<b>P</b>ump is <b>U</b>seful for <b>M</b>eta <b>P</b>rogramming.
# The Problem #
Template and macro libraries often need to define many classes,
functions, or macros that vary only (or almost only) in the number of
arguments they take. It's a lot of repetitive, mechanical, and
error-prone work.
Variadic templates and variadic macros can alleviate the problem.
However, while both are being considered by the C++ committee, neither
is in the standard yet or widely supported by compilers. Thus they
are often not a good choice, especially when your code needs to be
portable. And their capabilities are still limited.
As a result, authors of such libraries often have to write scripts to
generate their implementation. However, our experience is that it's
tedious to write such scripts, which tend to reflect the structure of
the generated code poorly and are often hard to read and edit. For
example, a small change needed in the generated code may require some
non-intuitive, non-trivial changes in the script. This is especially
painful when experimenting with the code.
# Our Solution #
Pump (for Pump is Useful for Meta Programming, Pretty Useful for Meta
Programming, or Practical Utility for Meta Programming, whichever you
prefer) is a simple meta-programming tool for C++. The idea is that a
programmer writes a `foo.pump` file which contains C++ code plus meta
code that manipulates the C++ code. The meta code can handle
iterations over a range, nested iterations, local meta variable
definitions, simple arithmetic, and conditional expressions. You can
view it as a small Domain-Specific Language. The meta language is
designed to be non-intrusive (s.t. it won't confuse Emacs' C++ mode,
for example) and concise, making Pump code intuitive and easy to
maintain.
## Highlights ##
* The implementation is in a single Python script and thus ultra portable: no build or installation is needed and it works cross platforms.
* Pump tries to be smart with respect to [Google's style guide](http://code.google.com/p/google-styleguide/): it breaks long lines (easy to have when they are generated) at acceptable places to fit within 80 columns and indent the continuation lines correctly.
* The format is human-readable and more concise than XML.
* The format works relatively well with Emacs' C++ mode.
## Examples ##
The following Pump code (where meta keywords start with `$`, `[[` and `]]` are meta brackets, and `$$` starts a meta comment that ends with the line):
```
$var n = 3 $$ Defines a meta variable n.
$range i 0..n $$ Declares the range of meta iterator i (inclusive).
$for i [[
$$ Meta loop.
// Foo$i does blah for $i-ary predicates.
$range j 1..i
template <size_t N $for j [[, typename A$j]]>
class Foo$i {
$if i == 0 [[
blah a;
]] $elif i <= 2 [[
blah b;
]] $else [[
blah c;
]]
};
]]
```
will be translated by the Pump compiler to:
```
// Foo0 does blah for 0-ary predicates.
template <size_t N>
class Foo0 {
blah a;
};
// Foo1 does blah for 1-ary predicates.
template <size_t N, typename A1>
class Foo1 {
blah b;
};
// Foo2 does blah for 2-ary predicates.
template <size_t N, typename A1, typename A2>
class Foo2 {
blah b;
};
// Foo3 does blah for 3-ary predicates.
template <size_t N, typename A1, typename A2, typename A3>
class Foo3 {
blah c;
};
```
In another example,
```
$range i 1..n
Func($for i + [[a$i]]);
$$ The text between i and [[ is the separator between iterations.
```
will generate one of the following lines (without the comments), depending on the value of `n`:
```
Func(); // If n is 0.
Func(a1); // If n is 1.
Func(a1 + a2); // If n is 2.
Func(a1 + a2 + a3); // If n is 3.
// And so on...
```
## Constructs ##
We support the following meta programming constructs:
| `$var id = exp` | Defines a named constant value. `$id` is valid util the end of the current meta lexical block. |
**Note:** To give the user some freedom in formatting the Pump source
code, Pump ignores a new-line character if it's right after `$for foo`
or next to `[[` or `]]`. Without this rule you'll often be forced to write
very long lines to get the desired output. Therefore sometimes you may
need to insert an extra new-line in such places for a new-line to show
up in your output.
## Grammar ##
```
code ::= atomic_code*
atomic_code ::= $var id = exp
| $var id = [[ code ]]
| $range id exp..exp
| $for id sep [[ code ]]
| $($)
| $id
| $(exp)
| $if exp [[ code ]] else_branch
| [[ code ]]
| cpp_code
sep ::= cpp_code | empty_string
else_branch ::= $else [[ code ]]
| $elif exp [[ code ]] else_branch
| empty_string
exp ::= simple_expression_in_Python_syntax
```
## Code ##
You can find the source code of Pump in [scripts/pump.py](../scripts/pump.py). It is still
very unpolished and lacks automated tests, although it has been
successfully used many times. If you find a chance to use it in your
project, please let us know what you think! We also welcome help on
improving Pump.
## Real Examples ##
You can find real-world applications of Pump in [Google Test](http://www.google.com/codesearch?q=file%3A\.pump%24+package%3Ahttp%3A%2F%2Fgoogletest\.googlecode\.com) and [Google Mock](http://www.google.com/codesearch?q=file%3A\.pump%24+package%3Ahttp%3A%2F%2Fgooglemock\.googlecode\.com). The source file `foo.h.pump` generates `foo.h`.
## Tips ##
* If a meta variable is followed by a letter or digit, you can separate them using `[[]]`, which inserts an empty string. For example `Foo$j[[]]Helper` generate `Foo1Helper` when `j` is 1.
* To avoid extra-long Pump source lines, you can break a line anywhere you want by inserting `[[]]` followed by a new line. Since any new-line character next to `[[` or `]]` is ignored, the generated code won't contain this new line.
*[Sample #7](../samples/sample7_unittest.cc) teaches the basics of value-parameterized tests.
*[Sample #8](../samples/sample8_unittest.cc) shows using `Combine()` in value-parameterized tests.
*[Sample #9](../samples/sample9_unittest.cc) shows use of the listener API to modify Google Test's console output and the use of its reflection API to inspect test results.
*[Sample #10](../samples/sample10_unittest.cc) shows use of the listener API to implement a primitive memory leak checker.
This guide will explain how to use the Google Testing Framework in your Xcode projects on Mac OS X. This tutorial begins by quickly explaining what to do for experienced users. After the quick start, the guide goes provides additional explanation about each step.
# Quick Start #
Here is the quick guide for using Google Test in your Xcode project.
1. Download the source from the [website](http://code.google.com/p/googletest) using this command: `svn checkout http://googletest.googlecode.com/svn/trunk/ googletest-read-only`
1. Open up the `gtest.xcodeproj` in the `googletest-read-only/xcode/` directory and build the gtest.framework.
1. Create a new "Shell Tool" target in your Xcode project called something like "UnitTests"
1. Add the gtest.framework to your project and add it to the "Link Binary with Libraries" build phase of "UnitTests"
1. Add your unit test source code to the "Compile Sources" build phase of "UnitTests"
1. Edit the "UnitTests" executable and add an environment variable named "DYLD\_FRAMEWORK\_PATH" with a value equal to the path to the framework containing the gtest.framework relative to the compiled executable.
1. Build and Go
The following sections further explain each of the steps listed above in depth, describing in more detail how to complete it including some variations.
# Get the Source #
Currently, the gtest.framework discussed here isn't available in a tagged release of Google Test, it is only available in the trunk. As explained at the Google Test [site](http://code.google.com/p/googletest/source/checkout">svn), you can get the code from anonymous SVN with this command:
Alternatively, if you are working with Subversion in your own code base, you can add Google Test as an external dependency to your own Subversion repository. By following this approach, everyone that checks out your svn repository will also receive a copy of Google Test (a specific version, if you wish) without having to check it out explicitly. This makes the set up of your project simpler and reduces the copied code in the repository.
To use `svn:externals`, decide where you would like to have the external source reside. You might choose to put the external source inside the trunk, because you want it to be part of the branch when you make a release. However, keeping it outside the trunk in a version-tagged directory called something like `third-party/googletest/1.0.1`, is another option. Once the location is established, use `svn propedit svn:externals _directory_` to set the svn:externals property on a directory in your repository. This directory won't contain the code, but be its versioned parent directory.
The command `svn propedit` will bring up your Subversion editor, making editing the long, (potentially multi-line) property simpler. This same method can be used to check out a tagged branch, by using the appropriate URL (e.g. `http://googletest.googlecode.com/svn/tags/release-1.0.1`). Additionally, the svn:externals property allows the specification of a particular revision of the trunk with the `-r_##_` option (e.g. `externals/src/googletest -r60 http://googletest.googlecode.com/svn/trunk`).
Here is an example of using the svn:externals properties on a trunk (read via `svn propget`) of a project. This value checks out a copy of Google Test into the `trunk/externals/src/googletest/` directory.
The next step is to build and add the gtest.framework to your own project. This guide describes two common ways below.
***Option 1** --- The simplest way to add Google Test to your own project, is to open gtest.xcodeproj (found in the xcode/ directory of the Google Test trunk) and build the framework manually. Then, add the built framework into your project using the "Add->Existing Framework..." from the context menu or "Project->Add..." from the main menu. The gtest.framework is relocatable and contains the headers and object code that you'll need to make tests. This method requires rebuilding every time you upgrade Google Test in your project.
***Option 2** --- If you are going to be living off the trunk of Google Test, incorporating its latest features into your unit tests (or are a Google Test developer yourself). You'll want to rebuild the framework every time the source updates. to do this, you'll need to add the gtest.xcodeproj file, not the framework itself, to your own Xcode project. Then, from the build products that are revealed by the project's disclosure triangle, you can find the gtest.framework, which can be added to your targets (discussed below).
# Make a Test Target #
To start writing tests, make a new "Shell Tool" target. This target template is available under BSD, Cocoa, or Carbon. Add your unit test source code to the "Compile Sources" build phase of the target.
Next, you'll want to add gtest.framework in two different ways, depending upon which option you chose above.
***Option 1** --- During compilation, Xcode will need to know that you are linking against the gtest.framework. Add the gtest.framework to the "Link Binary with Libraries" build phase of your test target. This will include the Google Test headers in your header search path, and will tell the linker where to find the library.
***Option 2** --- If your working out of the trunk, you'll also want to add gtest.framework to your "Link Binary with Libraries" build phase of your test target. In addition, you'll want to add the gtest.framework as a dependency to your unit test target. This way, Xcode will make sure that gtest.framework is up to date, every time your build your target. Finally, if you don't share build directories with Google Test, you'll have to copy the gtest.framework into your own build products directory using a "Run Script" build phase.
# Set Up the Executable Run Environment #
Since the unit test executable is a shell tool, it doesn't have a bundle with a `Contents/Frameworks` directory, in which to place gtest.framework. Instead, the dynamic linker must be told at runtime to search for the framework in another location. This can be accomplished by setting the "DYLD\_FRAMEWORK\_PATH" environment variable in the "Edit Active Executable ..." Arguments tab, under "Variables to be set in the environment:". The path for this value is the path (relative or absolute) of the directory containing the gtest.framework.
If you haven't set up the DYLD\_FRAMEWORK\_PATH, correctly, you might get a message like this:
```
[Session started at 2008-08-15 06:23:57 -0600.]
dyld: Library not loaded: @loader_path/../Frameworks/gtest.framework/Versions/A/gtest
To correct this problem, got to the directory containing the executable named in "Referenced from:" value in the error message above. Then, with the terminal in this location, find the relative path to the directory containing the gtest.framework. That is the value you'll need to set as the DYLD\_FRAMEWORK\_PATH.
# Build and Go #
Now, when you click "Build and Go", the test will be executed. Dumping out something like this:
```
[Session started at 2008-08-06 06:36:13 -0600.]
[==========] Running 2 tests from 1 test case.
[----------] Global test environment set-up.
[----------] 2 tests from WidgetInitializerTest
[ RUN ] WidgetInitializerTest.TestConstructor
[ OK ] WidgetInitializerTest.TestConstructor
[ RUN ] WidgetInitializerTest.TestConversion
[ OK ] WidgetInitializerTest.TestConversion
[----------] Global test environment tear-down
[==========] 2 tests from 1 test case ran.
[ PASSED ] 2 tests.
The Debugger has exited with status 0.
```
# Summary #
Unit testing is a valuable way to ensure your data model stays valid even during rapid development or refactoring. The Google Testing Framework is a great unit testing framework for C and C++ which integrates well with an Xcode development environment.
| `ASSERT_THROW(`_statement_, _exception\_type_`);` | `EXPECT_THROW(`_statement_, _exception\_type_`);` | _statement_ throws an exception of the given type |
| `ASSERT_ANY_THROW(`_statement_`);` | `EXPECT_ANY_THROW(`_statement_`);` | _statement_ throws an exception of any type |
In the above, _predn_ is an _n_-ary predicate function or functor, where
_val1_, _val2_, ..., and _valn_ are its arguments. The assertion succeeds
if the predicate returns `true` when applied to the given arguments, and fails
otherwise. When the assertion fails, it prints the value of each argument. In
either case, the arguments are evaluated exactly once.
Here's an example. Given
```
// Returns true iff m and n have no common divisors except 1.
bool MutuallyPrime(int m, int n) { ... }
const int a = 3;
const int b = 4;
const int c = 10;
```
the assertion `EXPECT_PRED2(MutuallyPrime, a, b);` will succeed, while the
assertion `EXPECT_PRED2(MutuallyPrime, b, c);` will fail with the message
<pre>
!MutuallyPrime(b, c) is false, where<br>
b is 4<br>
c is 10<br>
</pre>
**Notes:**
1. If you see a compiler error "no matching function to call" when using `ASSERT_PRED*` or `EXPECT_PRED*`, please see [this](V1_7_FAQ.md#the-compiler-complains-about-undefined-references-to-some-static-const-member-variables-but-i-did-define-them-in-the-class-body-whats-wrong) for how to resolve it.
1. Currently we only provide predicate assertions of arity <= 5. If you need a higher-arity assertion, let us know.
_Availability_: Linux, Windows, Mac
### Using a Function That Returns an AssertionResult ###
While `EXPECT_PRED*()` and friends are handy for a quick job, the
syntax is not satisfactory: you have to use different macros for
different arities, and it feels more like Lisp than C++. The
`::testing::AssertionResult` class solves this problem.
An `AssertionResult` object represents the result of an assertion
(whether it's a success or a failure, and an associated message). You
can create an `AssertionResult` using one of these factory
functions:
```
namespace testing {
// Returns an AssertionResult object to indicate that an assertion has
// succeeded.
AssertionResult AssertionSuccess();
// Returns an AssertionResult object to indicate that an assertion has
// failed.
AssertionResult AssertionFailure();
}
```
You can then use the `<<` operator to stream messages to the
`AssertionResult` object.
To provide more readable messages in Boolean assertions
(e.g. `EXPECT_TRUE()`), write a predicate function that returns
`AssertionResult` instead of `bool`. For example, if you define
`IsEven()` as:
```
::testing::AssertionResult IsEven(int n) {
if ((n % 2) == 0)
return ::testing::AssertionSuccess();
else
return ::testing::AssertionFailure() << n << " is odd";
}
```
instead of:
```
bool IsEven(int n) {
return (n % 2) == 0;
}
```
the failed assertion `EXPECT_TRUE(IsEven(Fib(4)))` will print:
<pre>
Value of: IsEven(Fib(4))<br>
Actual: false (*3 is odd*)<br>
Expected: true<br>
</pre>
instead of a more opaque
<pre>
Value of: IsEven(Fib(4))<br>
Actual: false<br>
Expected: true<br>
</pre>
If you want informative messages in `EXPECT_FALSE` and `ASSERT_FALSE`
as well, and are fine with making the predicate slower in the success
case, you can supply a success message:
```
::testing::AssertionResult IsEven(int n) {
if ((n % 2) == 0)
return ::testing::AssertionSuccess() << n << " is even";
else
return ::testing::AssertionFailure() << n << " is odd";
}
```
Then the statement `EXPECT_FALSE(IsEven(Fib(6)))` will print
<pre>
Value of: IsEven(Fib(6))<br>
Actual: true (8 is even)<br>
Expected: false<br>
</pre>
_Availability_: Linux, Windows, Mac; since version 1.4.1.
### Using a Predicate-Formatter ###
If you find the default message generated by `(ASSERT|EXPECT)_PRED*` and
`(ASSERT|EXPECT)_(TRUE|FALSE)` unsatisfactory, or some arguments to your
predicate do not support streaming to `ostream`, you can instead use the
following _predicate-formatter assertions_ to _fully_ customize how the
| `ASSERT_NEAR(`_val1, val2, abs\_error_`);` | `EXPECT_NEAR`_(val1, val2, abs\_error_`);` | the difference between _val1_ and _val2_ doesn't exceed the given absolute error |
_Availability_: Linux, Windows, Mac.
### Floating-Point Predicate-Format Functions ###
Some floating-point operations are useful, but not that often used. In order
to avoid an explosion of new macros, we provide them as predicate-format
functions that can be used in predicate assertion macros (e.g.
| `ASSERT_DEATH(`_statement, regex_`); | `EXPECT_DEATH(`_statement, regex_`); | _statement_ crashes with the given error |
| `ASSERT_DEATH_IF_SUPPORTED(`_statement, regex_`); | `EXPECT_DEATH_IF_SUPPORTED(`_statement, regex_`); | if death tests are supported, verifies that _statement_ crashes with the given error; otherwise verifies nothing |
| `ASSERT_EXIT(`_statement, predicate, regex_`); | `EXPECT_EXIT(`_statement, predicate, regex_`); |_statement_ exits with the given error and its exit code matches _predicate_ |
where _statement_ is a statement that is expected to cause the process to
die, _predicate_ is a function or function object that evaluates an integer
exit status, and _regex_ is a regular expression that the stderr output of
_statement_ is expected to match. Note that _statement_ can be _any valid
statement_ (including _compound statement_) and doesn't have to be an
expression.
As usual, the `ASSERT` variants abort the current test function, while the
`EXPECT` variants do not.
**Note:** We use the word "crash" here to mean that the process
terminates with a _non-zero_ exit status code. There are two
possibilities: either the process has called `exit()` or `_exit()`
with a non-zero value, or it may be killed by a signal.
This means that if _statement_ terminates the process with a 0 exit
code, it is _not_ considered a crash by `EXPECT_DEATH`. Use
`EXPECT_EXIT` instead if this is the case, or if you want to restrict
the exit code more precisely.
A predicate here must accept an `int` and return a `bool`. The death test
succeeds only if the predicate returns `true`. Google Test defines a few
predicates that handle the most common cases:
```
::testing::ExitedWithCode(exit_code)
```
This expression is `true` if the program exited normally with the given exit
code.
```
::testing::KilledBySignal(signal_number) // Not available on Windows.
```
This expression is `true` if the program was killed by the given signal.
The `*_DEATH` macros are convenient wrappers for `*_EXIT` that use a predicate
that verifies the process' exit code is non-zero.
Note that a death test only cares about three things:
1. does _statement_ abort or exit the process?
1. (in the case of `ASSERT_EXIT` and `EXPECT_EXIT`) does the exit status satisfy _predicate_? Or (in the case of `ASSERT_DEATH` and `EXPECT_DEATH`) is the exit status non-zero? And
1. does the stderr output match _regex_?
In particular, if _statement_ generates an `ASSERT_*` or `EXPECT_*` failure, it will **not** cause the death test to fail, as Google Test assertions don't abort the process.
To write a death test, simply use one of the above macros inside your test
function. For example,
```
TEST(MyDeathTest, Foo) {
// This death test uses a compound statement.
ASSERT_DEATH({ int n = 5; Foo(&n); }, "Error on line .* of Foo()");
* calling `Foo(5)` causes the process to die with the given error message,
* calling `NormalExit()` causes the process to print `"Success"` to stderr and exit with exit code 0, and
* calling `KillMyself()` kills the process with signal `SIGKILL`.
The test function body may contain other assertions and statements as well, if
necessary.
_Important:_ We strongly recommend you to follow the convention of naming your
test case (not test) `*DeathTest` when it contains a death test, as
demonstrated in the above example. The `Death Tests And Threads` section below
explains why.
If a test fixture class is shared by normal tests and death tests, you
can use typedef to introduce an alias for the fixture class and avoid
duplicating its code:
```
class FooTest : public ::testing::Test { ... };
typedef FooTest FooDeathTest;
TEST_F(FooTest, DoesThis) {
// normal test
}
TEST_F(FooDeathTest, DoesThat) {
// death test
}
```
_Availability:_ Linux, Windows (requires MSVC 8.0 or above), Cygwin, and Mac (the latter three are supported since v1.3.0). `(ASSERT|EXPECT)_DEATH_IF_SUPPORTED` are new in v1.4.0.
## Regular Expression Syntax ##
On POSIX systems (e.g. Linux, Cygwin, and Mac), Google Test uses the
syntax in death tests. To learn about this syntax, you may want to read this [Wikipedia entry](http://en.wikipedia.org/wiki/Regular_expression#POSIX_Extended_Regular_Expressions).
On Windows, Google Test uses its own simple regular expression
implementation. It lacks many features you can find in POSIX extended
regular expressions. For example, we don't support union (`"x|y"`),
grouping (`"(xy)"`), brackets (`"[xy]"`), and repetition count
(`"x{5,7}"`), among others. Below is what we do support (Letter `A` denotes a
literal character, period (`.`), or a single `\\` escape sequence; `x`
and `y` denote regular expressions.):
| `c` | matches any literal character `c` |
|:----|:----------------------------------|
| `\\d` | matches any decimal digit |
| `\\D` | matches any character that's not a decimal digit |
| `\\f` | matches `\f` |
| `\\n` | matches `\n` |
| `\\r` | matches `\r` |
| `\\s` | matches any ASCII whitespace, including `\n` |
| `\\S` | matches any character that's not a whitespace |
| `\\t` | matches `\t` |
| `\\v` | matches `\v` |
| `\\w` | matches any letter, `_`, or decimal digit |
| `\\W` | matches any character that `\\w` doesn't match |
| `\\c` | matches any literal character `c`, which must be a punctuation |
| `\\.` | matches the `.` character |
| `.` | matches any single character except `\n` |
| `A?` | matches 0 or 1 occurrences of `A` |
| `A*` | matches 0 or many occurrences of `A` |
| `A+` | matches 1 or many occurrences of `A` |
| `^` | matches the beginning of a string (not that of each line) |
| `$` | matches the end of a string (not that of each line) |
| `xy` | matches `x` followed by `y` |
To help you determine which capability is available on your system,
Google Test defines macro `GTEST_USES_POSIX_RE=1` when it uses POSIX
extended regular expressions, or `GTEST_USES_SIMPLE_RE=1` when it uses
the simple version. If you want your death tests to work in both
cases, you can either `#if` on these macros or use the more limited
syntax only.
## How It Works ##
Under the hood, `ASSERT_EXIT()` spawns a new process and executes the
death test statement in that process. The details of how precisely
that happens depend on the platform and the variable
`::testing::GTEST_FLAG(death_test_style)` (which is initialized from the
command-line flag `--gtest_death_test_style`).
* On POSIX systems, `fork()` (or `clone()` on Linux) is used to spawn the child, after which:
* If the variable's value is `"fast"`, the death test statement is immediately executed.
* If the variable's value is `"threadsafe"`, the child process re-executes the unit test binary just as it was originally invoked, but with some extra flags to cause just the single death test under consideration to be run.
* On Windows, the child is spawned using the `CreateProcess()` API, and re-executes the binary to cause just the single death test under consideration to be run - much like the `threadsafe` mode on POSIX.
Other values for the variable are illegal and will cause the death test to
fail. Currently, the flag's default value is `"fast"`. However, we reserve the
right to change it in the future. Therefore, your tests should not depend on
this.
In either case, the parent process waits for the child process to complete, and checks that
1. the child's exit status satisfies the predicate, and
1. the child's stderr matches the regular expression.
If the death test statement runs to completion without dying, the child
process will nonetheless terminate, and the assertion fails.
## Death Tests And Threads ##
The reason for the two death test styles has to do with thread safety. Due to
well-known problems with forking in the presence of threads, death tests should
be run in a single-threaded context. Sometimes, however, it isn't feasible to
arrange that kind of environment. For example, statically-initialized modules
may start threads before main is ever reached. Once threads have been created,
it may be difficult or impossible to clean them up.
Google Test has three features intended to raise awareness of threading issues.
1. A warning is emitted if multiple threads are running when a death test is encountered.
1. Test cases with a name ending in "DeathTest" are run before all other tests.
1. It uses `clone()` instead of `fork()` to spawn the child process on Linux (`clone()` is not available on Cygwin and Mac), as `fork()` is more likely to cause the child to hang when the parent process has multiple threads.
It's perfectly fine to create threads inside a death test statement; they are
executed in a separate process and cannot affect the parent.
## Death Test Styles ##
The "threadsafe" death test style was introduced in order to help mitigate the
risks of testing in a possibly multithreaded environment. It trades increased
test execution time (potentially dramatically so) for improved thread safety.
We suggest using the faster, default "fast" style unless your test has specific
problems with it.
You can choose a particular style of death tests by setting the flag
The _statement_ argument of `ASSERT_EXIT()` can be any valid C++ statement.
If it leaves the current function via a `return` statement or by throwing an exception,
the death test is considered to have failed. Some Google Test macros may return
from the current function (e.g. `ASSERT_TRUE()`), so be sure to avoid them in _statement_.
Since _statement_ runs in the child process, any in-memory side effect (e.g.
modifying a variable, releasing memory, etc) it causes will _not_ be observable
in the parent process. In particular, if you release memory in a death test,
your program will fail the heap check as the parent process will never see the
memory reclaimed. To solve this problem, you can
1. try not to free memory in a death test;
1. free the memory again in the parent process; or
1. do not use the heap checker in your program.
Due to an implementation detail, you cannot place multiple death test
assertions on the same line; otherwise, compilation will fail with an unobvious
error message.
Despite the improved thread safety afforded by the "threadsafe" style of death
test, thread problems such as deadlock are still possible in the presence of
handlers registered with `pthread_atfork(3)`.
# Using Assertions in Sub-routines #
## Adding Traces to Assertions ##
If a test sub-routine is called from several places, when an assertion
inside it fails, it can be hard to tell which invocation of the
sub-routine the failure is from. You can alleviate this problem using
extra logging or custom failure messages, but that usually clutters up
your tests. A better solution is to use the `SCOPED_TRACE` macro:
| `SCOPED_TRACE(`_message_`);` |
|:-----------------------------|
where _message_ can be anything streamable to `std::ostream`. This
macro will cause the current file name, line number, and the given
message to be added in every failure message. The effect will be
undone when the control leaves the current lexical scope.
For example,
```
10: void Sub1(int n) {
11: EXPECT_EQ(1, Bar(n));
12: EXPECT_EQ(2, Bar(n + 1));
13: }
14:
15: TEST(FooTest, Bar) {
16: {
17: SCOPED_TRACE("A"); // This trace point will be included in
18: // every failure in this scope.
19: Sub1(1);
20: }
21: // Now it won't.
22: Sub1(9);
23: }
```
could result in messages like these:
```
path/to/foo_test.cc:11: Failure
Value of: Bar(n)
Expected: 1
Actual: 2
Trace:
path/to/foo_test.cc:17: A
path/to/foo_test.cc:12: Failure
Value of: Bar(n + 1)
Expected: 2
Actual: 3
```
Without the trace, it would've been difficult to know which invocation
of `Sub1()` the two failures come from respectively. (You could add an
extra message to each assertion in `Sub1()` to indicate the value of
`n`, but that's tedious.)
Some tips on using `SCOPED_TRACE`:
1. With a suitable message, it's often enough to use `SCOPED_TRACE` at the beginning of a sub-routine, instead of at each call site.
1. When calling sub-routines inside a loop, make the loop iterator part of the message in `SCOPED_TRACE` such that you can know which iteration the failure is from.
1. Sometimes the line number of the trace point is enough for identifying the particular invocation of a sub-routine. In this case, you don't have to choose a unique message for `SCOPED_TRACE`. You can simply use `""`.
1. You can use `SCOPED_TRACE` in an inner scope when there is one in the outer scope. In this case, all active trace points will be included in the failure messages, in reverse order they are encountered.
1. The trace dump is clickable in Emacs' compilation buffer - hit return on a line number and you'll be taken to that line in the source file!
_Availability:_ Linux, Windows, Mac.
## Propagating Fatal Failures ##
A common pitfall when using `ASSERT_*` and `FAIL*` is not understanding that
when they fail they only abort the _current function_, not the entire test. For
example, the following test will segfault:
```
void Subroutine() {
// Generates a fatal failure and aborts the current function.
ASSERT_EQ(1, 2);
// The following won't be executed.
...
}
TEST(FooTest, Bar) {
Subroutine();
// The intended behavior is for the fatal failure
// in Subroutine() to abort the entire test.
// The actual behavior: the function goes on after Subroutine() returns.
int* p = NULL;
*p = 3; // Segfault!
}
```
Since we don't use exceptions, it is technically impossible to
implement the intended behavior here. To alleviate this, Google Test
provides two solutions. You could use either the
`(ASSERT|EXPECT)_NO_FATAL_FAILURE` assertions or the
`HasFatalFailure()` function. They are described in the following two
subsections.
### Asserting on Subroutines ###
As shown above, if your test calls a subroutine that has an `ASSERT_*`
failure in it, the test will continue after the subroutine
returns. This may not be what you want.
Often people want fatal failures to propagate like exceptions. For
| `ASSERT_NO_FATAL_FAILURE(`_statement_`);` | `EXPECT_NO_FATAL_FAILURE(`_statement_`);` | _statement_ doesn't generate any new fatal failures in the current thread. |
Only failures in the thread that executes the assertion are checked to
determine the result of this type of assertions. If _statement_
creates new threads, failures in these threads are ignored.
Examples:
```
ASSERT_NO_FATAL_FAILURE(Foo());
int i;
EXPECT_NO_FATAL_FAILURE({
i = Bar();
});
```
_Availability:_ Linux, Windows, Mac. Assertions from multiple threads
are currently not supported.
### Checking for Failures in the Current Test ###
`HasFatalFailure()` in the `::testing::Test` class returns `true` if an
assertion in the current test has suffered a fatal failure. This
allows functions to catch fatal failures in a sub-routine and return
early.
```
class Test {
public:
...
static bool HasFatalFailure();
};
```
The typical usage, which basically simulates the behavior of a thrown
exception, is:
```
TEST(FooTest, Bar) {
Subroutine();
// Aborts if Subroutine() had a fatal failure.
if (HasFatalFailure())
return;
// The following won't be executed.
...
}
```
If `HasFatalFailure()` is used outside of `TEST()` , `TEST_F()` , or a test
fixture, you must add the `::testing::Test::` prefix, as in:
```
if (::testing::Test::HasFatalFailure())
return;
```
Similarly, `HasNonfatalFailure()` returns `true` if the current test
has at least one non-fatal failure, and `HasFailure()` returns `true`
if the current test has at least one failure of either kind.
_Availability:_ Linux, Windows, Mac. `HasNonfatalFailure()` and
`HasFailure()` are available since version 1.4.0.
# Logging Additional Information #
In your test code, you can call `RecordProperty("key", value)` to log
additional information, where `value` can be either a string or an `int`. The _last_ value recorded for a key will be emitted to the XML output
* `RecordProperty()` is a static member of the `Test` class. Therefore it needs to be prefixed with `::testing::Test::` if used outside of the `TEST` body and the test fixture class.
* `key` must be a valid XML attribute name, and cannot conflict with the ones already used by Google Test (`name`, `status`, `time`, `classname`, `type_param`, and `value_param`).
* Calling `RecordProperty()` outside of the lifespan of a test is allowed. If it's called outside of a test but between a test case's `SetUpTestCase()` and `TearDownTestCase()` methods, it will be attributed to the XML element for the test case. If it's called outside of all test cases (e.g. in a test environment), it will be attributed to the top-level XML element.
_Availability_: Linux, Windows, Mac.
# Sharing Resources Between Tests in the Same Test Case #
Google Test creates a new test fixture object for each test in order to make
tests independent and easier to debug. However, sometimes tests use resources
that are expensive to set up, making the one-copy-per-test model prohibitively
expensive.
If the tests don't change the resource, there's no harm in them sharing a
single resource copy. So, in addition to per-test set-up/tear-down, Google Test
also supports per-test-case set-up/tear-down. To use it:
1. In your test fixture class (say `FooTest` ), define as `static` some member variables to hold the shared resources.
1. In the same test fixture class, define a `static void SetUpTestCase()` function (remember not to spell it as **`SetupTestCase`** with a small `u`!) to set up the shared resources and a `static void TearDownTestCase()` function to tear them down.
That's it! Google Test automatically calls `SetUpTestCase()` before running the
_first test_ in the `FooTest` test case (i.e. before creating the first
`FooTest` object), and calls `TearDownTestCase()` after running the _last test_
in it (i.e. after deleting the last `FooTest` object). In between, the tests
can use the shared resources.
Remember that the test order is undefined, so your code can't depend on a test
preceding or following another. Also, the tests must either not modify the
state of any shared resource, or, if they do modify the state, they must
restore the state to its original value before passing control to the next
test.
Here's an example of per-test-case set-up and tear-down:
```
class FooTest : public ::testing::Test {
protected:
// Per-test-case set-up.
// Called before the first test in this test case.
// Can be omitted if not needed.
static void SetUpTestCase() {
shared_resource_ = new ...;
}
// Per-test-case tear-down.
// Called after the last test in this test case.
// Can be omitted if not needed.
static void TearDownTestCase() {
delete shared_resource_;
shared_resource_ = NULL;
}
// You can define per-test set-up and tear-down logic as usual.
virtual void SetUp() { ... }
virtual void TearDown() { ... }
// Some expensive resource shared by all tests.
static T* shared_resource_;
};
T* FooTest::shared_resource_ = NULL;
TEST_F(FooTest, Test1) {
... you can refer to shared_resource here ...
}
TEST_F(FooTest, Test2) {
... you can refer to shared_resource here ...
}
```
_Availability:_ Linux, Windows, Mac.
# Global Set-Up and Tear-Down #
Just as you can do set-up and tear-down at the test level and the test case
level, you can also do it at the test program level. Here's how.
First, you subclass the `::testing::Environment` class to define a test
environment, which knows how to set-up and tear-down:
```
class Environment {
public:
virtual ~Environment() {}
// Override this to define how to set up the environment.
virtual void SetUp() {}
// Override this to define how to tear down the environment.
virtual void TearDown() {}
};
```
Then, you register an instance of your environment class with Google Test by
calling the `::testing::AddGlobalTestEnvironment()` function:
However, we strongly recommend you to write your own `main()` and call
`AddGlobalTestEnvironment()` there, as relying on initialization of global
variables makes the code harder to read and may cause problems when you
register multiple environments from different translation units and the
environments have dependencies among them (remember that the compiler doesn't
guarantee the order in which global variables from different translation units
are initialized).
_Availability:_ Linux, Windows, Mac.
# Value Parameterized Tests #
_Value-parameterized tests_ allow you to test your code with different
parameters without writing multiple copies of the same test.
Suppose you write a test for your code and then realize that your code is affected by a presence of a Boolean command line flag.
```
TEST(MyCodeTest, TestFoo) {
// A code to test foo().
}
```
Usually people factor their test code into a function with a Boolean parameter in such situations. The function sets the flag, then executes the testing code.
```
void TestFooHelper(bool flag_value) {
flag = flag_value;
// A code to test foo().
}
TEST(MyCodeTest, TestFoo) {
TestFooHelper(false);
TestFooHelper(true);
}
```
But this setup has serious drawbacks. First, when a test assertion fails in your tests, it becomes unclear what value of the parameter caused it to fail. You can stream a clarifying message into your `EXPECT`/`ASSERT` statements, but it you'll have to do it with all of them. Second, you have to add one such helper function per test. What if you have ten tests? Twenty? A hundred?
Value-parameterized tests will let you write your test only once and then easily instantiate and run it with an arbitrary number of parameter values.
Here are some other situations when value-parameterized tests come handy:
* You want to test different implementations of an OO interface.
* You want to test your code over various inputs (a.k.a. data-driven testing). This feature is easy to abuse, so please exercise your good sense when doing it!
## How to Write Value-Parameterized Tests ##
To write value-parameterized tests, first you should define a fixture
class. It must be derived from both `::testing::Test` and
`::testing::WithParamInterface<T>` (the latter is a pure interface),
where `T` is the type of your parameter values. For convenience, you
can just derive the fixture class from `::testing::TestWithParam<T>`,
which itself is derived from both `::testing::Test` and
`::testing::WithParamInterface<T>`. `T` can be any copyable type. If
it's a raw pointer, you are responsible for managing the lifespan of
the pointed values.
```
class FooTest : public ::testing::TestWithParam<const char*> {
// You can implement all the usual fixture class members here.
// To access the test parameter, call GetParam() from class
// TestWithParam<T>.
};
// Or, when you want to add parameters to a pre-existing fixture class:
class BaseTest : public ::testing::Test {
...
};
class BarTest : public BaseTest,
public ::testing::WithParamInterface<const char*> {
...
};
```
Then, use the `TEST_P` macro to define as many test patterns using
this fixture as you want. The `_P` suffix is for "parameterized" or
"pattern", whichever you prefer to think.
```
TEST_P(FooTest, DoesBlah) {
// Inside a test, access the test parameter with the GetParam() method
// of the TestWithParam<T> class:
EXPECT_TRUE(foo.Blah(GetParam()));
...
}
TEST_P(FooTest, HasBlahBlah) {
...
}
```
Finally, you can use `INSTANTIATE_TEST_CASE_P` to instantiate the test
case with any set of parameters you want. Google Test defines a number of
functions for generating test parameters. They return what we call
(surprise!) _parameter generators_. Here is a summary of them,
which are all in the `testing` namespace:
| `Range(begin, end[, step])` | Yields values `{begin, begin+step, begin+step+step, ...}`. The values do not include `end`. `step` defaults to 1. |
| `ValuesIn(container)` and `ValuesIn(begin, end)` | Yields values from a C-style array, an STL-style container, or an iterator range `[begin, end)`. `container`, `begin`, and `end` can be expressions whose values are determined at run time. |
| `Bool()` | Yields sequence `{false, true}`. |
| `Combine(g1, g2, ..., gN)` | Yields all combinations (the Cartesian product for the math savvy) of the values generated by the `N` generators. This is only available if your system provides the `<tr1/tuple>` header. If you are sure your system does, and Google Test disagrees, you can override it by defining `GTEST_HAS_TR1_TUPLE=1`. See comments in [include/gtest/internal/gtest-port.h](../include/gtest/internal/gtest-port.h) for more information. |
For more details, see the comments at the definitions of these functions in the [source code](../include/gtest/gtest-param-test.h).
The following statement will instantiate tests from the `FooTest` test case
each with parameter values `"meeny"`, `"miny"`, and `"moe"`.
```
INSTANTIATE_TEST_CASE_P(InstantiationName,
FooTest,
::testing::Values("meeny", "miny", "moe"));
```
To distinguish different instances of the pattern (yes, you can
instantiate it more than once), the first argument to
`INSTANTIATE_TEST_CASE_P` is a prefix that will be added to the actual
test case name. Remember to pick unique prefixes for different
instantiations. The tests from the instantiation above will have these
names:
*`InstantiationName/FooTest.DoesBlah/0` for `"meeny"`
*`InstantiationName/FooTest.DoesBlah/1` for `"miny"`
*`InstantiationName/FooTest.DoesBlah/2` for `"moe"`
*`InstantiationName/FooTest.HasBlahBlah/0` for `"meeny"`
*`InstantiationName/FooTest.HasBlahBlah/1` for `"miny"`
*`InstantiationName/FooTest.HasBlahBlah/2` for `"moe"`
You can use these names in [--gtest\_filter](#running-a-subset-of-the-tests).
This statement will instantiate all tests from `FooTest` again, each
The tests from the instantiation above will have these names:
*`AnotherInstantiationName/FooTest.DoesBlah/0` for `"cat"`
*`AnotherInstantiationName/FooTest.DoesBlah/1` for `"dog"`
*`AnotherInstantiationName/FooTest.HasBlahBlah/0` for `"cat"`
*`AnotherInstantiationName/FooTest.HasBlahBlah/1` for `"dog"`
Please note that `INSTANTIATE_TEST_CASE_P` will instantiate _all_
tests in the given test case, whether their definitions come before or
_after_ the `INSTANTIATE_TEST_CASE_P` statement.
You can see
[these](../samples/sample7_unittest.cc)
[files](../samples/sample8_unittest.cc) for more examples.
_Availability_: Linux, Windows (requires MSVC 8.0 or above), Mac; since version 1.2.0.
## Creating Value-Parameterized Abstract Tests ##
In the above, we define and instantiate `FooTest` in the same source
file. Sometimes you may want to define value-parameterized tests in a
library and let other people instantiate them later. This pattern is
known as <i>abstract tests</i>. As an example of its application, when you
are designing an interface you can write a standard suite of abstract
tests (perhaps using a factory function as the test parameter) that
all implementations of the interface are expected to pass. When
someone implements the interface, he can instantiate your suite to get
all the interface-conformance tests for free.
To define abstract tests, you should organize your code like this:
1. Put the definition of the parameterized test fixture class (e.g. `FooTest`) in a header file, say `foo_param_test.h`. Think of this as _declaring_ your abstract tests.
1. Put the `TEST_P` definitions in `foo_param_test.cc`, which includes `foo_param_test.h`. Think of this as _implementing_ your abstract tests.
Once they are defined, you can instantiate them by including
`foo_param_test.h`, invoking `INSTANTIATE_TEST_CASE_P()`, and linking
with `foo_param_test.cc`. You can instantiate the same abstract test
case multiple times, possibly in different source files.
# Typed Tests #
Suppose you have multiple implementations of the same interface and
want to make sure that all of them satisfy some common requirements.
Or, you may have defined several types that are supposed to conform to
the same "concept" and you want to verify it. In both cases, you want
the same test logic repeated for different types.
While you can write one `TEST` or `TEST_F` for each type you want to
test (and you may even factor the test logic into a function template
that you invoke from the `TEST`), it's tedious and doesn't scale:
if you want _m_ tests over _n_ types, you'll end up writing _m\*n_
`TEST`s.
_Typed tests_ allow you to repeat the same test logic over a list of
types. You only need to write the test logic once, although you must
know the type list when writing typed tests. Here's how you do it:
First, define a fixture class template. It should be parameterized
by a type. Remember to derive it from `::testing::Test`:
```
template <typename T>
class FooTest : public ::testing::Test {
public:
...
typedef std::list<T> List;
static T shared_;
T value_;
};
```
Next, associate a list of types with the test case, which will be
| `$ foo_test --gtest_repeat=-1` | A negative count means repeating forever. |
| `$ foo_test --gtest_repeat=1000 --gtest_break_on_failure` | Repeat foo\_test 1000 times, stopping at the first failure. This is especially useful when running under a debugger: when the testfails, it will drop into the debugger and you can then inspect variables and stacks. |
| `$ foo_test --gtest_repeat=1000 --gtest_filter=FooBar` | Repeat the tests whose name matches the filter 1000 times. |
If your test program contains global set-up/tear-down code registered
using `AddGlobalTestEnvironment()`, it will be repeated in each
iteration as well, as the flakiness may be in it. You can also specify
the repeat count by setting the `GTEST_REPEAT` environment variable.
_Availability:_ Linux, Windows, Mac.
## Shuffling the Tests ##
You can specify the `--gtest_shuffle` flag (or set the `GTEST_SHUFFLE`
environment variable to `1`) to run the tests in a program in a random
order. This helps to reveal bad dependencies between tests.
By default, Google Test uses a random seed calculated from the current
time. Therefore you'll get a different order every time. The console
output includes the random seed value, such that you can reproduce an
order-related test failure later. To specify the random seed
explicitly, use the `--gtest_random_seed=SEED` flag (or set the
`GTEST_RANDOM_SEED` environment variable), where `SEED` is an integer
between 0 and 99999. The seed value 0 is special: it tells Google Test
to do the default behavior of calculating the seed from the current
time.
If you combine this with `--gtest_repeat=N`, Google Test will pick a
different random seed and re-shuffle the tests in each iteration.
_Availability:_ Linux, Windows, Mac; since v1.4.0.
## Controlling Test Output ##
This section teaches how to tweak the way test results are reported.
### Colored Terminal Output ###
Google Test can use colors in its terminal output to make it easier to spot
the separation between tests, and whether tests passed.
You can set the GTEST\_COLOR environment variable or set the `--gtest_color`
command line flag to `yes`, `no`, or `auto` (the default) to enable colors,
disable colors, or let Google Test decide. When the value is `auto`, Google
Test will use colors if and only if the output goes to a terminal and (on
non-Windows platforms) the `TERM` environment variable is set to `xterm` or
`xterm-color`.
_Availability:_ Linux, Windows, Mac.
### Suppressing the Elapsed Time ###
By default, Google Test prints the time it takes to run each test. To
suppress that, run the test program with the `--gtest_print_time=0`
command line flag. Setting the `GTEST_PRINT_TIME` environment
variable to `0` has the same effect.
_Availability:_ Linux, Windows, Mac. (In Google Test 1.3.0 and lower,
the default behavior is that the elapsed time is **not** printed.)
### Generating an XML Report ###
Google Test can emit a detailed XML report to a file in addition to its normal
textual output. The report contains the duration of each test, and thus can
help you identify slow tests.
To generate the XML report, set the `GTEST_OUTPUT` environment variable or the
`--gtest_output` flag to the string `"xml:_path_to_output_file_"`, which will
create the file at the given location. You can also just use the string
`"xml"`, in which case the output can be found in the `test_detail.xml` file in
the current directory.
If you specify a directory (for example, `"xml:output/directory/"` on Linux or
`"xml:output\directory\"` on Windows), Google Test will create the XML file in
that directory, named after the test executable (e.g. `foo_test.xml` for test
program `foo_test` or `foo_test.exe`). If the file already exists (perhaps left
over from a previous run), Google Test will pick a different name (e.g.
`foo_test_1.xml`) to avoid overwriting it.
The report uses the format described here. It is based on the
`junitreport` Ant task and can be parsed by popular continuous build
systems like [Jenkins](http://jenkins-ci.org/). Since that format
was originally intended for Java, a little interpretation is required
to make it apply to Google Test tests, as shown here:
```
<testsuites name="AllTests" ...>
<testsuite name="test_case_name" ...>
<testcase name="test_name" ...>
<failure message="..."/>
<failure message="..."/>
<failure message="..."/>
</testcase>
</testsuite>
</testsuites>
```
* The root `<testsuites>` element corresponds to the entire test program.
*`<testsuite>` elements correspond to Google Test test cases.
*`<testcase>` elements correspond to Google Test test functions.
* The `tests` attribute of a `<testsuites>` or `<testsuite>` element tells how many test functions the Google Test program or test case contains, while the `failures` attribute tells how many of them failed.
* The `time` attribute expresses the duration of the test, test case, or entire test program in milliseconds.
* Each `<failure>` element corresponds to a single failed Google Test assertion.
* Some JUnit concepts don't apply to Google Test, yet we have to conform to the DTD. Therefore you'll see some dummy elements and attributes in the report. You can safely ignore these parts.
_Availability:_ Linux, Windows, Mac.
## Controlling How Failures Are Reported ##
### Turning Assertion Failures into Break-Points ###
When running test programs under a debugger, it's very convenient if the
debugger can catch an assertion failure and automatically drop into interactive
mode. Google Test's _break-on-failure_ mode supports this behavior.
To enable it, set the `GTEST_BREAK_ON_FAILURE` environment variable to a value
other than `0` . Alternatively, you can use the `--gtest_break_on_failure`
command line flag.
_Availability:_ Linux, Windows, Mac.
### Disabling Catching Test-Thrown Exceptions ###
Google Test can be used either with or without exceptions enabled. If
a test throws a C++ exception or (on Windows) a structured exception
(SEH), by default Google Test catches it, reports it as a test
failure, and continues with the next test method. This maximizes the
coverage of a test run. Also, on Windows an uncaught exception will
cause a pop-up window, so catching the exceptions allows you to run
the tests automatically.
When debugging the test failures, however, you may instead want the
exceptions to be handled by the debugger, such that you can examine
the call stack when an exception is thrown. To achieve that, set the
`GTEST_CATCH_EXCEPTIONS` environment variable to `0`, or use the
`--gtest_catch_exceptions=0` flag when running the tests.
**Availability**: Linux, Windows, Mac.
### Letting Another Testing Framework Drive ###
If you work on a project that has already been using another testing
framework and is not ready to completely switch to Google Test yet,
you can get much of Google Test's benefit by using its assertions in
your existing tests. Just change your `main()` function to look
like:
```
#include "gtest/gtest.h"
int main(int argc, char** argv) {
::testing::GTEST_FLAG(throw_on_failure) = true;
// Important: Google Test must be initialized.
::testing::InitGoogleTest(&argc, argv);
... whatever your existing testing framework requires ...
}
```
With that, you can use Google Test assertions in addition to the
native assertions your testing framework provides, for example:
```
void TestFooDoesBar() {
Foo foo;
EXPECT_LE(foo.Bar(1), 100); // A Google Test assertion.
CPPUNIT_ASSERT(foo.IsEmpty()); // A native assertion.
}
```
If a Google Test assertion fails, it will print an error message and
throw an exception, which will be treated as a failure by your host
testing framework. If you compile your code with exceptions disabled,
a failed Google Test assertion will instead exit your program with a
non-zero code, which will also signal a test failure to your test
runner.
If you don't write `::testing::GTEST_FLAG(throw_on_failure) = true;` in
your `main()`, you can alternatively enable this feature by specifying
the `--gtest_throw_on_failure` flag on the command-line or setting the
`GTEST_THROW_ON_FAILURE` environment variable to a non-zero value.
Death tests are _not_ supported when other test framework is used to organize tests.
_Availability:_ Linux, Windows, Mac; since v1.3.0.
## Distributing Test Functions to Multiple Machines ##
If you have more than one machine you can use to run a test program,
you might want to run the test functions in parallel and get the
result faster. We call this technique _sharding_, where each machine
is called a _shard_.
Google Test is compatible with test sharding. To take advantage of
this feature, your test runner (not part of Google Test) needs to do
the following:
1. Allocate a number of machines (shards) to run the tests.
1. On each shard, set the `GTEST_TOTAL_SHARDS` environment variable to the total number of shards. It must be the same for all shards.
1. On each shard, set the `GTEST_SHARD_INDEX` environment variable to the index of the shard. Different shards must be assigned different indices, which must be in the range `[0, GTEST_TOTAL_SHARDS - 1]`.
1. Run the same test program on all shards. When Google Test sees the above two environment variables, it will select a subset of the test functions to run. Across all shards, each test function in the program will be run exactly once.
1. Wait for all shards to finish, then collect and report the results.
Your project may have tests that were written without Google Test and
thus don't understand this protocol. In order for your test runner to
figure out which test supports sharding, it can set the environment
variable `GTEST_SHARD_STATUS_FILE` to a non-existent file path. If a
test program supports sharding, it will create this file to
acknowledge the fact (the actual contents of the file are not
important at this time; although we may stick some useful information
in it in the future.); otherwise it will not create it.
Here's an example to make it clear. Suppose you have a test program
`foo_test` that contains the following 5 test functions:
```
TEST(A, V)
TEST(A, W)
TEST(B, X)
TEST(B, Y)
TEST(B, Z)
```
and you have 3 machines at your disposal. To run the test functions in
parallel, you would set `GTEST_TOTAL_SHARDS` to 3 on all machines, and
set `GTEST_SHARD_INDEX` to 0, 1, and 2 on the machines respectively.
Then you would run the same `foo_test` on each machine.
Google Test reserves the right to change how the work is distributed
across the shards, but here's one possible scenario:
* Machine #0 runs `A.V` and `B.X`.
* Machine #1 runs `A.W` and `B.Y`.
* Machine #2 runs `B.Z`.
_Availability:_ Linux, Windows, Mac; since version 1.3.0.
# Fusing Google Test Source Files #
Google Test's implementation consists of ~30 files (excluding its own
tests). Sometimes you may want them to be packaged up in two files (a
`.h` and a `.cc`) instead, such that you can easily copy them to a new
machine and start hacking there. For this we provide an experimental
Python script `fuse_gtest_files.py` in the `scripts/` directory (since release 1.3.0).
Assuming you have Python 2.4 or above installed on your machine, just
go to that directory and run
```
python fuse_gtest_files.py OUTPUT_DIR
```
and you should see an `OUTPUT_DIR` directory being created with files
`gtest/gtest.h` and `gtest/gtest-all.cc` in it. These files contain
everything you need to use Google Test. Just copy them to anywhere
you want and you are ready to write tests. You can use the
[scripts/test/Makefile](../scripts/test/Makefile)
file as an example on how to compile your tests against them.
# Where to Go from Here #
Congratulations! You've now learned more advanced Google Test tools and are
ready to tackle more complex testing tasks. If you want to dive even deeper, you
can read the [Frequently-Asked Questions](V1_7_FAQ.md).
If you cannot find the answer to your question here, and you have read
[Primer](V1_7_Primer.md) and [AdvancedGuide](V1_7_AdvancedGuide.md), send it to
googletestframework@googlegroups.com.
## Why should I use Google Test instead of my favorite C++ testing framework? ##
First, let us say clearly that we don't want to get into the debate of
which C++ testing framework is **the best**. There exist many fine
frameworks for writing C++ tests, and we have tremendous respect for
the developers and users of them. We don't think there is (or will
be) a single best framework - you have to pick the right tool for the
particular task you are tackling.
We created Google Test because we couldn't find the right combination
of features and conveniences in an existing framework to satisfy _our_
needs. The following is a list of things that _we_ like about Google
Test. We don't claim them to be unique to Google Test - rather, the
combination of them makes Google Test the choice for us. We hope this
list can help you decide whether it is for you too.
* Google Test is designed to be portable: it doesn't require exceptions or RTTI; it works around various bugs in various compilers and environments; etc. As a result, it works on Linux, Mac OS X, Windows and several embedded operating systems.
* Nonfatal assertions (`EXPECT_*`) have proven to be great time savers, as they allow a test to report multiple failures in a single edit-compile-test cycle.
* It's easy to write assertions that generate informative messages: you just use the stream syntax to append any additional information, e.g. `ASSERT_EQ(5, Foo(i)) << " where i = " << i;`. It doesn't require a new set of macros or special functions.
* Google Test automatically detects your tests and doesn't require you to enumerate them in order to run them.
* Death tests are pretty handy for ensuring that your asserts in production code are triggered by the right conditions.
*`SCOPED_TRACE` helps you understand the context of an assertion failure when it comes from inside a sub-routine or loop.
* You can decide which tests to run using name patterns. This saves time when you want to quickly reproduce a test failure.
* Google Test can generate XML test result reports that can be parsed by popular continuous build system like Hudson.
* Simple things are easy in Google Test, while hard things are possible: in addition to advanced features like [global test environments](V1_7_AdvancedGuide.md#global-set-up-and-tear-down) and tests parameterized by [values](V1_7_AdvancedGuide.md#value-parameterized-tests) or [types](V1_7_AdvancedGuide.md#typed-tests), Google Test supports various ways for the user to extend the framework -- if Google Test doesn't do something out of the box, chances are that a user can implement the feature using Google Test's public API, without changing Google Test itself. In particular, you can:
* expand your testing vocabulary by defining [custom predicates](V1_7_AdvancedGuide.md#predicate-assertions-for-better-error-messages),
* teach Google Test how to [print your types](V1_7_AdvancedGuide.md#teaching-google-test-how-to-print-your-values),
* define your own testing macros or utilities and verify them using Google Test's [Service Provider Interface](V1_7_AdvancedGuide.md#catching-failures), and
* reflect on the test cases or change the test output format by intercepting the [test events](V1_7_AdvancedGuide.md#extending-google-test-by-handling-test-events).
## I'm getting warnings when compiling Google Test. Would you fix them? ##
We strive to minimize compiler warnings Google Test generates. Before releasing a new version, we test to make sure that it doesn't generate warnings when compiled using its CMake script on Windows, Linux, and Mac OS.
Unfortunately, this doesn't mean you are guaranteed to see no warnings when compiling Google Test in your environment:
* You may be using a different compiler as we use, or a different version of the same compiler. We cannot possibly test for all compilers.
* You may be compiling on a different platform as we do.
* Your project may be using different compiler flags as we do.
It is not always possible to make Google Test warning-free for everyone. Or, it may not be desirable if the warning is rarely enabled and fixing the violations makes the code more complex.
If you see warnings when compiling Google Test, we suggest that you use the `-isystem` flag (assuming your are using GCC) to mark Google Test headers as system headers. That'll suppress warnings from Google Test headers.
## Why should not test case names and test names contain underscore? ##
Underscore (`_`) is special, as C++ reserves the following to be used by
the compiler and the standard library:
1. any identifier that starts with an `_` followed by an upper-case letter, and
1. any identifier that containers two consecutive underscores (i.e. `__`) _anywhere_ in its name.
User code is _prohibited_ from using such identifiers.
Now let's look at what this means for `TEST` and `TEST_F`.
Currently `TEST(TestCaseName, TestName)` generates a class named
`TestCaseName_TestName_Test`. What happens if `TestCaseName` or `TestName`
contains `_`?
1. If `TestCaseName` starts with an `_` followed by an upper-case letter (say, `_Foo`), we end up with `_Foo_TestName_Test`, which is reserved and thus invalid.
1. If `TestCaseName` ends with an `_` (say, `Foo_`), we get `Foo__TestName_Test`, which is invalid.
1. If `TestName` starts with an `_` (say, `_Bar`), we get `TestCaseName__Bar_Test`, which is invalid.
1. If `TestName` ends with an `_` (say, `Bar_`), we get `TestCaseName_Bar__Test`, which is invalid.
So clearly `TestCaseName` and `TestName` cannot start or end with `_`
(Actually, `TestCaseName` can start with `_` -- as long as the `_` isn't
followed by an upper-case letter. But that's getting complicated. So
for simplicity we just say that it cannot start with `_`.).
It may seem fine for `TestCaseName` and `TestName` to contain `_` in the
middle. However, consider this:
```
TEST(Time, Flies_Like_An_Arrow) { ... }
TEST(Time_Flies, Like_An_Arrow) { ... }
```
Now, the two `TEST`s will both generate the same class
(`Time_Files_Like_An_Arrow_Test`). That's not good.
So for simplicity, we just ask the users to avoid `_` in `TestCaseName`
and `TestName`. The rule is more constraining than necessary, but it's
simple and easy to remember. It also gives Google Test some wiggle
room in case its implementation needs to change in the future.
If you violate the rule, there may not be immediately consequences,
but your test may (just may) break with a new compiler (or a new
version of the compiler you are using) or with a new version of Google
Test. Therefore it's best to follow the rule.
## Why is it not recommended to install a pre-compiled copy of Google Test (for example, into /usr/local)? ##
In the early days, we said that you could install
compiled Google Test libraries on `*`nix systems using `make install`.
Then every user of your machine can write tests without
recompiling Google Test.
This seemed like a good idea, but it has a
got-cha: every user needs to compile his tests using the _same_ compiler
flags used to compile the installed Google Test libraries; otherwise
he may run into undefined behaviors (i.e. the tests can behave
strangely and may even crash for no obvious reasons).
Why? Because C++ has this thing called the One-Definition Rule: if
two C++ source files contain different definitions of the same
class/function/variable, and you link them together, you violate the
rule. The linker may or may not catch the error (in many cases it's
not required by the C++ standard to catch the violation). If it
doesn't, you get strange run-time behaviors that are unexpected and
hard to debug.
If you compile Google Test and your test code using different compiler
flags, they may see different definitions of the same
class/function/variable (e.g. due to the use of `#if` in Google Test).
Therefore, for your sanity, we recommend to avoid installing pre-compiled
Google Test libraries. Instead, each project should compile
Google Test itself such that it can be sure that the same flags are
used for both Google Test and the tests.
## How do I generate 64-bit binaries on Windows (using Visual Studio 2008)? ##
(Answered by Trevor Robinson)
Load the supplied Visual Studio solution file, either `msvc\gtest-md.sln` or
`msvc\gtest.sln`. Go through the migration wizard to migrate the
solution and project files to Visual Studio 2008. Select
`Configuration Manager...` from the `Build` menu. Select `<New...>` from
the `Active solution platform` dropdown. Select `x64` from the new
platform dropdown, leave `Copy settings from` set to `Win32` and
`Create new project platforms` checked, then click `OK`. You now have
`Win32` and `x64` platform configurations, selectable from the
`Standard` toolbar, which allow you to toggle between building 32-bit or
64-bit binaries (or both at once using Batch Build).
In order to prevent build output files from overwriting one another,
you'll need to change the `Intermediate Directory` settings for the
newly created platform configuration across all the projects. To do
this, multi-select (e.g. using shift-click) all projects (but not the
solution) in the `Solution Explorer`. Right-click one of them and
select `Properties`. In the left pane, select `Configuration Properties`,
and from the `Configuration` dropdown, select `All Configurations`.
Make sure the selected platform is `x64`. For the
`Intermediate Directory` setting, change the value from
`$(PlatformName)\$(ConfigurationName)` to
`$(OutDir)\$(ProjectName)`. Click `OK` and then build the
solution. When the build is complete, the 64-bit binaries will be in
the `msvc\x64\Debug` directory.
## Can I use Google Test on MinGW? ##
We haven't tested this ourselves, but Per Abrahamsen reported that he
was able to compile and install Google Test successfully when using
MinGW from Cygwin. You'll need to configure it with:
Please contact `googletestframework@googlegroups.com` if you are
interested in improving the support for MinGW.
## Why does Google Test support EXPECT\_EQ(NULL, ptr) and ASSERT\_EQ(NULL, ptr) but not EXPECT\_NE(NULL, ptr) and ASSERT\_NE(NULL, ptr)? ##
Due to some peculiarity of C++, it requires some non-trivial template
meta programming tricks to support using `NULL` as an argument of the
`EXPECT_XX()` and `ASSERT_XX()` macros. Therefore we only do it where
it's most needed (otherwise we make the implementation of Google Test
harder to maintain and more error-prone than necessary).
The `EXPECT_EQ()` macro takes the _expected_ value as its first
argument and the _actual_ value as the second. It's reasonable that
someone wants to write `EXPECT_EQ(NULL, some_expression)`, and this
indeed was requested several times. Therefore we implemented it.
The need for `EXPECT_NE(NULL, ptr)` isn't nearly as strong. When the
assertion fails, you already know that `ptr` must be `NULL`, so it
doesn't add any information to print ptr in this case. That means
`EXPECT_TRUE(ptr != NULL)` works just as well.
If we were to support `EXPECT_NE(NULL, ptr)`, for consistency we'll
have to support `EXPECT_NE(ptr, NULL)` as well, as unlike `EXPECT_EQ`,
we don't have a convention on the order of the two arguments for
`EXPECT_NE`. This means using the template meta programming tricks
twice in the implementation, making it even harder to understand and
maintain. We believe the benefit doesn't justify the cost.
Finally, with the growth of Google Mock's [matcher](../../CookBook.md#using-matchers-in-google-test-assertions) library, we are
encouraging people to use the unified `EXPECT_THAT(value, matcher)`
syntax more often in tests. One significant advantage of the matcher
approach is that matchers can be easily combined to form new matchers,
while the `EXPECT_NE`, etc, macros cannot be easily
combined. Therefore we want to invest more in the matchers than in the
`EXPECT_XX()` macros.
## Does Google Test support running tests in parallel? ##
Test runners tend to be tightly coupled with the build/test
environment, and Google Test doesn't try to solve the problem of
running tests in parallel. Instead, we tried to make Google Test work
nicely with test runners. For example, Google Test's XML report
contains the time spent on each test, and its `gtest_list_tests` and
`gtest_filter` flags can be used for splitting the execution of test
methods into multiple processes. These functionalities can help the
test runner run the tests in parallel.
## Why don't Google Test run the tests in different threads to speed things up? ##
It's difficult to write thread-safe code. Most tests are not written
with thread-safety in mind, and thus may not work correctly in a
multi-threaded setting.
If you think about it, it's already hard to make your code work when
you know what other threads are doing. It's much harder, and
sometimes even impossible, to make your code work when you don't know
what other threads are doing (remember that test methods can be added,
deleted, or modified after your test was written). If you want to run
the tests in parallel, you'd better run them in different processes.
## Why aren't Google Test assertions implemented using exceptions? ##
Our original motivation was to be able to use Google Test in projects
that disable exceptions. Later we realized some additional benefits
of this approach:
1. Throwing in a destructor is undefined behavior in C++. Not using exceptions means Google Test's assertions are safe to use in destructors.
1. The `EXPECT_*` family of macros will continue even after a failure, allowing multiple failures in a `TEST` to be reported in a single run. This is a popular feature, as in C++ the edit-compile-test cycle is usually quite long and being able to fixing more than one thing at a time is a blessing.
1. If assertions are implemented using exceptions, a test may falsely ignore a failure if it's caught by user code:
```
try { ... ASSERT_TRUE(...) ... }
catch (...) { ... }
```
The above code will pass even if the `ASSERT_TRUE` throws. While it's unlikely for someone to write this in a test, it's possible to run into this pattern when you write assertions in callbacks that are called by the code under test.
The downside of not using exceptions is that `ASSERT_*` (implemented
using `return`) will only abort the current function, not the current
`TEST`.
## Why do we use two different macros for tests with and without fixtures? ##
Unfortunately, C++'s macro system doesn't allow us to use the same
macro for both cases. One possibility is to provide only one macro
for tests with fixtures, and require the user to define an empty
fixture sometimes:
```
class FooTest : public ::testing::Test {};
TEST_F(FooTest, DoesThis) { ... }
```
or
```
typedef ::testing::Test FooTest;
TEST_F(FooTest, DoesThat) { ... }
```
Yet, many people think this is one line too many. :-) Our goal was to
make it really easy to write tests, so we tried to make simple tests
trivial to create. That means using a separate macro for such tests.
We think neither approach is ideal, yet either of them is reasonable.
In the end, it probably doesn't matter much either way.
## Why don't we use structs as test fixtures? ##
We like to use structs only when representing passive data. This
distinction between structs and classes is good for documenting the
intent of the code's author. Since test fixtures have logic like
`SetUp()` and `TearDown()`, they are better defined as classes.
## Why are death tests implemented as assertions instead of using a test runner? ##
Our goal was to make death tests as convenient for a user as C++
possibly allows. In particular:
* The runner-style requires to split the information into two pieces: the definition of the death test itself, and the specification for the runner on how to run the death test and what to expect. The death test would be written in C++, while the runner spec may or may not be. A user needs to carefully keep the two in sync. `ASSERT_DEATH(statement, expected_message)` specifies all necessary information in one place, in one language, without boilerplate code. It is very declarative.
*`ASSERT_DEATH` has a similar syntax and error-reporting semantics as other Google Test assertions, and thus is easy to learn.
*`ASSERT_DEATH` can be mixed with other assertions and other logic at your will. You are not limited to one death test per test method. For example, you can write something like:
```
if (FooCondition()) {
ASSERT_DEATH(Bar(), "blah");
} else {
ASSERT_EQ(5, Bar());
}
```
If you prefer one death test per test method, you can write your tests in that style too, but we don't want to impose that on the users. The fewer artificial limitations the better.
*`ASSERT_DEATH` can reference local variables in the current function, and you can decide how many death tests you want based on run-time information. For example,
```
const int count = GetCount(); // Only known at run time.
for (int i = 1; i <= count; i++) {
ASSERT_DEATH({
double* buffer = new double[i];
... initializes buffer ...
Foo(buffer, i)
}, "blah blah");
}
```
The runner-based approach tends to be more static and less flexible, or requires more user effort to get this kind of flexibility.
Another interesting thing about `ASSERT_DEATH` is that it calls `fork()`
to create a child process to run the death test. This is lightening
fast, as `fork()` uses copy-on-write pages and incurs almost zero
overhead, and the child process starts from the user-supplied
statement directly, skipping all global and local initialization and
any code leading to the given statement. If you launch the child
process from scratch, it can take seconds just to load everything and
start running if the test links to many libraries dynamically.
## My death test modifies some state, but the change seems lost after the death test finishes. Why? ##
Death tests (`EXPECT_DEATH`, etc) are executed in a sub-process s.t. the
expected crash won't kill the test program (i.e. the parent process). As a
result, any in-memory side effects they incur are observable in their
respective sub-processes, but not in the parent process. You can think of them
as running in a parallel universe, more or less.
## The compiler complains about "undefined references" to some static const member variables, but I did define them in the class body. What's wrong? ##
If your class has a static data member:
```
// foo.h
class Foo {
...
static const int kBar = 100;
};
```
You also need to define it _outside_ of the class body in `foo.cc`:
```
const int Foo::kBar; // No initializer here.
```
Otherwise your code is **invalid C++**, and may break in unexpected ways. In
particular, using it in Google Test comparison assertions (`EXPECT_EQ`, etc)
will generate an "undefined reference" linker error.
## I have an interface that has several implementations. Can I write a set of tests once and repeat them over all the implementations? ##
Google Test doesn't yet have good support for this kind of tests, or
data-driven tests in general. We hope to be able to make improvements in this
area soon.
## Can I derive a test fixture from another? ##
Yes.
Each test fixture has a corresponding and same named test case. This means only
one test case can use a particular fixture. Sometimes, however, multiple test
cases may want to use the same or slightly different fixtures. For example, you
may want to make sure that all of a GUI library's test cases don't leak
important system resources like fonts and brushes.
In Google Test, you share a fixture among test cases by putting the shared
logic in a base test fixture, then deriving from that base a separate fixture
for each test case that wants to use this common logic. You then use `TEST_F()`
to write tests using each derived fixture.
Typically, your code looks like this:
```
// Defines a base test fixture.
class BaseTest : public ::testing::Test {
protected:
...
};
// Derives a fixture FooTest from BaseTest.
class FooTest : public BaseTest {
protected:
virtual void SetUp() {
BaseTest::SetUp(); // Sets up the base fixture first.
... additional set-up work ...
}
virtual void TearDown() {
... clean-up work for FooTest ...
BaseTest::TearDown(); // Remember to tear down the base fixture
// after cleaning up FooTest!
}
... functions and variables for FooTest ...
};
// Tests that use the fixture FooTest.
TEST_F(FooTest, Bar) { ... }
TEST_F(FooTest, Baz) { ... }
... additional fixtures derived from BaseTest ...
```
If necessary, you can continue to derive test fixtures from a derived fixture.
Google Test has no limit on how deep the hierarchy can be.
For a complete example using derived test fixtures, see
[sample5](../samples/sample5_unittest.cc).
## My compiler complains "void value not ignored as it ought to be." What does this mean? ##
You're probably using an `ASSERT_*()` in a function that doesn't return `void`.
`ASSERT_*()` can only be used in `void` functions.
## My death test hangs (or seg-faults). How do I fix it? ##
In Google Test, death tests are run in a child process and the way they work is
delicate. To write death tests you really need to understand how they work.
Please make sure you have read this.
In particular, death tests don't like having multiple threads in the parent
process. So the first thing you can try is to eliminate creating threads
outside of `EXPECT_DEATH()`.
Sometimes this is impossible as some library you must use may be creating
threads before `main()` is even reached. In this case, you can try to minimize
the chance of conflicts by either moving as many activities as possible inside
`EXPECT_DEATH()` (in the extreme case, you want to move everything inside), or
leaving as few things as possible in it. Also, you can try to set the death
test style to `"threadsafe"`, which is safer but slower, and see if it helps.
If you go with thread-safe death tests, remember that they rerun the test
program from the beginning in the child process. Therefore make sure your
program can run side-by-side with itself and is deterministic.
In the end, this boils down to good concurrent programming. You have to make
sure that there is no race conditions or dead locks in your program. No silver
bullet - sorry!
## Should I use the constructor/destructor of the test fixture or the set-up/tear-down function? ##
The first thing to remember is that Google Test does not reuse the
same test fixture object across multiple tests. For each `TEST_F`,
Google Test will create a fresh test fixture object, _immediately_
call `SetUp()`, run the test, call `TearDown()`, and then
_immediately_ delete the test fixture object. Therefore, there is no
need to write a `SetUp()` or `TearDown()` function if the constructor
or destructor already does the job.
You may still want to use `SetUp()/TearDown()` in the following cases:
* If the tear-down operation could throw an exception, you must use `TearDown()` as opposed to the destructor, as throwing in a destructor leads to undefined behavior and usually will kill your program right away. Note that many standard libraries (like STL) may throw when exceptions are enabled in the compiler. Therefore you should prefer `TearDown()` if you want to write portable tests that work with or without exceptions.
* The assertion macros throw an exception when flag `--gtest_throw_on_failure` is specified. Therefore, you shouldn't use Google Test assertions in a destructor if you plan to run your tests with this flag.
* In a constructor or destructor, you cannot make a virtual function call on this object. (You can call a method declared as virtual, but it will be statically bound.) Therefore, if you need to call a method that will be overriden in a derived class, you have to use `SetUp()/TearDown()`.
## The compiler complains "no matching function to call" when I use ASSERT\_PREDn. How do I fix it? ##
If the predicate function you use in `ASSERT_PRED*` or `EXPECT_PRED*` is
overloaded or a template, the compiler will have trouble figuring out which
overloaded version it should use. `ASSERT_PRED_FORMAT*` and
`EXPECT_PRED_FORMAT*` don't have this problem.
If you see this error, you might want to switch to
`(ASSERT|EXPECT)_PRED_FORMAT*`, which will also give you a better failure
message. If, however, that is not an option, you can resolve the problem by
explicitly telling the compiler which version to pick.
(The stuff inside the angled brackets for the `static_cast` operator is the
type of the function pointer for the `int`-version of `IsPositive()`.)
As another example, when you have a template function
```
template <typename T>
bool IsNegative(T x) {
return x < 0;
}
```
you can use it in a predicate assertion like this:
```
ASSERT_PRED1(IsNegative*<int>*, -5);
```
Things are more interesting if your template has more than one parameters. The
following won't compile:
```
ASSERT_PRED2(*GreaterThan<int, int>*, 5, 0);
```
as the C++ pre-processor thinks you are giving `ASSERT_PRED2` 4 arguments,
which is one more than expected. The workaround is to wrap the predicate
function in parentheses:
```
ASSERT_PRED2(*(GreaterThan<int, int>)*, 5, 0);
```
## My compiler complains about "ignoring return value" when I call RUN\_ALL\_TESTS(). Why? ##
Some people had been ignoring the return value of `RUN_ALL_TESTS()`. That is,
instead of
```
return RUN_ALL_TESTS();
```
they write
```
RUN_ALL_TESTS();
```
This is wrong and dangerous. A test runner needs to see the return value of
`RUN_ALL_TESTS()` in order to determine if a test has passed. If your `main()`
function ignores it, your test will be considered successful even if it has a
Google Test assertion failure. Very bad.
To help the users avoid this dangerous bug, the implementation of
`RUN_ALL_TESTS()` causes gcc to raise this warning, when the return value is
ignored. If you see this warning, the fix is simple: just make sure its value
is used as the return value of `main()`.
## My compiler complains that a constructor (or destructor) cannot return a value. What's going on? ##
Due to a peculiarity of C++, in order to support the syntax for streaming
messages to an `ASSERT_*`, e.g.
```
ASSERT_EQ(1, Foo()) << "blah blah" << foo;
```
we had to give up using `ASSERT*` and `FAIL*` (but not `EXPECT*` and
`ADD_FAILURE*`) in constructors and destructors. The workaround is to move the
content of your constructor/destructor to a private void member function, or
switch to `EXPECT_*()` if that works. This section in the user's guide explains
it.
## My set-up function is not called. Why? ##
C++ is case-sensitive. It should be spelled as `SetUp()`. Did you
spell it as `Setup()`?
Similarly, sometimes people spell `SetUpTestCase()` as `SetupTestCase()` and
wonder why it's never called.
## How do I jump to the line of a failure in Emacs directly? ##
Google Test's failure message format is understood by Emacs and many other
IDEs, like acme and XCode. If a Google Test message is in a compilation buffer
in Emacs, then it's clickable. You can now hit `enter` on a message to jump to
the corresponding source code, or use `C-x `` to jump to the next failure.
## I have several test cases which share the same test fixture logic, do I have to define a new test fixture class for each of them? This seems pretty tedious. ##
You don't have to. Instead of
```
class FooTest : public BaseTest {};
TEST_F(FooTest, Abc) { ... }
TEST_F(FooTest, Def) { ... }
class BarTest : public BaseTest {};
TEST_F(BarTest, Abc) { ... }
TEST_F(BarTest, Def) { ... }
```
you can simply `typedef` the test fixtures:
```
typedef BaseTest FooTest;
TEST_F(FooTest, Abc) { ... }
TEST_F(FooTest, Def) { ... }
typedef BaseTest BarTest;
TEST_F(BarTest, Abc) { ... }
TEST_F(BarTest, Def) { ... }
```
## The Google Test output is buried in a whole bunch of log messages. What do I do? ##
The Google Test output is meant to be a concise and human-friendly report. If
your test generates textual output itself, it will mix with the Google Test
output, making it hard to read. However, there is an easy solution to this
problem.
Since most log messages go to stderr, we decided to let Google Test output go
to stdout. This way, you can easily separate the two using redirection. For
example:
```
./my_test > googletest_output.txt
```
## Why should I prefer test fixtures over global variables? ##
There are several good reasons:
1. It's likely your test needs to change the states of its global variables. This makes it difficult to keep side effects from escaping one test and contaminating others, making debugging difficult. By using fixtures, each test has a fresh set of variables that's different (but with the same names). Thus, tests are kept independent of each other.
1. Global variables pollute the global namespace.
1. Test fixtures can be reused via subclassing, which cannot be done easily with global variables. This is useful if many test cases have something in common.
## How do I test private class members without writing FRIEND\_TEST()s? ##
You should try to write testable code, which means classes should be easily
tested from their public interface. One way to achieve this is the Pimpl idiom:
you move all private members of a class into a helper class, and make all
members of the helper class public.
You have several other options that don't require using `FRIEND_TEST`:
* Write the tests as members of the fixture class:
```
class Foo {
friend class FooTest;
...
};
class FooTest : public ::testing::Test {
protected:
...
void Test1() {...} // This accesses private members of class Foo.
void Test2() {...} // So does this one.
};
TEST_F(FooTest, Test1) {
Test1();
}
TEST_F(FooTest, Test2) {
Test2();
}
```
* In the fixture class, write accessors for the tested class' private members, then use the accessors in your tests:
```
class Foo {
friend class FooTest;
...
};
class FooTest : public ::testing::Test {
protected:
...
T1 get_private_member1(Foo* obj) {
return obj->private_member1_;
}
};
TEST_F(FooTest, Test1) {
...
get_private_member1(x)
...
}
```
* If the methods are declared **protected**, you can change their access level in a test-only subclass:
```
class YourClass {
...
protected: // protected access for testability.
int DoSomethingReturningInt();
...
};
// in the your_class_test.cc file:
class TestableYourClass : public YourClass {
...
public: using YourClass::DoSomethingReturningInt; // changes access rights
## I have a fixture class Foo, but TEST\_F(Foo, Bar) gives me error "no matching function for call to Foo::Foo()". Why? ##
Google Test needs to be able to create objects of your test fixture class, so
it must have a default constructor. Normally the compiler will define one for
you. However, there are cases where you have to define your own:
* If you explicitly declare a non-default constructor for class `Foo`, then you need to define a default constructor, even if it would be empty.
* If `Foo` has a const non-static data member, then you have to define the default constructor _and_ initialize the const member in the initializer list of the constructor. (Early versions of `gcc` doesn't force you to initialize the const member. It's a bug that has been fixed in `gcc 4`.)
## Why does ASSERT\_DEATH complain about previous threads that were already joined? ##
With the Linux pthread library, there is no turning back once you cross the
line from single thread to multiple threads. The first time you create a
thread, a manager thread is created in addition, so you get 3, not 2, threads.
Later when the thread you create joins the main thread, the thread count
decrements by 1, but the manager thread will never be killed, so you still have
2 threads, which means you cannot safely run a death test.
The new NPTL thread library doesn't suffer from this problem, as it doesn't
create a manager thread. However, if you don't control which machine your test
runs on, you shouldn't depend on this.
## Why does Google Test require the entire test case, instead of individual tests, to be named FOODeathTest when it uses ASSERT\_DEATH? ##
Google Test does not interleave tests from different test cases. That is, it
runs all tests in one test case first, and then runs all tests in the next test
case, and so on. Google Test does this because it needs to set up a test case
before the first test in it is run, and tear it down afterwords. Splitting up
the test case would require multiple set-up and tear-down processes, which is
inefficient and makes the semantics unclean.
If we were to determine the order of tests based on test name instead of test
case name, then we would have a problem with the following situation:
```
TEST_F(FooTest, AbcDeathTest) { ... }
TEST_F(FooTest, Uvw) { ... }
TEST_F(BarTest, DefDeathTest) { ... }
TEST_F(BarTest, Xyz) { ... }
```
Since `FooTest.AbcDeathTest` needs to run before `BarTest.Xyz`, and we don't
interleave tests from different test cases, we need to run all tests in the
`FooTest` case before running any test in the `BarTest` case. This contradicts
with the requirement to run `BarTest.DefDeathTest` before `FooTest.Uvw`.
## But I don't like calling my entire test case FOODeathTest when it contains both death tests and non-death tests. What do I do? ##
You don't have to, but if you like, you may split up the test case into
`FooTest` and `FooDeathTest`, where the names make it clear that they are
Therefore we strongly advise against the practice, and Google Test doesn't
provide a way to do it.
In general, the recommended way to cause the code to behave
differently under test is [dependency injection](http://jamesshore.com/Blog/Dependency-Injection-Demystified.html).
You can inject different functionality from the test and from the
production code. Since your production code doesn't link in the
for-test logic at all, there is no danger in accidentally running it.
However, if you _really_, _really_, _really_ have no choice, and if
you follow the rule of ending your test program names with `_test`,
you can use the _horrible_ hack of sniffing your executable name
(`argv[0]` in `main()`) to know whether the code is under test.
## Google Test defines a macro that clashes with one defined by another library. How do I deal with that? ##
In C++, macros don't obey namespaces. Therefore two libraries that
both define a macro of the same name will clash if you `#include` both
definitions. In case a Google Test macro clashes with another
library, you can force Google Test to rename its macro to avoid the
conflict.
Specifically, if both Google Test and some other code define macro
`FOO`, you can add
```
-DGTEST_DONT_DEFINE_FOO=1
```
to the compiler flags to tell Google Test to change the macro's name
from `FOO` to `GTEST_FOO`. For example, with `-DGTEST_DONT_DEFINE_TEST=1`, you'll need to write
```
GTEST_TEST(SomeTest, DoesThis) { ... }
```
instead of
```
TEST(SomeTest, DoesThis) { ... }
```
in order to define a test.
Currently, the following `TEST`, `FAIL`, `SUCCEED`, and the basic comparison assertion macros can have alternative names. You can see the full list of covered macros [here](http://www.google.com/codesearch?q=if+!GTEST_DONT_DEFINE_\w%2B+package:http://googletest\.googlecode\.com+file:/include/gtest/gtest.h). More information can be found in the "Avoiding Macro Name Clashes" section of the README file.
## Is it OK if I have two separate `TEST(Foo, Bar)` test methods defined in different namespaces? ##
Yes.
The rule is **all test methods in the same test case must use the same fixture class**. This means that the following is **allowed** because both tests use the same fixture class (`::testing::Test`).
```
namespace foo {
TEST(CoolTest, DoSomething) {
SUCCEED();
}
} // namespace foo
namespace bar {
TEST(CoolTest, DoSomething) {
SUCCEED();
}
} // namespace foo
```
However, the following code is **not allowed** and will produce a runtime error from Google Test because the test methods are using different test fixture classes with the same test case name.
```
namespace foo {
class CoolTest : public ::testing::Test {}; // Fixture foo::CoolTest
TEST_F(CoolTest, DoSomething) {
SUCCEED();
}
} // namespace foo
namespace bar {
class CoolTest : public ::testing::Test {}; // Fixture: bar::CoolTest
TEST_F(CoolTest, DoSomething) {
SUCCEED();
}
} // namespace foo
```
## How do I build Google Testing Framework with Xcode 4? ##
If you try to build Google Test's Xcode project with Xcode 4.0 or later, you may encounter an error message that looks like
"Missing SDK in target gtest\_framework: /Developer/SDKs/MacOSX10.4u.sdk". That means that Xcode does not support the SDK the project is targeting. See the Xcode section in the [README](../../README.MD) file on how to resolve this.
## My question is not covered in your FAQ! ##
If you cannot find the answer to your question in this FAQ, there are
some other resources you can use:
1. read other [wiki pages](http://code.google.com/p/googletest/w/list),
1. search the mailing list [archive](http://groups.google.com/group/googletestframework/topics),
1. ask it on [googletestframework@googlegroups.com](mailto:googletestframework@googlegroups.com) and someone will answer it (to prevent spam, we require you to join the [discussion group](http://groups.google.com/group/googletestframework) before you can post.).
Please note that creating an issue in the
[issue tracker](http://code.google.com/p/googletest/issues/list) is _not_
a good way to get your answer, as it is monitored infrequently by a
very small number of people.
When asking a question, it's helpful to provide as much of the
following information as possible (people cannot help you if there's
not enough information in your question):
* the version (or the revision number if you check out from SVN directly) of Google Test you use (Google Test is under active development, so it's possible that your problem has been solved in a later version),
* your operating system,
* the name and version of your compiler,
* the complete command line flags you give to your compiler,
* the complete compiler error messages (if the question is about compilation),
* the _actual_ code (ideally, a minimal but complete program) that has the problem you encounter.
# Introduction: Why Google C++ Testing Framework? #
_Google C++ Testing Framework_ helps you write better C++ tests.
No matter whether you work on Linux, Windows, or a Mac, if you write C++ code,
Google Test can help you.
So what makes a good test, and how does Google C++ Testing Framework fit in? We believe:
1. Tests should be _independent_ and _repeatable_. It's a pain to debug a test that succeeds or fails as a result of other tests. Google C++ Testing Framework isolates the tests by running each of them on a different object. When a test fails, Google C++ Testing Framework allows you to run it in isolation for quick debugging.
1. Tests should be well _organized_ and reflect the structure of the tested code. Google C++ Testing Framework groups related tests into test cases that can share data and subroutines. This common pattern is easy to recognize and makes tests easy to maintain. Such consistency is especially helpful when people switch projects and start to work on a new code base.
1. Tests should be _portable_ and _reusable_. The open-source community has a lot of code that is platform-neutral, its tests should also be platform-neutral. Google C++ Testing Framework works on different OSes, with different compilers (gcc, MSVC, and others), with or without exceptions, so Google C++ Testing Framework tests can easily work with a variety of configurations. (Note that the current release only contains build scripts for Linux - we are actively working on scripts for other platforms.)
1. When tests fail, they should provide as much _information_ about the problem as possible. Google C++ Testing Framework doesn't stop at the first test failure. Instead, it only stops the current test and continues with the next. You can also set up tests that report non-fatal failures after which the current test continues. Thus, you can detect and fix multiple bugs in a single run-edit-compile cycle.
1. The testing framework should liberate test writers from housekeeping chores and let them focus on the test _content_. Google C++ Testing Framework automatically keeps track of all tests defined, and doesn't require the user to enumerate them in order to run them.
1. Tests should be _fast_. With Google C++ Testing Framework, you can reuse shared resources across tests and pay for the set-up/tear-down only once, without making tests depend on each other.
Since Google C++ Testing Framework is based on the popular xUnit
architecture, you'll feel right at home if you've used JUnit or PyUnit before.
If not, it will take you about 10 minutes to learn the basics and get started.
So let's go!
_Note:_ We sometimes refer to Google C++ Testing Framework informally
as _Google Test_.
# Setting up a New Test Project #
To write a test program using Google Test, you need to compile Google
Test into a library and link your test with it. We provide build
files for some popular build systems: `msvc/` for Visual Studio,
`xcode/` for Mac Xcode, `make/` for GNU make, `codegear/` for Borland
C++ Builder, and the autotools script (deprecated) and
`CMakeLists.txt` for CMake (recommended) in the Google Test root
directory. If your build system is not on this list, you can take a
look at `make/Makefile` to learn how Google Test should be compiled
(basically you want to compile `src/gtest-all.cc` with `GTEST_ROOT`
and `GTEST_ROOT/include` in the header search path, where `GTEST_ROOT`
is the Google Test root directory).
Once you are able to compile the Google Test library, you should
create a project or build target for your test program. Make sure you
have `GTEST_ROOT/include` in the header search path so that the
compiler can find `"gtest/gtest.h"` when compiling your test. Set up
your test project to link with the Google Test library (for example,
in Visual Studio, this is done by adding a dependency on
`gtest.vcproj`).
If you still have questions, take a look at how Google Test's own
tests are built and use them as examples.
# Basic Concepts #
When using Google Test, you start by writing _assertions_, which are statements
that check whether a condition is true. An assertion's result can be _success_,
_nonfatal failure_, or _fatal failure_. If a fatal failure occurs, it aborts
the current function; otherwise the program continues normally.
_Tests_ use assertions to verify the tested code's behavior. If a test crashes
or has a failed assertion, then it _fails_; otherwise it _succeeds_.
A _test case_ contains one or many tests. You should group your tests into test
cases that reflect the structure of the tested code. When multiple tests in a
test case need to share common objects and subroutines, you can put them into a
_test fixture_ class.
A _test program_ can contain multiple test cases.
We'll now explain how to write a test program, starting at the individual
assertion level and building up to tests and test cases.
# Assertions #
Google Test assertions are macros that resemble function calls. You test a
class or function by making assertions about its behavior. When an assertion
fails, Google Test prints the assertion's source file and line number location,
along with a failure message. You may also supply a custom failure message
which will be appended to Google Test's message.
The assertions come in pairs that test the same thing but have different
effects on the current function. `ASSERT_*` versions generate fatal failures
when they fail, and **abort the current function**. `EXPECT_*` versions generate
nonfatal failures, which don't abort the current function. Usually `EXPECT_*`
are preferred, as they allow more than one failures to be reported in a test.
However, you should use `ASSERT_*` if it doesn't make sense to continue when
the assertion in question fails.
Since a failed `ASSERT_*` returns from the current function immediately,
possibly skipping clean-up code that comes after it, it may cause a space leak.
Depending on the nature of the leak, it may or may not be worth fixing - so
keep this in mind if you get a heap checker error in addition to assertion
errors.
To provide a custom failure message, simply stream it into the macro using the
`<<` operator, or a sequence of such operators. An example:
```
ASSERT_EQ(x.size(), y.size()) << "Vectors x and y are of unequal length";
for (int i = 0; i < x.size(); ++i) {
EXPECT_EQ(x[i], y[i]) << "Vectors x and y differ at index " << i;
}
```
Anything that can be streamed to an `ostream` can be streamed to an assertion
macro--in particular, C strings and `string` objects. If a wide string
(`wchar_t*`, `TCHAR*` in `UNICODE` mode on Windows, or `std::wstring`) is
streamed to an assertion, it will be translated to UTF-8 when printed.
## Basic Assertions ##
These assertions do basic true/false condition testing.
| `ASSERT_STREQ(`_expected\_str_`, `_actual\_str_`);` | `EXPECT_STREQ(`_expected\_str_`, `_actual\_str_`);` | the two C strings have the same content |
| `ASSERT_STRNE(`_str1_`, `_str2_`);` | `EXPECT_STRNE(`_str1_`, `_str2_`);` | the two C strings have different content |
| `ASSERT_STRCASEEQ(`_expected\_str_`, `_actual\_str_`);`| `EXPECT_STRCASEEQ(`_expected\_str_`, `_actual\_str_`);` | the two C strings have the same content, ignoring case |
| `ASSERT_STRCASENE(`_str1_`, `_str2_`);`| `EXPECT_STRCASENE(`_str1_`, `_str2_`);` | the two C strings have different content, ignoring case |
Note that "CASE" in an assertion name means that case is ignored.
`*STREQ*` and `*STRNE*` also accept wide C strings (`wchar_t*`). If a
comparison of two wide strings fails, their values will be printed as UTF-8
narrow strings.
A `NULL` pointer and an empty string are considered _different_.
_Availability_: Linux, Windows, Mac.
See also: For more string comparison tricks (substring, prefix, suffix, and
regular expression matching, for example), see the [Advanced Google Test Guide](V1_7_AdvancedGuide.md).
# Simple Tests #
To create a test:
1. Use the `TEST()` macro to define and name a test function, These are ordinary C++ functions that don't return a value.
1. In this function, along with any valid C++ statements you want to include, use the various Google Test assertions to check values.
1. The test's result is determined by the assertions; if any assertion in the test fails (either fatally or non-fatally), or if the test crashes, the entire test fails. Otherwise, it succeeds.
```
TEST(test_case_name, test_name) {
... test body ...
}
```
`TEST()` arguments go from general to specific. The _first_ argument is the
name of the test case, and the _second_ argument is the test's name within the
test case. Both names must be valid C++ identifiers, and they should not contain underscore (`_`). A test's _full name_ consists of its containing test case and its
individual name. Tests from different test cases can have the same individual
name.
For example, let's take a simple integer function:
```
int Factorial(int n); // Returns the factorial of n
```
A test case for this function might look like:
```
// Tests factorial of 0.
TEST(FactorialTest, HandlesZeroInput) {
EXPECT_EQ(1, Factorial(0));
}
// Tests factorial of positive numbers.
TEST(FactorialTest, HandlesPositiveInput) {
EXPECT_EQ(1, Factorial(1));
EXPECT_EQ(2, Factorial(2));
EXPECT_EQ(6, Factorial(3));
EXPECT_EQ(40320, Factorial(8));
}
```
Google Test groups the test results by test cases, so logically-related tests
should be in the same test case; in other words, the first argument to their
`TEST()` should be the same. In the above example, we have two tests,
`HandlesZeroInput` and `HandlesPositiveInput`, that belong to the same test
case `FactorialTest`.
_Availability_: Linux, Windows, Mac.
# Test Fixtures: Using the Same Data Configuration for Multiple Tests #
If you find yourself writing two or more tests that operate on similar data,
you can use a _test fixture_. It allows you to reuse the same configuration of
objects for several different tests.
To create a fixture, just:
1. Derive a class from `::testing::Test` . Start its body with `protected:` or `public:` as we'll want to access fixture members from sub-classes.
1. Inside the class, declare any objects you plan to use.
1. If necessary, write a default constructor or `SetUp()` function to prepare the objects for each test. A common mistake is to spell `SetUp()` as `Setup()` with a small `u` - don't let that happen to you.
1. If necessary, write a destructor or `TearDown()` function to release any resources you allocated in `SetUp()` . To learn when you should use the constructor/destructor and when you should use `SetUp()/TearDown()`, read this [FAQ entry](V1_7_FAQ.md#should-i-use-the-constructordestructor-of-the-test-fixture-or-the-set-uptear-down-function).
1. If needed, define subroutines for your tests to share.
When using a fixture, use `TEST_F()` instead of `TEST()` as it allows you to
access objects and subroutines in the test fixture:
```
TEST_F(test_case_name, test_name) {
... test body ...
}
```
Like `TEST()`, the first argument is the test case name, but for `TEST_F()`
this must be the name of the test fixture class. You've probably guessed: `_F`
is for fixture.
Unfortunately, the C++ macro system does not allow us to create a single macro
that can handle both types of tests. Using the wrong macro causes a compiler
error.
Also, you must first define a test fixture class before using it in a
`TEST_F()`, or you'll get the compiler error "`virtual outside class
declaration`".
For each test defined with `TEST_F()`, Google Test will:
1. Create a _fresh_ test fixture at runtime
1. Immediately initialize it via `SetUp()` ,
1. Run the test
1. Clean up by calling `TearDown()`
1. Delete the test fixture. Note that different tests in the same test case have different test fixture objects, and Google Test always deletes a test fixture before it creates the next one. Google Test does not reuse the same test fixture for multiple tests. Any changes one test makes to the fixture do not affect other tests.
As an example, let's write tests for a FIFO queue class named `Queue`, which
has the following interface:
```
template <typename E> // E is the element type.
class Queue {
public:
Queue();
void Enqueue(const E& element);
E* Dequeue(); // Returns NULL if the queue is empty.
size_t size() const;
...
};
```
First, define a fixture class. By convention, you should give it the name
`FooTest` where `Foo` is the class being tested.
```
class QueueTest : public ::testing::Test {
protected:
virtual void SetUp() {
q1_.Enqueue(1);
q2_.Enqueue(2);
q2_.Enqueue(3);
}
// virtual void TearDown() {}
Queue<int> q0_;
Queue<int> q1_;
Queue<int> q2_;
};
```
In this case, `TearDown()` is not needed since we don't have to clean up after
each test, other than what's already done by the destructor.
Now we'll write tests using `TEST_F()` and this fixture.
```
TEST_F(QueueTest, IsEmptyInitially) {
EXPECT_EQ(0, q0_.size());
}
TEST_F(QueueTest, DequeueWorks) {
int* n = q0_.Dequeue();
EXPECT_EQ(NULL, n);
n = q1_.Dequeue();
ASSERT_TRUE(n != NULL);
EXPECT_EQ(1, *n);
EXPECT_EQ(0, q1_.size());
delete n;
n = q2_.Dequeue();
ASSERT_TRUE(n != NULL);
EXPECT_EQ(2, *n);
EXPECT_EQ(1, q2_.size());
delete n;
}
```
The above uses both `ASSERT_*` and `EXPECT_*` assertions. The rule of thumb is
to use `EXPECT_*` when you want the test to continue to reveal more errors
after the assertion failure, and use `ASSERT_*` when continuing after failure
doesn't make sense. For example, the second assertion in the `Dequeue` test is
`ASSERT_TRUE(n != NULL)`, as we need to dereference the pointer `n` later,
which would lead to a segfault when `n` is `NULL`.
When these tests run, the following happens:
1. Google Test constructs a `QueueTest` object (let's call it `t1` ).
1.`t1.SetUp()` initializes `t1` .
1. The first test ( `IsEmptyInitially` ) runs on `t1` .
1.`t1.TearDown()` cleans up after the test finishes.
1.`t1` is destructed.
1. The above steps are repeated on another `QueueTest` object, this time running the `DequeueWorks` test.
_Availability_: Linux, Windows, Mac.
_Note_: Google Test automatically saves all _Google Test_ flags when a test
object is constructed, and restores them when it is destructed.
# Invoking the Tests #
`TEST()` and `TEST_F()` implicitly register their tests with Google Test. So, unlike with many other C++ testing frameworks, you don't have to re-list all your defined tests in order to run them.
After defining your tests, you can run them with `RUN_ALL_TESTS()` , which returns `0` if all the tests are successful, or `1` otherwise. Note that `RUN_ALL_TESTS()` runs _all tests_ in your link unit -- they can be from different test cases, or even different source files.
When invoked, the `RUN_ALL_TESTS()` macro:
1. Saves the state of all Google Test flags.
1. Creates a test fixture object for the first test.
1. Initializes it via `SetUp()`.
1. Runs the test on the fixture object.
1. Cleans up the fixture via `TearDown()`.
1. Deletes the fixture.
1. Restores the state of all Google Test flags.
1. Repeats the above steps for the next test, until all tests have run.
In addition, if the text fixture's constructor generates a fatal failure in
step 2, there is no point for step 3 - 5 and they are thus skipped. Similarly,
if step 3 generates a fatal failure, step 4 will be skipped.
_Important_: You must not ignore the return value of `RUN_ALL_TESTS()`, or `gcc`
will give you a compiler error. The rationale for this design is that the
automated testing service determines whether a test has passed based on its
exit code, not on its stdout/stderr output; thus your `main()` function must
return the value of `RUN_ALL_TESTS()`.
Also, you should call `RUN_ALL_TESTS()` only **once**. Calling it more than once
conflicts with some advanced Google Test features (e.g. thread-safe death
tests) and thus is not supported.
_Availability_: Linux, Windows, Mac.
# Writing the main() Function #
You can start from this boilerplate:
```
#include "this/package/foo.h"
#include "gtest/gtest.h"
namespace {
// The fixture for testing class Foo.
class FooTest : public ::testing::Test {
protected:
// You can remove any or all of the following functions if its body
// is empty.
FooTest() {
// You can do set-up work for each test here.
}
virtual ~FooTest() {
// You can do clean-up work that doesn't throw exceptions here.
}
// If the constructor and destructor are not enough for setting up
// and cleaning up each test, you can define the following methods:
virtual void SetUp() {
// Code here will be called immediately after the constructor (right
// before each test).
}
virtual void TearDown() {
// Code here will be called immediately after each test (right
// before the destructor).
}
// Objects declared here can be used by all tests in the test case for Foo.
The `::testing::InitGoogleTest()` function parses the command line for Google
Test flags, and removes all recognized flags. This allows the user to control a
test program's behavior via various flags, which we'll cover in [AdvancedGuide](V1_7_AdvancedGuide.md).
You must call this function before calling `RUN_ALL_TESTS()`, or the flags
won't be properly initialized.
On Windows, `InitGoogleTest()` also works with wide strings, so it can be used
in programs compiled in `UNICODE` mode as well.
But maybe you think that writing all those main() functions is too much work? We agree with you completely and that's why Google Test provides a basic implementation of main(). If it fits your needs, then just link your test with gtest\_main library and you are good to go.
## Important note for Visual C++ users ##
If you put your tests into a library and your `main()` function is in a different library or in your .exe file, those tests will not run. The reason is a [bug](https://connect.microsoft.com/feedback/viewfeedback.aspx?FeedbackID=244410&siteid=210) in Visual C++. When you define your tests, Google Test creates certain static objects to register them. These objects are not referenced from elsewhere but their constructors are still supposed to run. When Visual C++ linker sees that nothing in the library is referenced from other places it throws the library out. You have to reference your library with tests from your main program to keep the linker from discarding it. Here is how to do it. Somewhere in your library code declare a function:
```
__declspec(dllexport) int PullInMyLibrary() { return 0; }
```
If you put your tests in a static library (not DLL) then `__declspec(dllexport)` is not required. Now, in your main program, write a code that invokes that function:
```
int PullInMyLibrary();
static int dummy = PullInMyLibrary();
```
This will keep your tests referenced and will make them register themselves at startup.
In addition, if you define your tests in a static library, add `/OPT:NOREF` to your main program linker options. If you use MSVC++ IDE, go to your .exe project properties/Configuration Properties/Linker/Optimization and set References setting to `Keep Unreferenced Data (/OPT:NOREF)`. This will keep Visual C++ linker from discarding individual symbols generated by your tests from the final executable.
There is one more pitfall, though. If you use Google Test as a static library (that's how it is defined in gtest.vcproj) your tests must also reside in a static library. If you have to have them in a DLL, you _must_ change Google Test to build into a DLL as well. Otherwise your tests will not run correctly or will not run at all. The general conclusion here is: make your life easier - do not write your tests in libraries!
# Where to Go from Here #
Congratulations! You've learned the Google Test basics. You can start writing
and running Google Test tests, read some [samples](V1_7_Samples.md), or continue with
[AdvancedGuide](V1_7_AdvancedGuide.md), which describes many more useful Google Test features.
# Known Limitations #
Google Test is designed to be thread-safe. The implementation is
thread-safe on systems where the `pthreads` library is available. It
is currently _unsafe_ to use Google Test assertions from two threads
concurrently on other systems (e.g. Windows). In most tests this is
not an issue as usually the assertions are done in the main thread. If
you want to help, you can volunteer to implement the necessary
synchronization primitives in `gtest-port.h` for your platform.
<b>P</b>ump is <b>U</b>seful for <b>M</b>eta <b>P</b>rogramming.
# The Problem #
Template and macro libraries often need to define many classes,
functions, or macros that vary only (or almost only) in the number of
arguments they take. It's a lot of repetitive, mechanical, and
error-prone work.
Variadic templates and variadic macros can alleviate the problem.
However, while both are being considered by the C++ committee, neither
is in the standard yet or widely supported by compilers. Thus they
are often not a good choice, especially when your code needs to be
portable. And their capabilities are still limited.
As a result, authors of such libraries often have to write scripts to
generate their implementation. However, our experience is that it's
tedious to write such scripts, which tend to reflect the structure of
the generated code poorly and are often hard to read and edit. For
example, a small change needed in the generated code may require some
non-intuitive, non-trivial changes in the script. This is especially
painful when experimenting with the code.
# Our Solution #
Pump (for Pump is Useful for Meta Programming, Pretty Useful for Meta
Programming, or Practical Utility for Meta Programming, whichever you
prefer) is a simple meta-programming tool for C++. The idea is that a
programmer writes a `foo.pump` file which contains C++ code plus meta
code that manipulates the C++ code. The meta code can handle
iterations over a range, nested iterations, local meta variable
definitions, simple arithmetic, and conditional expressions. You can
view it as a small Domain-Specific Language. The meta language is
designed to be non-intrusive (s.t. it won't confuse Emacs' C++ mode,
for example) and concise, making Pump code intuitive and easy to
maintain.
## Highlights ##
* The implementation is in a single Python script and thus ultra portable: no build or installation is needed and it works cross platforms.
* Pump tries to be smart with respect to [Google's style guide](http://code.google.com/p/google-styleguide/): it breaks long lines (easy to have when they are generated) at acceptable places to fit within 80 columns and indent the continuation lines correctly.
* The format is human-readable and more concise than XML.
* The format works relatively well with Emacs' C++ mode.
## Examples ##
The following Pump code (where meta keywords start with `$`, `[[` and `]]` are meta brackets, and `$$` starts a meta comment that ends with the line):
```
$var n = 3 $$ Defines a meta variable n.
$range i 0..n $$ Declares the range of meta iterator i (inclusive).
$for i [[
$$ Meta loop.
// Foo$i does blah for $i-ary predicates.
$range j 1..i
template <size_t N $for j [[, typename A$j]]>
class Foo$i {
$if i == 0 [[
blah a;
]] $elif i <= 2 [[
blah b;
]] $else [[
blah c;
]]
};
]]
```
will be translated by the Pump compiler to:
```
// Foo0 does blah for 0-ary predicates.
template <size_t N>
class Foo0 {
blah a;
};
// Foo1 does blah for 1-ary predicates.
template <size_t N, typename A1>
class Foo1 {
blah b;
};
// Foo2 does blah for 2-ary predicates.
template <size_t N, typename A1, typename A2>
class Foo2 {
blah b;
};
// Foo3 does blah for 3-ary predicates.
template <size_t N, typename A1, typename A2, typename A3>
class Foo3 {
blah c;
};
```
In another example,
```
$range i 1..n
Func($for i + [[a$i]]);
$$ The text between i and [[ is the separator between iterations.
```
will generate one of the following lines (without the comments), depending on the value of `n`:
```
Func(); // If n is 0.
Func(a1); // If n is 1.
Func(a1 + a2); // If n is 2.
Func(a1 + a2 + a3); // If n is 3.
// And so on...
```
## Constructs ##
We support the following meta programming constructs:
| `$var id = exp` | Defines a named constant value. `$id` is valid util the end of the current meta lexical block. |
**Note:** To give the user some freedom in formatting the Pump source
code, Pump ignores a new-line character if it's right after `$for foo`
or next to `[[` or `]]`. Without this rule you'll often be forced to write
very long lines to get the desired output. Therefore sometimes you may
need to insert an extra new-line in such places for a new-line to show
up in your output.
## Grammar ##
```
code ::= atomic_code*
atomic_code ::= $var id = exp
| $var id = [[ code ]]
| $range id exp..exp
| $for id sep [[ code ]]
| $($)
| $id
| $(exp)
| $if exp [[ code ]] else_branch
| [[ code ]]
| cpp_code
sep ::= cpp_code | empty_string
else_branch ::= $else [[ code ]]
| $elif exp [[ code ]] else_branch
| empty_string
exp ::= simple_expression_in_Python_syntax
```
## Code ##
You can find the source code of Pump in [scripts/pump.py](../scripts/pump.py). It is still
very unpolished and lacks automated tests, although it has been
successfully used many times. If you find a chance to use it in your
project, please let us know what you think! We also welcome help on
improving Pump.
## Real Examples ##
You can find real-world applications of Pump in [Google Test](http://www.google.com/codesearch?q=file%3A\.pump%24+package%3Ahttp%3A%2F%2Fgoogletest\.googlecode\.com) and [Google Mock](http://www.google.com/codesearch?q=file%3A\.pump%24+package%3Ahttp%3A%2F%2Fgooglemock\.googlecode\.com). The source file `foo.h.pump` generates `foo.h`.
## Tips ##
* If a meta variable is followed by a letter or digit, you can separate them using `[[]]`, which inserts an empty string. For example `Foo$j[[]]Helper` generate `Foo1Helper` when `j` is 1.
* To avoid extra-long Pump source lines, you can break a line anywhere you want by inserting `[[]]` followed by a new line. Since any new-line character next to `[[` or `]]` is ignored, the generated code won't contain this new line.
*[Sample #7](../samples/sample7_unittest.cc) teaches the basics of value-parameterized tests.
*[Sample #8](../samples/sample8_unittest.cc) shows using `Combine()` in value-parameterized tests.
*[Sample #9](../samples/sample9_unittest.cc) shows use of the listener API to modify Google Test's console output and the use of its reflection API to inspect test results.
*[Sample #10](../samples/sample10_unittest.cc) shows use of the listener API to implement a primitive memory leak checker.
This guide will explain how to use the Google Testing Framework in your Xcode projects on Mac OS X. This tutorial begins by quickly explaining what to do for experienced users. After the quick start, the guide goes provides additional explanation about each step.
# Quick Start #
Here is the quick guide for using Google Test in your Xcode project.
1. Download the source from the [website](http://code.google.com/p/googletest) using this command: `svn checkout http://googletest.googlecode.com/svn/trunk/ googletest-read-only`
1. Open up the `gtest.xcodeproj` in the `googletest-read-only/xcode/` directory and build the gtest.framework.
1. Create a new "Shell Tool" target in your Xcode project called something like "UnitTests"
1. Add the gtest.framework to your project and add it to the "Link Binary with Libraries" build phase of "UnitTests"
1. Add your unit test source code to the "Compile Sources" build phase of "UnitTests"
1. Edit the "UnitTests" executable and add an environment variable named "DYLD\_FRAMEWORK\_PATH" with a value equal to the path to the framework containing the gtest.framework relative to the compiled executable.
1. Build and Go
The following sections further explain each of the steps listed above in depth, describing in more detail how to complete it including some variations.
# Get the Source #
Currently, the gtest.framework discussed here isn't available in a tagged release of Google Test, it is only available in the trunk. As explained at the Google Test [site](http://code.google.com/p/googletest/source/checkout">svn), you can get the code from anonymous SVN with this command:
Alternatively, if you are working with Subversion in your own code base, you can add Google Test as an external dependency to your own Subversion repository. By following this approach, everyone that checks out your svn repository will also receive a copy of Google Test (a specific version, if you wish) without having to check it out explicitly. This makes the set up of your project simpler and reduces the copied code in the repository.
To use `svn:externals`, decide where you would like to have the external source reside. You might choose to put the external source inside the trunk, because you want it to be part of the branch when you make a release. However, keeping it outside the trunk in a version-tagged directory called something like `third-party/googletest/1.0.1`, is another option. Once the location is established, use `svn propedit svn:externals _directory_` to set the svn:externals property on a directory in your repository. This directory won't contain the code, but be its versioned parent directory.
The command `svn propedit` will bring up your Subversion editor, making editing the long, (potentially multi-line) property simpler. This same method can be used to check out a tagged branch, by using the appropriate URL (e.g. `http://googletest.googlecode.com/svn/tags/release-1.0.1`). Additionally, the svn:externals property allows the specification of a particular revision of the trunk with the `-r_##_` option (e.g. `externals/src/googletest -r60 http://googletest.googlecode.com/svn/trunk`).
Here is an example of using the svn:externals properties on a trunk (read via `svn propget`) of a project. This value checks out a copy of Google Test into the `trunk/externals/src/googletest/` directory.
The next step is to build and add the gtest.framework to your own project. This guide describes two common ways below.
***Option 1** --- The simplest way to add Google Test to your own project, is to open gtest.xcodeproj (found in the xcode/ directory of the Google Test trunk) and build the framework manually. Then, add the built framework into your project using the "Add->Existing Framework..." from the context menu or "Project->Add..." from the main menu. The gtest.framework is relocatable and contains the headers and object code that you'll need to make tests. This method requires rebuilding every time you upgrade Google Test in your project.
***Option 2** --- If you are going to be living off the trunk of Google Test, incorporating its latest features into your unit tests (or are a Google Test developer yourself). You'll want to rebuild the framework every time the source updates. to do this, you'll need to add the gtest.xcodeproj file, not the framework itself, to your own Xcode project. Then, from the build products that are revealed by the project's disclosure triangle, you can find the gtest.framework, which can be added to your targets (discussed below).
# Make a Test Target #
To start writing tests, make a new "Shell Tool" target. This target template is available under BSD, Cocoa, or Carbon. Add your unit test source code to the "Compile Sources" build phase of the target.
Next, you'll want to add gtest.framework in two different ways, depending upon which option you chose above.
***Option 1** --- During compilation, Xcode will need to know that you are linking against the gtest.framework. Add the gtest.framework to the "Link Binary with Libraries" build phase of your test target. This will include the Google Test headers in your header search path, and will tell the linker where to find the library.
***Option 2** --- If your working out of the trunk, you'll also want to add gtest.framework to your "Link Binary with Libraries" build phase of your test target. In addition, you'll want to add the gtest.framework as a dependency to your unit test target. This way, Xcode will make sure that gtest.framework is up to date, every time your build your target. Finally, if you don't share build directories with Google Test, you'll have to copy the gtest.framework into your own build products directory using a "Run Script" build phase.
# Set Up the Executable Run Environment #
Since the unit test executable is a shell tool, it doesn't have a bundle with a `Contents/Frameworks` directory, in which to place gtest.framework. Instead, the dynamic linker must be told at runtime to search for the framework in another location. This can be accomplished by setting the "DYLD\_FRAMEWORK\_PATH" environment variable in the "Edit Active Executable ..." Arguments tab, under "Variables to be set in the environment:". The path for this value is the path (relative or absolute) of the directory containing the gtest.framework.
If you haven't set up the DYLD\_FRAMEWORK\_PATH, correctly, you might get a message like this:
```
[Session started at 2008-08-15 06:23:57 -0600.]
dyld: Library not loaded: @loader_path/../Frameworks/gtest.framework/Versions/A/gtest
To correct this problem, got to the directory containing the executable named in "Referenced from:" value in the error message above. Then, with the terminal in this location, find the relative path to the directory containing the gtest.framework. That is the value you'll need to set as the DYLD\_FRAMEWORK\_PATH.
# Build and Go #
Now, when you click "Build and Go", the test will be executed. Dumping out something like this:
```
[Session started at 2008-08-06 06:36:13 -0600.]
[==========] Running 2 tests from 1 test case.
[----------] Global test environment set-up.
[----------] 2 tests from WidgetInitializerTest
[ RUN ] WidgetInitializerTest.TestConstructor
[ OK ] WidgetInitializerTest.TestConstructor
[ RUN ] WidgetInitializerTest.TestConversion
[ OK ] WidgetInitializerTest.TestConversion
[----------] Global test environment tear-down
[==========] 2 tests from 1 test case ran.
[ PASSED ] 2 tests.
The Debugger has exited with status 0.
```
# Summary #
Unit testing is a valuable way to ensure your data model stays valid even during rapid development or refactoring. The Google Testing Framework is a great unit testing framework for C and C++ which integrates well with an Xcode development environment.
@@ -6,13 +6,13 @@ This guide will explain how to use the Google Testing Framework in your Xcode pr
Here is the quick guide for using Google Test in your Xcode project.
1. Download the source from the [website](http://code.google.com/p/googletest) using this command: `svn checkout http://googletest.googlecode.com/svn/trunk/ googletest-read-only`
1. Download the source from the [website](http://code.google.com/p/googletest) using this command: `svn checkout http://googletest.googlecode.com/svn/trunk/ googletest-read-only`.
1. Open up the `gtest.xcodeproj` in the `googletest-read-only/xcode/` directory and build the gtest.framework.
1. Create a new "Shell Tool" target in your Xcode project called something like "UnitTests"
1. Add the gtest.framework to your project and add it to the "Link Binary with Libraries" build phase of "UnitTests"
1. Add your unit test source code to the "Compile Sources" build phase of "UnitTests"
1. Create a new "Shell Tool" target in your Xcode project called something like "UnitTests".
1. Add the gtest.framework to your project and add it to the "Link Binary with Libraries" build phase of "UnitTests".
1. Add your unit test source code to the "Compile Sources" build phase of "UnitTests".
1. Edit the "UnitTests" executable and add an environment variable named "DYLD\_FRAMEWORK\_PATH" with a value equal to the path to the framework containing the gtest.framework relative to the compiled executable.
1. Build and Go
1. Build and Go.
The following sections further explain each of the steps listed above in depth, describing in more detail how to complete it including some variations.
...
...
@@ -66,7 +66,7 @@ If you haven't set up the DYLD\_FRAMEWORK\_PATH, correctly, you might get a mess
Reason: image not found
```
To correct this problem, got to the directory containing the executable named in "Referenced from:" value in the error message above. Then, with the terminal in this location, find the relative path to the directory containing the gtest.framework. That is the value you'll need to set as the DYLD\_FRAMEWORK\_PATH.
To correct this problem, go to to the directory containing the executable named in "Referenced from:" value in the error message above. Then, with the terminal in this location, find the relative path to the directory containing the gtest.framework. That is the value you'll need to set as the DYLD\_FRAMEWORK\_PATH.
