crashpad/client/crash_report_database_test.cc

1029 lines
36 KiB
C++
Raw Normal View History

// Copyright 2015 The Crashpad Authors
//
// 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.
#include "client/crash_report_database.h"
#include "build/build_config.h"
#include "client/settings.h"
#include "gtest/gtest.h"
#include "test/errors.h"
#include "test/file.h"
#include "test/filesystem.h"
#include "test/scoped_temp_dir.h"
#include "util/file/file_io.h"
#include "util/file/filesystem.h"
#if BUILDFLAG(IS_IOS)
#include "util/mac/xattr.h"
#endif
namespace crashpad {
namespace test {
namespace {
class CrashReportDatabaseTest : public testing::Test {
public:
CrashReportDatabaseTest() {}
CrashReportDatabaseTest(const CrashReportDatabaseTest&) = delete;
CrashReportDatabaseTest& operator=(const CrashReportDatabaseTest&) = delete;
protected:
// testing::Test:
void SetUp() override {
db_ = CrashReportDatabase::Initialize(path());
ASSERT_TRUE(db_);
}
void ResetDatabase() { db_.reset(); }
CrashReportDatabase* db() { return db_.get(); }
base::FilePath path() const {
return temp_dir_.path().Append(FILE_PATH_LITERAL("crashpad_test_database"));
}
void CreateCrashReport(CrashReportDatabase::Report* report) {
std::unique_ptr<CrashReportDatabase::NewReport> new_report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
ASSERT_EQ(db_->PrepareNewCrashReport(&new_report),
CrashReportDatabase::kNoError);
static constexpr char kTest[] = "test";
ASSERT_TRUE(new_report->Writer()->Write(kTest, sizeof(kTest)));
char contents[sizeof(kTest)];
FileReaderInterface* reader = new_report->Reader();
ASSERT_TRUE(reader->ReadExactly(contents, sizeof(contents)));
EXPECT_EQ(memcmp(contents, kTest, sizeof(contents)), 0);
EXPECT_EQ(reader->ReadExactly(contents, 1), 0);
UUID uuid;
EXPECT_EQ(db_->FinishedWritingCrashReport(std::move(new_report), &uuid),
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
CrashReportDatabase::kNoError);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db_->LookUpCrashReport(uuid, report),
CrashReportDatabase::kNoError);
ExpectPreparedCrashReport(*report);
}
void UploadReport(const UUID& uuid, bool successful, const std::string& id) {
Settings* const settings = db_->GetSettings();
ASSERT_TRUE(settings);
time_t times[2];
ASSERT_TRUE(settings->GetLastUploadAttemptTime(&times[0]));
std::unique_ptr<const CrashReportDatabase::UploadReport> report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
ASSERT_EQ(db_->GetReportForUploading(uuid, &report),
CrashReportDatabase::kNoError);
EXPECT_NE(report->uuid, UUID());
EXPECT_FALSE(report->file_path.empty());
EXPECT_TRUE(FileExists(report->file_path)) << report->file_path.value();
EXPECT_GT(report->creation_time, 0);
if (successful) {
EXPECT_EQ(db_->RecordUploadComplete(std::move(report), id),
CrashReportDatabase::kNoError);
} else {
report.reset();
}
ASSERT_TRUE(settings->GetLastUploadAttemptTime(&times[1]));
EXPECT_NE(times[1], 0);
EXPECT_GE(times[1], times[0]);
}
void ExpectPreparedCrashReport(const CrashReportDatabase::Report& report) {
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_NE(report.uuid, UUID());
EXPECT_FALSE(report.file_path.empty());
EXPECT_TRUE(FileExists(report.file_path)) << report.file_path.value();
EXPECT_TRUE(report.id.empty());
EXPECT_GT(report.creation_time, 0);
EXPECT_FALSE(report.uploaded);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(report.last_upload_attempt_time, 0);
EXPECT_EQ(report.upload_attempts, 0);
EXPECT_FALSE(report.upload_explicitly_requested);
EXPECT_GE(report.total_size, 0u);
}
void RelocateDatabase() {
ResetDatabase();
temp_dir_.Rename();
SetUp();
}
CrashReportDatabase::OperationStatus RequestUpload(const UUID& uuid) {
CrashReportDatabase::OperationStatus os = db()->RequestUpload(uuid);
CrashReportDatabase::Report report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db_->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kNoError);
return os;
}
private:
ScopedTempDir temp_dir_;
std::unique_ptr<CrashReportDatabase> db_;
};
TEST_F(CrashReportDatabaseTest, Initialize) {
// Initialize the database for the first time, creating it.
ASSERT_TRUE(db());
Settings* settings = db()->GetSettings();
UUID client_ids[3];
ASSERT_TRUE(settings->GetClientID(&client_ids[0]));
EXPECT_NE(client_ids[0], UUID());
time_t last_upload_attempt_time;
ASSERT_TRUE(settings->GetLastUploadAttemptTime(&last_upload_attempt_time));
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(last_upload_attempt_time, 0);
// Close and reopen the database at the same path.
ResetDatabase();
EXPECT_FALSE(db());
auto db = CrashReportDatabase::InitializeWithoutCreating(path());
ASSERT_TRUE(db);
settings = db->GetSettings();
ASSERT_TRUE(settings->GetClientID(&client_ids[1]));
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(client_ids[1], client_ids[0]);
ASSERT_TRUE(settings->GetLastUploadAttemptTime(&last_upload_attempt_time));
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(last_upload_attempt_time, 0);
// Check that the database can also be opened by the method that is permitted
// to create it.
db = CrashReportDatabase::Initialize(path());
ASSERT_TRUE(db);
settings = db->GetSettings();
ASSERT_TRUE(settings->GetClientID(&client_ids[2]));
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(client_ids[2], client_ids[0]);
ASSERT_TRUE(settings->GetLastUploadAttemptTime(&last_upload_attempt_time));
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(last_upload_attempt_time, 0);
std::vector<CrashReportDatabase::Report> reports;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db->GetPendingReports(&reports), CrashReportDatabase::kNoError);
EXPECT_TRUE(reports.empty());
reports.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db->GetCompletedReports(&reports), CrashReportDatabase::kNoError);
EXPECT_TRUE(reports.empty());
// InitializeWithoutCreating() shouldnt create a nonexistent database.
base::FilePath non_database_path =
path().DirName().Append(FILE_PATH_LITERAL("not_a_database"));
db = CrashReportDatabase::InitializeWithoutCreating(non_database_path);
EXPECT_FALSE(db);
}
TEST_F(CrashReportDatabaseTest, Settings) {
// Initialize three databases and ensure settings.dat isn't created yet.
ASSERT_TRUE(db());
base::FilePath settings_path =
path().Append(FILE_PATH_LITERAL("settings.dat"));
EXPECT_FALSE(FileExists(settings_path));
std::unique_ptr<CrashReportDatabase> db2 =
CrashReportDatabase::Initialize(path());
ASSERT_TRUE(db2);
EXPECT_FALSE(FileExists(settings_path));
std::unique_ptr<CrashReportDatabase> db3 =
CrashReportDatabase::Initialize(path());
ASSERT_TRUE(db3);
EXPECT_FALSE(FileExists(settings_path));
// Ensure settings.dat exists after getter.
Settings* settings = db3->GetSettings();
ASSERT_TRUE(settings);
EXPECT_TRUE(FileExists(settings_path));
time_t last_upload_attempt_time = 42;
ASSERT_TRUE(settings->SetLastUploadAttemptTime(last_upload_attempt_time));
// Ensure the first two databases read the same value.
ASSERT_TRUE(
db2->GetSettings()->GetLastUploadAttemptTime(&last_upload_attempt_time));
EXPECT_EQ(last_upload_attempt_time, 42);
ASSERT_TRUE(
db()->GetSettings()->GetLastUploadAttemptTime(&last_upload_attempt_time));
EXPECT_EQ(last_upload_attempt_time, 42);
}
TEST_F(CrashReportDatabaseTest, NewCrashReport) {
std::unique_ptr<CrashReportDatabase::NewReport> new_report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->PrepareNewCrashReport(&new_report),
CrashReportDatabase::kNoError);
UUID expect_uuid = new_report->ReportID();
UUID uuid;
EXPECT_EQ(db()->FinishedWritingCrashReport(std::move(new_report), &uuid),
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
CrashReportDatabase::kNoError);
EXPECT_EQ(uuid, expect_uuid);
CrashReportDatabase::Report report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kNoError);
ExpectPreparedCrashReport(report);
std::vector<CrashReportDatabase::Report> reports;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetPendingReports(&reports), CrashReportDatabase::kNoError);
ASSERT_EQ(reports.size(), 1u);
EXPECT_EQ(reports[0].uuid, report.uuid);
reports.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetCompletedReports(&reports), CrashReportDatabase::kNoError);
EXPECT_TRUE(reports.empty());
}
TEST_F(CrashReportDatabaseTest, LookUpCrashReport) {
UUID uuid;
{
CrashReportDatabase::Report report;
CreateCrashReport(&report);
uuid = report.uuid;
}
{
CrashReportDatabase::Report report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kNoError);
EXPECT_EQ(report.uuid, uuid);
EXPECT_NE(report.file_path.value().find(path().value()), std::string::npos);
EXPECT_EQ(report.id, std::string());
EXPECT_FALSE(report.uploaded);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(report.last_upload_attempt_time, 0);
EXPECT_EQ(report.upload_attempts, 0);
EXPECT_FALSE(report.upload_explicitly_requested);
}
UploadReport(uuid, true, "test");
{
CrashReportDatabase::Report report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kNoError);
EXPECT_EQ(report.uuid, uuid);
EXPECT_NE(report.file_path.value().find(path().value()), std::string::npos);
EXPECT_EQ(report.id, "test");
EXPECT_TRUE(report.uploaded);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_NE(report.last_upload_attempt_time, 0);
EXPECT_EQ(report.upload_attempts, 1);
EXPECT_FALSE(report.upload_explicitly_requested);
}
}
TEST_F(CrashReportDatabaseTest, RecordUploadAttempt) {
std::vector<CrashReportDatabase::Report> reports(3);
CreateCrashReport(&reports[0]);
CreateCrashReport(&reports[1]);
CreateCrashReport(&reports[2]);
// Record two attempts: one successful, one not.
UploadReport(reports[1].uuid, false, std::string());
UploadReport(reports[2].uuid, true, "abc123");
std::vector<CrashReportDatabase::Report> query(3);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(reports[0].uuid, &query[0]),
CrashReportDatabase::kNoError);
EXPECT_EQ(db()->LookUpCrashReport(reports[1].uuid, &query[1]),
CrashReportDatabase::kNoError);
EXPECT_EQ(db()->LookUpCrashReport(reports[2].uuid, &query[2]),
CrashReportDatabase::kNoError);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(query[0].id, std::string());
EXPECT_EQ(query[1].id, std::string());
EXPECT_EQ(query[2].id, "abc123");
EXPECT_FALSE(query[0].uploaded);
EXPECT_FALSE(query[1].uploaded);
EXPECT_TRUE(query[2].uploaded);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(query[0].last_upload_attempt_time, 0);
EXPECT_NE(query[1].last_upload_attempt_time, 0);
EXPECT_NE(query[2].last_upload_attempt_time, 0);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(query[0].upload_attempts, 0);
EXPECT_EQ(query[1].upload_attempts, 1);
EXPECT_EQ(query[2].upload_attempts, 1);
// Attempt to upload and fail again.
UploadReport(reports[1].uuid, false, std::string());
time_t report_2_upload_time = query[2].last_upload_attempt_time;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(reports[0].uuid, &query[0]),
CrashReportDatabase::kNoError);
EXPECT_EQ(db()->LookUpCrashReport(reports[1].uuid, &query[1]),
CrashReportDatabase::kNoError);
EXPECT_EQ(db()->LookUpCrashReport(reports[2].uuid, &query[2]),
CrashReportDatabase::kNoError);
EXPECT_FALSE(query[0].uploaded);
EXPECT_FALSE(query[1].uploaded);
EXPECT_TRUE(query[2].uploaded);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(query[0].last_upload_attempt_time, 0);
EXPECT_GE(query[1].last_upload_attempt_time, report_2_upload_time);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(query[2].last_upload_attempt_time, report_2_upload_time);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(query[0].upload_attempts, 0);
EXPECT_EQ(query[1].upload_attempts, 2);
EXPECT_EQ(query[2].upload_attempts, 1);
// Third time's the charm: upload and succeed.
UploadReport(reports[1].uuid, true, "666hahaha");
time_t report_1_upload_time = query[1].last_upload_attempt_time;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(reports[0].uuid, &query[0]),
CrashReportDatabase::kNoError);
EXPECT_EQ(db()->LookUpCrashReport(reports[1].uuid, &query[1]),
CrashReportDatabase::kNoError);
EXPECT_EQ(db()->LookUpCrashReport(reports[2].uuid, &query[2]),
CrashReportDatabase::kNoError);
EXPECT_FALSE(query[0].uploaded);
EXPECT_TRUE(query[1].uploaded);
EXPECT_TRUE(query[2].uploaded);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(query[0].last_upload_attempt_time, 0);
EXPECT_GE(query[1].last_upload_attempt_time, report_1_upload_time);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(query[2].last_upload_attempt_time, report_2_upload_time);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(query[0].upload_attempts, 0);
EXPECT_EQ(query[1].upload_attempts, 3);
EXPECT_EQ(query[2].upload_attempts, 1);
}
// This test covers both query functions since they are related.
TEST_F(CrashReportDatabaseTest, GetCompletedAndNotUploadedReports) {
std::vector<CrashReportDatabase::Report> reports(5);
CreateCrashReport(&reports[0]);
CreateCrashReport(&reports[1]);
CreateCrashReport(&reports[2]);
CreateCrashReport(&reports[3]);
CreateCrashReport(&reports[4]);
const UUID& report_0_uuid = reports[0].uuid;
const UUID& report_1_uuid = reports[1].uuid;
const UUID& report_2_uuid = reports[2].uuid;
const UUID& report_3_uuid = reports[3].uuid;
const UUID& report_4_uuid = reports[4].uuid;
std::vector<CrashReportDatabase::Report> pending;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetPendingReports(&pending), CrashReportDatabase::kNoError);
std::vector<CrashReportDatabase::Report> completed;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetCompletedReports(&completed),
CrashReportDatabase::kNoError);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(pending.size(), reports.size());
EXPECT_EQ(completed.size(), 0u);
// Upload one report successfully.
UploadReport(report_1_uuid, true, "report1");
pending.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetPendingReports(&pending), CrashReportDatabase::kNoError);
completed.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetCompletedReports(&completed),
CrashReportDatabase::kNoError);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(pending.size(), 4u);
ASSERT_EQ(completed.size(), 1u);
for (const auto& report : pending) {
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_NE(report.uuid, report_1_uuid);
EXPECT_FALSE(report.file_path.empty());
}
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(completed[0].uuid, report_1_uuid);
EXPECT_EQ(completed[0].id, "report1");
EXPECT_EQ(completed[0].uploaded, true);
EXPECT_GT(completed[0].last_upload_attempt_time, 0);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(completed[0].upload_attempts, 1);
// Fail to upload one report.
UploadReport(report_2_uuid, false, std::string());
pending.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetPendingReports(&pending), CrashReportDatabase::kNoError);
completed.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetCompletedReports(&completed),
CrashReportDatabase::kNoError);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(pending.size(), 4u);
ASSERT_EQ(completed.size(), 1u);
for (const auto& report : pending) {
if (report.upload_attempts != 0) {
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(report.uuid, report_2_uuid);
EXPECT_GT(report.last_upload_attempt_time, 0);
EXPECT_FALSE(report.uploaded);
EXPECT_TRUE(report.id.empty());
}
EXPECT_FALSE(report.file_path.empty());
}
// Upload a second report.
UploadReport(report_4_uuid, true, "report_4");
pending.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetPendingReports(&pending), CrashReportDatabase::kNoError);
completed.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetCompletedReports(&completed),
CrashReportDatabase::kNoError);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(pending.size(), 3u);
ASSERT_EQ(completed.size(), 2u);
// Succeed the failed report.
UploadReport(report_2_uuid, true, "report 2");
pending.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetPendingReports(&pending), CrashReportDatabase::kNoError);
completed.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetCompletedReports(&completed),
CrashReportDatabase::kNoError);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(pending.size(), 2u);
ASSERT_EQ(completed.size(), 3u);
for (const auto& report : pending) {
EXPECT_TRUE(report.uuid == report_0_uuid || report.uuid == report_3_uuid);
EXPECT_FALSE(report.file_path.empty());
}
// Skip upload for one report.
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->SkipReportUpload(
report_3_uuid, Metrics::CrashSkippedReason::kUploadsDisabled),
CrashReportDatabase::kNoError);
pending.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetPendingReports(&pending), CrashReportDatabase::kNoError);
completed.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetCompletedReports(&completed),
CrashReportDatabase::kNoError);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
ASSERT_EQ(pending.size(), 1u);
ASSERT_EQ(completed.size(), 4u);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(pending[0].uuid, report_0_uuid);
for (const auto& report : completed) {
if (report.uuid == report_3_uuid) {
EXPECT_FALSE(report.uploaded);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(report.upload_attempts, 0);
EXPECT_EQ(report.last_upload_attempt_time, 0);
} else {
EXPECT_TRUE(report.uploaded);
EXPECT_GT(report.upload_attempts, 0);
EXPECT_GT(report.last_upload_attempt_time, 0);
}
EXPECT_FALSE(report.file_path.empty());
}
}
TEST_F(CrashReportDatabaseTest, DuelingUploads) {
CrashReportDatabase::Report report;
CreateCrashReport(&report);
std::unique_ptr<const CrashReportDatabase::UploadReport> upload_report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetReportForUploading(report.uuid, &upload_report),
CrashReportDatabase::kNoError);
std::unique_ptr<const CrashReportDatabase::UploadReport> upload_report_2;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->GetReportForUploading(report.uuid, &upload_report_2),
CrashReportDatabase::kBusyError);
EXPECT_FALSE(upload_report_2);
EXPECT_EQ(db()->RecordUploadComplete(std::move(upload_report), std::string()),
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
CrashReportDatabase::kNoError);
}
#if BUILDFLAG(IS_IOS)
TEST_F(CrashReportDatabaseTest, InterruptedIOSUploads) {
CrashReportDatabase::Report report;
CreateCrashReport(&report);
std::unique_ptr<const CrashReportDatabase::UploadReport> upload_report;
EXPECT_EQ(db()->GetReportForUploading(report.uuid, &upload_report),
CrashReportDatabase::kNoError);
// Set upload_start_time to 10 minutes ago.
time_t ten_minutes_ago = time(nullptr) - 10 * 60;
ASSERT_TRUE(
WriteXattrTimeT(report.file_path,
"org.chromium.crashpad.database.upload_start_time",
ten_minutes_ago));
std::vector<CrashReportDatabase::Report> reports;
EXPECT_EQ(db()->GetPendingReports(&reports), CrashReportDatabase::kNoError);
ASSERT_EQ(reports.size(), 1u);
reports.clear();
EXPECT_EQ(db()->GetCompletedReports(&reports), CrashReportDatabase::kNoError);
EXPECT_TRUE(reports.empty());
// Getting a stale report will automatically skip it.
std::unique_ptr<const CrashReportDatabase::UploadReport> upload_report_2;
EXPECT_EQ(db()->GetReportForUploading(report.uuid, &upload_report_2),
CrashReportDatabase::kReportNotFound);
EXPECT_FALSE(upload_report_2);
// Confirm report was moved from pending to completed.
EXPECT_EQ(db()->GetPendingReports(&reports), CrashReportDatabase::kNoError);
EXPECT_TRUE(reports.empty());
EXPECT_EQ(db()->GetCompletedReports(&reports), CrashReportDatabase::kNoError);
ASSERT_EQ(reports.size(), 1u);
EXPECT_EQ(db()->RecordUploadComplete(std::move(upload_report), std::string()),
CrashReportDatabase::kReportNotFound);
}
#endif
TEST_F(CrashReportDatabaseTest, UploadAlreadyUploaded) {
CrashReportDatabase::Report report;
CreateCrashReport(&report);
std::unique_ptr<const CrashReportDatabase::UploadReport> upload_report;
EXPECT_EQ(db()->GetReportForUploading(report.uuid, &upload_report),
CrashReportDatabase::kNoError);
EXPECT_EQ(db()->RecordUploadComplete(std::move(upload_report), std::string()),
CrashReportDatabase::kNoError);
std::unique_ptr<const CrashReportDatabase::UploadReport> upload_report_2;
EXPECT_EQ(db()->GetReportForUploading(report.uuid, &upload_report_2),
CrashReportDatabase::kReportNotFound);
EXPECT_FALSE(upload_report_2.get());
}
TEST_F(CrashReportDatabaseTest, MoveDatabase) {
std::unique_ptr<CrashReportDatabase::NewReport> new_report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->PrepareNewCrashReport(&new_report),
CrashReportDatabase::kNoError);
UUID uuid;
EXPECT_EQ(db()->FinishedWritingCrashReport(std::move(new_report), &uuid),
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
CrashReportDatabase::kNoError);
RelocateDatabase();
CrashReportDatabase::Report report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kNoError);
ExpectPreparedCrashReport(report);
}
TEST_F(CrashReportDatabaseTest, ReportRemoved) {
std::unique_ptr<CrashReportDatabase::NewReport> new_report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->PrepareNewCrashReport(&new_report),
CrashReportDatabase::kNoError);
UUID uuid;
EXPECT_EQ(db()->FinishedWritingCrashReport(std::move(new_report), &uuid),
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
CrashReportDatabase::kNoError);
CrashReportDatabase::Report report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kNoError);
EXPECT_TRUE(LoggingRemoveFile(report.file_path));
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kReportNotFound);
}
TEST_F(CrashReportDatabaseTest, DeleteReport) {
CrashReportDatabase::Report keep_pending;
CrashReportDatabase::Report delete_pending;
CrashReportDatabase::Report keep_completed;
CrashReportDatabase::Report delete_completed;
CreateCrashReport(&keep_pending);
CreateCrashReport(&delete_pending);
CreateCrashReport(&keep_completed);
CreateCrashReport(&delete_completed);
EXPECT_TRUE(FileExists(keep_pending.file_path));
EXPECT_TRUE(FileExists(delete_pending.file_path));
EXPECT_TRUE(FileExists(keep_completed.file_path));
EXPECT_TRUE(FileExists(delete_completed.file_path));
UploadReport(keep_completed.uuid, true, "1");
UploadReport(delete_completed.uuid, true, "2");
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(keep_completed.uuid, &keep_completed),
CrashReportDatabase::kNoError);
EXPECT_EQ(db()->LookUpCrashReport(delete_completed.uuid, &delete_completed),
CrashReportDatabase::kNoError);
EXPECT_TRUE(FileExists(keep_completed.file_path));
EXPECT_TRUE(FileExists(delete_completed.file_path));
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->DeleteReport(delete_pending.uuid),
CrashReportDatabase::kNoError);
EXPECT_FALSE(FileExists(delete_pending.file_path));
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(delete_pending.uuid, &delete_pending),
CrashReportDatabase::kReportNotFound);
EXPECT_EQ(db()->DeleteReport(delete_pending.uuid),
CrashReportDatabase::kReportNotFound);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->DeleteReport(delete_completed.uuid),
CrashReportDatabase::kNoError);
EXPECT_FALSE(FileExists(delete_completed.file_path));
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(delete_completed.uuid, &delete_completed),
CrashReportDatabase::kReportNotFound);
EXPECT_EQ(db()->DeleteReport(delete_completed.uuid),
CrashReportDatabase::kReportNotFound);
EXPECT_EQ(db()->LookUpCrashReport(keep_pending.uuid, &keep_pending),
CrashReportDatabase::kNoError);
EXPECT_EQ(db()->LookUpCrashReport(keep_completed.uuid, &keep_completed),
CrashReportDatabase::kNoError);
}
TEST_F(CrashReportDatabaseTest, DeleteReportEmptyingDatabase) {
CrashReportDatabase::Report report;
CreateCrashReport(&report);
EXPECT_TRUE(FileExists(report.file_path));
UploadReport(report.uuid, true, "1");
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(report.uuid, &report),
CrashReportDatabase::kNoError);
EXPECT_TRUE(FileExists(report.file_path));
// This causes an empty database to be written, make sure this is handled.
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->DeleteReport(report.uuid), CrashReportDatabase::kNoError);
EXPECT_FALSE(FileExists(report.file_path));
}
TEST_F(CrashReportDatabaseTest, ReadEmptyDatabase) {
CrashReportDatabase::Report report;
CreateCrashReport(&report);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->DeleteReport(report.uuid), CrashReportDatabase::kNoError);
// Deleting and the creating another report causes an empty database to be
// loaded. Make sure this is handled.
CrashReportDatabase::Report report2;
CreateCrashReport(&report2);
}
TEST_F(CrashReportDatabaseTest, RequestUpload) {
std::vector<CrashReportDatabase::Report> reports(2);
CreateCrashReport(&reports[0]);
CreateCrashReport(&reports[1]);
const UUID& report_0_uuid = reports[0].uuid;
const UUID& report_1_uuid = reports[1].uuid;
// Skipped report gets back to pending state after RequestUpload is called.
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->SkipReportUpload(
report_1_uuid, Metrics::CrashSkippedReason::kUploadsDisabled),
CrashReportDatabase::kNoError);
std::vector<CrashReportDatabase::Report> pending_reports;
CrashReportDatabase::OperationStatus os =
db()->GetPendingReports(&pending_reports);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(os, CrashReportDatabase::kNoError);
ASSERT_EQ(pending_reports.size(), 1u);
EXPECT_EQ(report_0_uuid, pending_reports[0].uuid);
pending_reports.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(RequestUpload(report_1_uuid), CrashReportDatabase::kNoError);
os = db()->GetPendingReports(&pending_reports);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(os, CrashReportDatabase::kNoError);
ASSERT_EQ(pending_reports.size(), 2u);
// Check individual reports.
const CrashReportDatabase::Report* explicitly_requested_report;
const CrashReportDatabase::Report* pending_report;
if (pending_reports[0].uuid == report_0_uuid) {
pending_report = &pending_reports[0];
explicitly_requested_report = &pending_reports[1];
} else {
pending_report = &pending_reports[1];
explicitly_requested_report = &pending_reports[0];
}
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(pending_report->uuid, report_0_uuid);
EXPECT_FALSE(pending_report->upload_explicitly_requested);
EXPECT_EQ(explicitly_requested_report->uuid, report_1_uuid);
EXPECT_TRUE(explicitly_requested_report->upload_explicitly_requested);
// Explicitly requested reports will not have upload_explicitly_requested bit
// after getting skipped.
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->SkipReportUpload(
report_1_uuid, Metrics::CrashSkippedReason::kUploadsDisabled),
CrashReportDatabase::kNoError);
CrashReportDatabase::Report report;
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(db()->LookUpCrashReport(report_1_uuid, &report),
CrashReportDatabase::kNoError);
EXPECT_FALSE(report.upload_explicitly_requested);
// Pending report gets correctly affected after RequestUpload is called.
pending_reports.clear();
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(RequestUpload(report_0_uuid), CrashReportDatabase::kNoError);
os = db()->GetPendingReports(&pending_reports);
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(os, CrashReportDatabase::kNoError);
EXPECT_EQ(pending_reports.size(), 1u);
EXPECT_EQ(report_0_uuid, pending_reports[0].uuid);
EXPECT_TRUE(pending_reports[0].upload_explicitly_requested);
// Already uploaded report cannot be requested for the new upload.
UploadReport(report_0_uuid, true, "1");
test: Use (actual, [un]expected) in gtest {ASSERT,EXPECT}_{EQ,NE} gtest used to require (expected, actual) ordering for arguments to EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify each side as “expected” or “actual.” Tests in Crashpad adhered to this traditional ordering. After a gtest change in February 2016, it is now agnostic with respect to the order of these arguments. This change mechanically updates all uses of these macros to (actual, expected) by reversing them. This provides consistency with our use of the logging CHECK_EQ and DCHECK_EQ macros, and makes for better readability by ordinary native speakers. The rough (but working!) conversion tool is https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py, and “git cl format” cleaned up its output. EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made a judgment that one side or the other needed to provide an “unexpected” value. Consequently, some code used (unexpected, actual) while other code used (actual, unexpected). For consistency with the new EXPECT_EQ and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE, this change also updates these use sites to (actual, unexpected) where one side can be called “unexpected” as, for example, std::string::npos can be. Unfortunately, this portion was a manual conversion. References: https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison https://github.com/google/googletest/commit/77d6b173380332b1c1bc540532641f410ec82d65 https://github.com/google/googletest/pull/713 Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3 Reviewed-on: https://chromium-review.googlesource.com/466727 Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-04-04 00:35:21 -04:00
EXPECT_EQ(RequestUpload(report_0_uuid),
CrashReportDatabase::kCannotRequestUpload);
}
TEST_F(CrashReportDatabaseTest, Attachments) {
std::unique_ptr<CrashReportDatabase::NewReport> new_report;
ASSERT_EQ(db()->PrepareNewCrashReport(&new_report),
CrashReportDatabase::kNoError);
FileWriter* attach_some_file = new_report->AddAttachment("some_file");
ASSERT_NE(attach_some_file, nullptr);
static constexpr char test_data[] = "test data";
attach_some_file->Write(test_data, sizeof(test_data));
FileWriter* failed_attach = new_report->AddAttachment("not/a valid fi!e");
EXPECT_EQ(failed_attach, nullptr);
UUID expect_uuid = new_report->ReportID();
UUID uuid;
ASSERT_EQ(db()->FinishedWritingCrashReport(std::move(new_report), &uuid),
CrashReportDatabase::kNoError);
EXPECT_EQ(uuid, expect_uuid);
CrashReportDatabase::Report report;
EXPECT_EQ(db()->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kNoError);
ExpectPreparedCrashReport(report);
std::vector<CrashReportDatabase::Report> reports;
EXPECT_EQ(db()->GetPendingReports(&reports), CrashReportDatabase::kNoError);
ASSERT_EQ(reports.size(), 1u);
EXPECT_EQ(reports[0].uuid, report.uuid);
std::unique_ptr<const CrashReportDatabase::UploadReport> upload_report;
ASSERT_EQ(db()->GetReportForUploading(reports[0].uuid, &upload_report),
CrashReportDatabase::kNoError);
std::map<std::string, FileReader*> result_attachments =
upload_report->GetAttachments();
EXPECT_EQ(result_attachments.size(), 1u);
EXPECT_NE(result_attachments.find("some_file"), result_attachments.end());
char result_buffer[sizeof(test_data)];
result_attachments["some_file"]->Read(result_buffer, sizeof(result_buffer));
EXPECT_EQ(memcmp(test_data, result_buffer, sizeof(test_data)), 0);
}
TEST_F(CrashReportDatabaseTest, OrphanedAttachments) {
std::unique_ptr<CrashReportDatabase::NewReport> new_report;
ASSERT_EQ(db()->PrepareNewCrashReport(&new_report),
CrashReportDatabase::kNoError);
FileWriter* file1 = new_report->AddAttachment("file1");
ASSERT_NE(file1, nullptr);
FileWriter* file2 = new_report->AddAttachment("file2");
ASSERT_NE(file2, nullptr);
UUID expect_uuid = new_report->ReportID();
UUID uuid;
ASSERT_EQ(db()->FinishedWritingCrashReport(std::move(new_report), &uuid),
CrashReportDatabase::kNoError);
EXPECT_EQ(uuid, expect_uuid);
CrashReportDatabase::Report report;
ASSERT_EQ(db()->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kNoError);
ASSERT_TRUE(LoggingRemoveFile(report.file_path));
Use BUILDFLAG for OS checking Use BUILDFLAG(IS_*) instead of defined(OS_*). This was generated mostly mechnically by performing the following steps: - sed -i '' -E -e 's/defined\(OS_/BUILDFLAG(IS_/g' \ -e 's%([ !])OS_([A-Z]+)%\1BUILDFLAG(IS_\2)%g' \ $(git grep -l 'OS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - sed -i '' -e 's/#ifdef BUILDFLAG(/#if BUILDFLAG(/' \ $(git grep -l '#ifdef BUILDFLAG(' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - gsed -i -z -E -e \ 's%(.*)#include "%\1#include "build/buildflag.h"\n#include "%' \ $(git grep -l 'BUILDFLAG(IS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - Spot checks to move #include "build/buildflag.h" to the correct parts of files. - sed -i '' -E -e \ 's%^(#include "build/buildflag.h")$%#include "build/build_config.h"\n\1%' \ $(grep -L '^#include "build/build_config.h"$' $(git grep -l 'BUILDFLAG(IS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm')) - Add “clang-format off” around tool usage messages. - git cl format - Update mini_chromium to 85ba51f98278 (intermediate step). TESTING ONLY). - for f in $(git grep -l '^#include "build/buildflag.h"$' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm'); do \ grep -v '^#include "build/buildflag.h"$' "${f}" > /tmp/z; \ cp /tmp/z "${f}"; done - git cl format - Update mini_chromium to 735143774c5f (intermediate step). - Update mini_chromium to f41420eb45fa (as checked in). - Update mini_chromium to 6e2f204b4ae1 (as checked in). For ease of review and inspection, each of these steps is uploaded as a new patch set in a review series. This includes an update of mini_chromium to 6e2f204b4ae1: f41420eb45fa Use BUILDFLAG for OS checking 6e2f204b4ae1 Include what you use: string_util.h uses build_config.h Bug: chromium:1234043 Change-Id: Ieef86186f094c64e59b853729737e36982f8cf69 Reviewed-on: https://chromium-review.googlesource.com/c/crashpad/crashpad/+/3400258 Reviewed-by: Joshua Peraza <jperaza@chromium.org> Commit-Queue: Mark Mentovai <mark@chromium.org>
2022-01-19 15:00:24 -05:00
#if !BUILDFLAG(IS_APPLE) && !BUILDFLAG(IS_WIN)
// CrashReportDatabaseMac stores metadata in xattrs and does not have .meta
// files.
// CrashReportDatabaseWin stores metadata in a global metadata file and not
// per report.
ASSERT_TRUE(LoggingRemoveFile(base::FilePath(
report.file_path.RemoveFinalExtension().value() + ".meta")));
#endif
ASSERT_EQ(db()->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kReportNotFound);
Use BUILDFLAG for OS checking Use BUILDFLAG(IS_*) instead of defined(OS_*). This was generated mostly mechnically by performing the following steps: - sed -i '' -E -e 's/defined\(OS_/BUILDFLAG(IS_/g' \ -e 's%([ !])OS_([A-Z]+)%\1BUILDFLAG(IS_\2)%g' \ $(git grep -l 'OS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - sed -i '' -e 's/#ifdef BUILDFLAG(/#if BUILDFLAG(/' \ $(git grep -l '#ifdef BUILDFLAG(' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - gsed -i -z -E -e \ 's%(.*)#include "%\1#include "build/buildflag.h"\n#include "%' \ $(git grep -l 'BUILDFLAG(IS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - Spot checks to move #include "build/buildflag.h" to the correct parts of files. - sed -i '' -E -e \ 's%^(#include "build/buildflag.h")$%#include "build/build_config.h"\n\1%' \ $(grep -L '^#include "build/build_config.h"$' $(git grep -l 'BUILDFLAG(IS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm')) - Add “clang-format off” around tool usage messages. - git cl format - Update mini_chromium to 85ba51f98278 (intermediate step). TESTING ONLY). - for f in $(git grep -l '^#include "build/buildflag.h"$' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm'); do \ grep -v '^#include "build/buildflag.h"$' "${f}" > /tmp/z; \ cp /tmp/z "${f}"; done - git cl format - Update mini_chromium to 735143774c5f (intermediate step). - Update mini_chromium to f41420eb45fa (as checked in). - Update mini_chromium to 6e2f204b4ae1 (as checked in). For ease of review and inspection, each of these steps is uploaded as a new patch set in a review series. This includes an update of mini_chromium to 6e2f204b4ae1: f41420eb45fa Use BUILDFLAG for OS checking 6e2f204b4ae1 Include what you use: string_util.h uses build_config.h Bug: chromium:1234043 Change-Id: Ieef86186f094c64e59b853729737e36982f8cf69 Reviewed-on: https://chromium-review.googlesource.com/c/crashpad/crashpad/+/3400258 Reviewed-by: Joshua Peraza <jperaza@chromium.org> Commit-Queue: Mark Mentovai <mark@chromium.org>
2022-01-19 15:00:24 -05:00
#if BUILDFLAG(IS_WIN)
const std::wstring uuid_string = uuid.ToWString();
#else
const std::string uuid_string = uuid.ToString();
#endif
base::FilePath report_attachments_dir(
path().Append(FILE_PATH_LITERAL("attachments")).Append(uuid_string));
base::FilePath file_path1(
report_attachments_dir.Append(FILE_PATH_LITERAL("file1")));
base::FilePath file_path2(
report_attachments_dir.Append(FILE_PATH_LITERAL("file2")));
EXPECT_TRUE(FileExists(file_path1));
EXPECT_TRUE(FileExists(file_path1));
Use BUILDFLAG for OS checking Use BUILDFLAG(IS_*) instead of defined(OS_*). This was generated mostly mechnically by performing the following steps: - sed -i '' -E -e 's/defined\(OS_/BUILDFLAG(IS_/g' \ -e 's%([ !])OS_([A-Z]+)%\1BUILDFLAG(IS_\2)%g' \ $(git grep -l 'OS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - sed -i '' -e 's/#ifdef BUILDFLAG(/#if BUILDFLAG(/' \ $(git grep -l '#ifdef BUILDFLAG(' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - gsed -i -z -E -e \ 's%(.*)#include "%\1#include "build/buildflag.h"\n#include "%' \ $(git grep -l 'BUILDFLAG(IS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - Spot checks to move #include "build/buildflag.h" to the correct parts of files. - sed -i '' -E -e \ 's%^(#include "build/buildflag.h")$%#include "build/build_config.h"\n\1%' \ $(grep -L '^#include "build/build_config.h"$' $(git grep -l 'BUILDFLAG(IS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm')) - Add “clang-format off” around tool usage messages. - git cl format - Update mini_chromium to 85ba51f98278 (intermediate step). TESTING ONLY). - for f in $(git grep -l '^#include "build/buildflag.h"$' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm'); do \ grep -v '^#include "build/buildflag.h"$' "${f}" > /tmp/z; \ cp /tmp/z "${f}"; done - git cl format - Update mini_chromium to 735143774c5f (intermediate step). - Update mini_chromium to f41420eb45fa (as checked in). - Update mini_chromium to 6e2f204b4ae1 (as checked in). For ease of review and inspection, each of these steps is uploaded as a new patch set in a review series. This includes an update of mini_chromium to 6e2f204b4ae1: f41420eb45fa Use BUILDFLAG for OS checking 6e2f204b4ae1 Include what you use: string_util.h uses build_config.h Bug: chromium:1234043 Change-Id: Ieef86186f094c64e59b853729737e36982f8cf69 Reviewed-on: https://chromium-review.googlesource.com/c/crashpad/crashpad/+/3400258 Reviewed-by: Joshua Peraza <jperaza@chromium.org> Commit-Queue: Mark Mentovai <mark@chromium.org>
2022-01-19 15:00:24 -05:00
#if BUILDFLAG(IS_WIN)
// On Windows, reports removed from metadata are counted, even if the file
// is not on the disk.
EXPECT_EQ(db()->CleanDatabase(0), 1);
#else
EXPECT_EQ(db()->CleanDatabase(0), 0);
#endif
EXPECT_FALSE(FileExists(file_path1));
EXPECT_FALSE(FileExists(file_path2));
EXPECT_FALSE(FileExists(report_attachments_dir));
}
// This test uses knowledge of the database format to break it, so it only
// applies to the unfified database implementation.
Use BUILDFLAG for OS checking Use BUILDFLAG(IS_*) instead of defined(OS_*). This was generated mostly mechnically by performing the following steps: - sed -i '' -E -e 's/defined\(OS_/BUILDFLAG(IS_/g' \ -e 's%([ !])OS_([A-Z]+)%\1BUILDFLAG(IS_\2)%g' \ $(git grep -l 'OS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - sed -i '' -e 's/#ifdef BUILDFLAG(/#if BUILDFLAG(/' \ $(git grep -l '#ifdef BUILDFLAG(' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - gsed -i -z -E -e \ 's%(.*)#include "%\1#include "build/buildflag.h"\n#include "%' \ $(git grep -l 'BUILDFLAG(IS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - Spot checks to move #include "build/buildflag.h" to the correct parts of files. - sed -i '' -E -e \ 's%^(#include "build/buildflag.h")$%#include "build/build_config.h"\n\1%' \ $(grep -L '^#include "build/build_config.h"$' $(git grep -l 'BUILDFLAG(IS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm')) - Add “clang-format off” around tool usage messages. - git cl format - Update mini_chromium to 85ba51f98278 (intermediate step). TESTING ONLY). - for f in $(git grep -l '^#include "build/buildflag.h"$' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm'); do \ grep -v '^#include "build/buildflag.h"$' "${f}" > /tmp/z; \ cp /tmp/z "${f}"; done - git cl format - Update mini_chromium to 735143774c5f (intermediate step). - Update mini_chromium to f41420eb45fa (as checked in). - Update mini_chromium to 6e2f204b4ae1 (as checked in). For ease of review and inspection, each of these steps is uploaded as a new patch set in a review series. This includes an update of mini_chromium to 6e2f204b4ae1: f41420eb45fa Use BUILDFLAG for OS checking 6e2f204b4ae1 Include what you use: string_util.h uses build_config.h Bug: chromium:1234043 Change-Id: Ieef86186f094c64e59b853729737e36982f8cf69 Reviewed-on: https://chromium-review.googlesource.com/c/crashpad/crashpad/+/3400258 Reviewed-by: Joshua Peraza <jperaza@chromium.org> Commit-Queue: Mark Mentovai <mark@chromium.org>
2022-01-19 15:00:24 -05:00
#if !BUILDFLAG(IS_APPLE) && !BUILDFLAG(IS_WIN)
TEST_F(CrashReportDatabaseTest, CleanBrokenDatabase) {
// Remove report files if metadata goes missing.
CrashReportDatabase::Report report;
ASSERT_NO_FATAL_FAILURE(CreateCrashReport(&report));
const base::FilePath metadata(
report.file_path.RemoveFinalExtension().value() +
FILE_PATH_LITERAL(".meta"));
ASSERT_TRUE(PathExists(report.file_path));
ASSERT_TRUE(PathExists(metadata));
ASSERT_TRUE(LoggingRemoveFile(metadata));
EXPECT_EQ(db()->CleanDatabase(0), 1);
EXPECT_FALSE(PathExists(report.file_path));
EXPECT_FALSE(PathExists(metadata));
// Remove metadata files if reports go missing.
ASSERT_NO_FATAL_FAILURE(CreateCrashReport(&report));
const base::FilePath metadata2(
report.file_path.RemoveFinalExtension().value() +
FILE_PATH_LITERAL(".meta"));
ASSERT_TRUE(PathExists(report.file_path));
ASSERT_TRUE(PathExists(metadata2));
ASSERT_TRUE(LoggingRemoveFile(report.file_path));
EXPECT_EQ(db()->CleanDatabase(0), 1);
EXPECT_FALSE(PathExists(report.file_path));
EXPECT_FALSE(PathExists(metadata2));
// Remove stale new files.
std::unique_ptr<CrashReportDatabase::NewReport> new_report;
EXPECT_EQ(db()->PrepareNewCrashReport(&new_report),
CrashReportDatabase::kNoError);
new_report->Writer()->Close();
EXPECT_EQ(db()->CleanDatabase(0), 1);
// Remove stale lock files and their associated reports.
ASSERT_NO_FATAL_FAILURE(CreateCrashReport(&report));
const base::FilePath metadata3(
report.file_path.RemoveFinalExtension().value() +
FILE_PATH_LITERAL(".meta"));
ASSERT_TRUE(PathExists(report.file_path));
ASSERT_TRUE(PathExists(metadata3));
const base::FilePath lockpath(
report.file_path.RemoveFinalExtension().value() +
FILE_PATH_LITERAL(".lock"));
ScopedFileHandle handle(LoggingOpenFileForWrite(
lockpath, FileWriteMode::kCreateOrFail, FilePermissions::kOwnerOnly));
ASSERT_TRUE(handle.is_valid());
time_t expired_timestamp = time(nullptr) - 60 * 60 * 24 * 3;
ASSERT_TRUE(LoggingWriteFile(
handle.get(), &expired_timestamp, sizeof(expired_timestamp)));
ASSERT_TRUE(LoggingCloseFile(handle.release()));
EXPECT_EQ(db()->CleanDatabase(0), 1);
EXPECT_FALSE(PathExists(report.file_path));
EXPECT_FALSE(PathExists(metadata3));
}
Use BUILDFLAG for OS checking Use BUILDFLAG(IS_*) instead of defined(OS_*). This was generated mostly mechnically by performing the following steps: - sed -i '' -E -e 's/defined\(OS_/BUILDFLAG(IS_/g' \ -e 's%([ !])OS_([A-Z]+)%\1BUILDFLAG(IS_\2)%g' \ $(git grep -l 'OS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - sed -i '' -e 's/#ifdef BUILDFLAG(/#if BUILDFLAG(/' \ $(git grep -l '#ifdef BUILDFLAG(' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - gsed -i -z -E -e \ 's%(.*)#include "%\1#include "build/buildflag.h"\n#include "%' \ $(git grep -l 'BUILDFLAG(IS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm') - Spot checks to move #include "build/buildflag.h" to the correct parts of files. - sed -i '' -E -e \ 's%^(#include "build/buildflag.h")$%#include "build/build_config.h"\n\1%' \ $(grep -L '^#include "build/build_config.h"$' $(git grep -l 'BUILDFLAG(IS_' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm')) - Add “clang-format off” around tool usage messages. - git cl format - Update mini_chromium to 85ba51f98278 (intermediate step). TESTING ONLY). - for f in $(git grep -l '^#include "build/buildflag.h"$' '**/*.c' '**/*.cc' '**/*.h' '**/*.m' '**/*.mm'); do \ grep -v '^#include "build/buildflag.h"$' "${f}" > /tmp/z; \ cp /tmp/z "${f}"; done - git cl format - Update mini_chromium to 735143774c5f (intermediate step). - Update mini_chromium to f41420eb45fa (as checked in). - Update mini_chromium to 6e2f204b4ae1 (as checked in). For ease of review and inspection, each of these steps is uploaded as a new patch set in a review series. This includes an update of mini_chromium to 6e2f204b4ae1: f41420eb45fa Use BUILDFLAG for OS checking 6e2f204b4ae1 Include what you use: string_util.h uses build_config.h Bug: chromium:1234043 Change-Id: Ieef86186f094c64e59b853729737e36982f8cf69 Reviewed-on: https://chromium-review.googlesource.com/c/crashpad/crashpad/+/3400258 Reviewed-by: Joshua Peraza <jperaza@chromium.org> Commit-Queue: Mark Mentovai <mark@chromium.org>
2022-01-19 15:00:24 -05:00
#endif // !BUILDFLAG(IS_APPLE) && !BUILDFLAG(IS_WIN)
TEST_F(CrashReportDatabaseTest, TotalSize_MainReportOnly) {
std::unique_ptr<CrashReportDatabase::NewReport> new_report;
ASSERT_EQ(db()->PrepareNewCrashReport(&new_report),
CrashReportDatabase::kNoError);
// Main report.
static constexpr char main_report_data[] = "dlbvandslhb";
ASSERT_TRUE(
new_report->Writer()->Write(main_report_data, sizeof(main_report_data)));
UUID uuid;
ASSERT_EQ(db()->FinishedWritingCrashReport(std::move(new_report), &uuid),
CrashReportDatabase::kNoError);
CrashReportDatabase::Report report;
ASSERT_EQ(db()->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kNoError);
EXPECT_EQ(report.total_size, sizeof(main_report_data));
}
TEST_F(CrashReportDatabaseTest, GetReportSize_RightSizeWithAttachments) {
std::unique_ptr<CrashReportDatabase::NewReport> new_report;
ASSERT_EQ(db()->PrepareNewCrashReport(&new_report),
CrashReportDatabase::kNoError);
// Main report.
static constexpr char main_report_data[] = "dlbvandslhb";
ASSERT_TRUE(
new_report->Writer()->Write(main_report_data, sizeof(main_report_data)));
// First attachment.
FileWriter* attachment_1 = new_report->AddAttachment("my_attachment_1");
ASSERT_NE(attachment_1, nullptr);
static constexpr char attachment_1_data[] = "vKDnidhvbiudshoihbvdsoiuh nhh";
attachment_1->Write(attachment_1_data, sizeof(attachment_1_data));
// Second attachment.
FileWriter* attachment_2 = new_report->AddAttachment("my_attachment_2");
ASSERT_NE(attachment_2, nullptr);
static constexpr char attachment_2_data[] = "sgvsvgusiyguysigfkhpmo-[";
attachment_2->Write(attachment_2_data, sizeof(attachment_2_data));
UUID uuid;
ASSERT_EQ(db()->FinishedWritingCrashReport(std::move(new_report), &uuid),
CrashReportDatabase::kNoError);
CrashReportDatabase::Report report;
ASSERT_EQ(db()->LookUpCrashReport(uuid, &report),
CrashReportDatabase::kNoError);
EXPECT_EQ(report.total_size,
sizeof(main_report_data) + sizeof(attachment_1_data) +
sizeof(attachment_2_data));
}
TEST_F(CrashReportDatabaseTest, InitializeFromLargerFileRetainsClientId) {
// Initialize the database for the first time, creating it.
ASSERT_TRUE(db());
const base::FilePath settings_path =
path().Append(FILE_PATH_LITERAL("settings.dat"));
EXPECT_FALSE(FileExists(settings_path));
Settings* settings = db()->GetSettings();
ASSERT_TRUE(settings);
EXPECT_TRUE(FileExists(settings_path));
UUID client_id;
ASSERT_TRUE(settings->GetClientID(&client_id));
EXPECT_NE(client_id, UUID());
// Close and reopen the database at the same path.
ResetDatabase();
EXPECT_FALSE(db());
EXPECT_TRUE(FileExists(settings_path));
// Append some data, to ensure that we can open settings even if a future
// version has added additional field to Settings (forwards compatible).
FileWriter settings_writer;
ASSERT_TRUE(settings_writer.Open(
settings_path, FileWriteMode::kReuseOrFail, FilePermissions::kOwnerOnly));
ASSERT_NE(settings_writer.Seek(0, SEEK_END), 0);
constexpr uint64_t extra_garbage = 0xBADF00D;
ASSERT_TRUE(settings_writer.Write(&extra_garbage, sizeof(extra_garbage)));
settings_writer.Close();
auto db = CrashReportDatabase::InitializeWithoutCreating(path());
ASSERT_TRUE(db);
settings = db->GetSettings();
ASSERT_TRUE(settings);
// Make sure that the reopened settings retained the original client id and
// wasn't recreated.
UUID reopened_client_id;
ASSERT_TRUE(settings->GetClientID(&reopened_client_id));
EXPECT_EQ(client_id, reopened_client_id);
}
} // namespace
} // namespace test
} // namespace crashpad