![\"01-prjname\"](\"http:\/\/visualgdb.com\/w\/wp-content\/uploads\/2016\/11\/01-prjname1.png\")
<\/a><\/li>\n<\/a><\/li>\n<\/a><\/li>\n<\/a><\/li>\n<\/a><\/li>\n<\/a><\/li>\n#include <gtest\/gtest.h>\r\n#include <stdio.h>\r\n\r\nbool IsWordAligned(void *addr)\r\n{\r\n\u00a0\u00a0\u00a0 return (((unsigned long)addr) & 0x3) == 0;\r\n}\r\n\r\nTEST(DemoTests, AllocationTest)\r\n{\r\n\u00a0\u00a0\u00a0 char *p = new char[1024];\r\n\u00a0\u00a0\u00a0 ASSERT_TRUE(p != NULL);\r\n\u00a0\u00a0\u00a0 ASSERT_PRED1(IsWordAligned, p);\r\n\u00a0\u00a0\u00a0 delete[] p;\r\n\u00a0\u00a0 \u00a0\r\n\u00a0\u00a0\u00a0 ASSERT_THROW(p = new char[2 * 1024 * 1024 * 1024], std::bad_alloc);\r\n}<\/pre>\nThe code above shows 3 different assertion macros:<\/p>\n
- \n
- ASSERT_TRUE() is used to quickly check that a condition is true. Note that ASSERT_NE() will not work with NULL due to limitations of GoogleTest.<\/li>\n
- ASSERT_PRED1() is used to check the value with a custom check function (a.k.a. predicate)<\/li>\n
- ASSERT_THROW() is used to verify that the checked expression throws an exception (we check that trying to allocate too many bytes results in a std::bad_alloc)<\/li>\n<\/ul>\n<\/li>\n
- Build the project and run the test. Ensure that it passes:
![\"07-throw\"](\"https:\/\/visualgdb.com\/w\/wp-content\/uploads\/2016\/11\/07-throw.png\")
<\/a><\/li>\n- Now we will provide a custom implementation of the new() operator that will simply call malloc() and will see if the our test can catch this:\n
void * operator new(size_t size)\r\n{\r\n\u00a0\u00a0\u00a0 return malloc(size);\r\n}<\/pre>\n<\/li>\n- Run the test and see how the test report provides a detailed message on what went wrong:
![\"08-nothrow\"](\"https:\/\/visualgdb.com\/w\/wp-content\/uploads\/2016\/11\/08-nothrow.png\")
<\/a><\/li>\n- Now we will cause the IsWordAligned() check to fail by adding 1 to the allocated pointer:\n
\u00a0\u00a0\u00a0 ASSERT_PRED1(IsWordAligned, p + 1);<\/pre>\n
![\"09-align\"](\"https:\/\/visualgdb.com\/w\/wp-content\/uploads\/2016\/11\/09-align.png\")
<\/a><\/li>\n- Note how the test now fails, but the shown message is not very informative. You could improve its readability by changing IsWordAligned() to return testing::AssertionResult<\/strong> with\u00a0 some additional information instead of just bool<\/strong>:\n
testing::AssertionResult IsWordAligned(void *addr)\r\n{\r\n\u00a0\u00a0\u00a0 if ((((unsigned long)addr) & 0x3) == 0)\r\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 return testing::AssertionSuccess();\r\n\u00a0\u00a0\u00a0 else\r\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 return testing::AssertionFailure() << addr << \" is not word-aligned\";\r\n}<\/pre>\nChange the line calling IsWordAligned() as follows:<\/p>\n
\u00a0\u00a0\u00a0 ASSERT_TRUE(IsWordAligned(p + 1));<\/pre>\n<\/li>\n
- The error message will now be more meaningful and will show the mismatched value:
![\"10-alignmsg\"](\"https:\/\/visualgdb.com\/w\/wp-content\/uploads\/2016\/11\/10-alignmsg.png\")
<\/a><\/li>\n- Now we will show how to use test fixtures to define initialization and cleanup code shared between several tests. Start with defining the following class:\n
class RNGTests : public::testing::Test\r\n{\r\nprotected:\r\n\u00a0\u00a0\u00a0 virtual void SetUp() override\r\n\u00a0\u00a0\u00a0 {\r\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 srand(123);\r\n\u00a0\u00a0\u00a0 }\r\n};<\/pre>\n<\/li>\n- Then define a basic test that will check the first value returned by rand():\n
TEST_F(RNGTests, BasicTest)\r\n{\r\n\u00a0\u00a0\u00a0 ASSERT_EQ(rand(), 128959393);\r\n}<\/pre>\nNote that we used TEST_F() instead of just TEST() to declare the test. This will ensure that the RNGTests::SetUp() will be called before the test and RNGTests::TearDown() will be called after the test.<\/li>\n
- Run the test and ensure it passes:
![\"11-rng\"](\"https:\/\/visualgdb.com\/w\/wp-content\/uploads\/2016\/11\/11-rng.png\")
<\/a>The value returned by rand() may not match the one shown here, as different implementations of rand() may work differently. If this happens, edit the test accordingly.<\/li>\n- You can use the Debug Tests command to quickly verify that srand() actually gets called:
![\"13-srand\"](\"https:\/\/visualgdb.com\/w\/wp-content\/uploads\/2016\/11\/13-srand.png\")
<\/a><\/li>\n- To understand how exactly GoogleTest handles the fixtures, you can right-click on the line defining your test and select “Preprocess Selected Lines”. VisualGDB will show that defining a test with TEST_F() simply defines a class derived from your fixture class:
![\"12-expand\"](\"https:\/\/visualgdb.com\/w\/wp-content\/uploads\/2016\/11\/12-expand.png\")
<\/a><\/li>\n- If you are not using C++11 or later, you may encounter strange IntelliSense errors:
![\"14-sense\"](\"https:\/\/visualgdb.com\/w\/wp-content\/uploads\/2016\/11\/14-sense.png\")
<\/a><\/li>\n- The easiest way to fix those is to switch the language standard to C++11 or later:
![\"15-c++11\"](\"https:\/\/visualgdb.com\/w\/wp-content\/uploads\/2016\/11\/15-c-11.png\")
<\/a><\/li>\n<\/ol>\nYou can read more about the advanced features of the GoogleTest framework here<\/a> or look through this tutorial<\/a> to see how to use VisualGDB with CppUTest.<\/p>\n","protected":false},"excerpt":{"rendered":"
This tutorial shows how to use the GoogleTest framework to develop unit tests for Linux with VisualGDB. We will create<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[128],"tags":[137,129],"_links":{"self":[{"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/posts\/2315"}],"collection":[{"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/comments?post=2315"}],"version-history":[{"count":2,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/posts\/2315\/revisions"}],"predecessor-version":[{"id":2333,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/posts\/2315\/revisions\/2333"}],"wp:attachment":[{"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/media?parent=2315"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/categories?post=2315"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/tags?post=2315"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
- Now we will provide a custom implementation of the new() operator that will simply call malloc() and will see if the our test can catch this:\n