) {
@@ -79,7 +88,7 @@ ssize_t WordBreaker::followingWithLocale(const Locale& locale, size_t from) {
     UErrorCode status = U_ZERO_ERROR;
     MINIKIN_ASSERT(mText != nullptr, "setText must be called first");
     // TODO: handle failure status
-    mIcuBreaker.breaker->setText(&mUText, status);
+    ubrk_setUText(mIcuBreaker.breaker.get(), &mUText, &status);
     if (mInEmailOrUrl) {
         // Note:
         // Don't reset mCurrent, mLast, or mScanOffset for keeping email/URL context.
@@ -116,7 +125,7 @@ ssize_t WordBreaker::current() const {
  **/
 static bool isValidBreak(const uint16_t* buf, size_t bufEnd, int32_t i) {
     const size_t position = static_cast<size_t>(i);
-    if (i == icu::BreakIterator::DONE || position == bufEnd) {
+    if (i == UBRK_DONE || position == bufEnd) {
         // If the iterator reaches the end, treat as break.
         return true;
     }
@@ -161,9 +170,9 @@ static bool isValidBreak(const uint16_t* buf, size_t bufEnd, int32_t i) {
 // Customized iteratorNext that takes care of both resets and our modifications
 // to ICU's behavior.
 int32_t WordBreaker::iteratorNext() {
-    int32_t result = mIcuBreaker.breaker->following(mCurrent);
+    int32_t result = ubrk_following(mIcuBreaker.breaker.get(), mCurrent);
     while (!isValidBreak(mText, mTextSize, result)) {
-        result = mIcuBreaker.breaker->next();
+        result = ubrk_next(mIcuBreaker.breaker.get());
     }
     return result;
 }
@@ -211,11 +220,11 @@ void WordBreaker::detectEmailOrUrl() {
             }
         }
         if (state == SAW_AT || state == SAW_COLON_SLASH_SLASH) {
-            if (!mIcuBreaker.breaker->isBoundary(i)) {
+            if (!ubrk_isBoundary(mIcuBreaker.breaker.get(), i)) {
                 // If there are combining marks or such at the end of the URL or the email address,
                 // consider them a part of the URL or the email, and skip to the next actual
                 // boundary.
-                i = mIcuBreaker.breaker->following(i);
+                i = ubrk_following(mIcuBreaker.breaker.get(), i);
             }
             mInEmailOrUrl = true;
         } else {
diff --git a/libs/minikin/WordBreaker.h b/libs/minikin/WordBreaker.h
index 487c1fa..99023a2 100644
--- a/libs/minikin/WordBreaker.h
+++ b/libs/minikin/WordBreaker.h
@@ -26,8 +26,9 @@
 #include <list>
 #include <mutex>
 
-#include <unicode/brkiter.h>
+#include <unicode/ubrk.h>
 
+#include "minikin/IcuUtils.h"
 #include "minikin/Macros.h"
 #include "minikin/Range.h"
 
@@ -41,7 +42,7 @@ class ICULineBreakerPool {
 public:
     struct Slot {
         Slot() : localeId(0), breaker(nullptr) {}
-        Slot(uint64_t localeId, std::unique_ptr<icu::BreakIterator>&& breaker)
+        Slot(uint64_t localeId, IcuUbrkUniquePtr&& breaker)
                 : localeId(localeId), breaker(std::move(breaker)) {}
 
         Slot(Slot&& other) = default;
@@ -52,7 +53,7 @@ public:
         Slot& operator=(const Slot&) = delete;
 
         uint64_t localeId;
-        std::unique_ptr<icu::BreakIterator> breaker;
+        IcuUbrkUniquePtr breaker;
     };
     virtual ~ICULineBreakerPool() {}
     virtual Slot acquire(const Locale& locale) = 0;
diff --git a/tests/unittest/WordBreakerTests.cpp b/tests/unittest/WordBreakerTests.cpp
index 394c64e..ee01d09 100644
--- a/tests/unittest/WordBreakerTests.cpp
+++ b/tests/unittest/WordBreakerTests.cpp
@@ -96,7 +96,6 @@ TEST(WordBreakerTest, postfixAndPrefix) {
 TEST(WordBreakerTest, myanmarKinzi) {
     uint16_t buf[] = {0x1004, 0x103A, 0x1039, 0x1000, 0x102C};  // NGA, ASAT, VIRAMA, KA, UU
     WordBreaker breaker;
-    icu::Locale burmese("my");
     breaker.setText(buf, NELEM(buf));
     EXPECT_EQ(0, breaker.current());
 
@@ -596,7 +595,7 @@ TEST(WordBreakerTest, LineBreakerPool_acquire_with_release) {
     ICULineBreakerPool::Slot enUSBreaker = pool.acquire(enUS);
 
     uint64_t enUSBreakerLocaleId = enUSBreaker.localeId;
-    icu::BreakIterator* enUSBreakerPtr = enUSBreaker.breaker.get();
+    UBreakIterator* enUSBreakerPtr = enUSBreaker.breaker.get();
 
     pool.release(std::move(enUSBreaker));
     EXPECT_EQ(nullptr, enUSBreaker.breaker.get());