The slowdown appears to be due to an implicit conversion of distinct (yet semantically identical) lambdas to `std::function`. Lifting out the lambdas into functors that don't get re-instantiated reduces compilation times by nearly half.
Fixes#4156
PiperOrigin-RevId: 523447948
Change-Id: Ib0ae0761a54d7b1f2b706b14b2858eedf47e2297
This seems to just have been an oversight. POSIX socket APIs work just
fine on macOS.
Fixes https://github.com/google/googletest/issues/4214.
PiperOrigin-RevId: 523398386
Change-Id: I9d56cd9c6933318c1f0b0024f7fef44122fd0c83
Ensure that tests are run in the order specified in the source code, even if
they are registered manually using RegisterTest. There should be no behavior
change for the common case.
PiperOrigin-RevId: 522136303
Change-Id: If155e2666780af0e514fbbf5ff2b157d5fe2fef1
This CL changes the link in the ReportUninterestingCall message from
.../gmock_cook_book.md#knowing-when-to-expect to
.../gmock_cook_book.md#knowing-when-to-expect-useoncall. This is necessary
following https://github.com/google/googletest/commit/31ff597.
PiperOrigin-RevId: 510138974
Change-Id: Ic98c84b07751d27dfc95eddbe7874f76d68b456f
When checked out on Windows, the repo might use \r\n line endings,
and so the golden output has them. Adjust for that.
PiperOrigin-RevId: 502577222
Change-Id: Iabfe537f6d95a49bf6bdcb934e855d28c65f6f89
@platforms//os:* instead of from @build_tools//platforms:*
See https://github.com/bazelbuild/bazel/issues/8622
Necessary to fix#4096
Also need the fix for #4098
PiperOrigin-RevId: 497182073
Change-Id: Ifd568b088d2f779755dd20264edfd5dad12ca9cc
It is good practice to specify a name for the `testing::Property` matcher's use in diagnostics, and we should do so in our examples.
PiperOrigin-RevId: 496406309
Change-Id: Ibb4d1ba63771b5d6855442b91564027cda8ddaca
Clang warns on this pattern because it looks like the author might
have meant to use the value of the first part of the comma operator,
so it warns that it isn't being used. The cast here signals to Clang
that this behavior is intentional.
This was discovered while updating gmock in Android. Clang's -Wcomma
warning is on by default with either -Wall or -Werror, so users of
gmock with those on in combination with -Werror are unable to build
without this fix.
PiperOrigin-RevId: 495655990
Change-Id: Iaf27e2199669f5b6185a877738234e551b6b6556
EXPECT_CALL doesn't seem to make much sense here. I think the intent was to use EXPECT_THAT instead.
PiperOrigin-RevId: 495427663
Change-Id: I39a16ea37282729d34017c6bcd331a60cedf1fbe
MatcherDescriberInterface specifies that DescribeTo "should print a verb phrase", but "size ..." is not a verb phrase. Currently, ElementsAre(SizeIs(9)) is described as "has 1 element that size is equal to 9". With this change, it will be described as "has 1 element that has a size that is equal to 9".
PiperOrigin-RevId: 492022324
Change-Id: I4083335f2419462464957521c1e033643b53b763
To make debug output readable, we still use the faster 6-digit precision
sometimes, but only if it will round-trip.
This way, when a test fails due to a very small difference in floating-point
numbers, users will have enough digits to see the difference.
PiperOrigin-RevId: 488958311
Change-Id: Ibcac43f48a97006d89217530c69386cc4fa2735c
This aims to avoid confusion as to what kinds of computations are valid inside
a parameter generator expression, calling out flags as a supported use case.
PiperOrigin-RevId: 483397027
Change-Id: I2d036fae95120d617f30a5566ea7498ce1f9bfb6
This matches the intention and documentation of terse printing which generally avoids printing the pointer.
PiperOrigin-RevId: 481178950
Change-Id: I27039dac1870934d2d5b212e2cc7e97ab82c5b34
In developing tests for the fuchsia debugger, it was found that in addition to catching gtest failures (which are implemented as software breakpoints) we also see PageFault exceptions, caused by this nullptr dereference.
PiperOrigin-RevId: 479365782
Change-Id: I84d805d94c2e46b6f3c982ca1ae49c6ac3ed3430
Remove the MSVC pragmas for disabling warning C4717 (infinite recursion) for
Invalid<T>() because that warning has been fixed in cl/441474979.
PiperOrigin-RevId: 473012585
Change-Id: I5f1bf88379bd4f2bf005e029c04766ac4caadd84
The way temporary directories were discovered from
environment variables in different operating systems
was inconsistent.
On Windows, only the system-specific environment variable
TEMP was considered, but not the bazel-common TEST_TMPDIR.
On Unix, only TEST_TMPDIR was considered, but not
the otherwise typical system specific TMPDIR.
Now, always consider TEST_TMPDIR first, followed by the
system-typical environment variable (TEMP on Windows, TMPDIR on Unix)
before falling back to a default on that particular system.
Also: the value for the temporary directory coming from the
environment variable was only checked for a trailing directory
separator on Windows (and appended if needed), not on the other
platforms.
Make this also consistent accross the systems: Now always apply
the same logic to all of these (to accomodate
tests that just concatenate TempDir() without directory separator).
PiperOrigin-RevId: 447909830
Change-Id: I9ea17acdf5944eb2a965615fd9cf142878c33a58
This CL adds a couple missing overrides in the googletest sources.
These were found downstream when -Wsuggest-override and
-Wsuggest-destructor-override were enabled.
PiperOrigin-RevId: 447754883
Change-Id: I7bf35a8757cbc5ae157827037aa3d13f47392406
Currently MacOS falls back to generic /tmp, but
for all intents and purposes it should behave like
other Unixes using the TEST_TMPDIR environment variable
if available (this environment variable is set in bazel,
which sets up a unique temp directory for the test
process).
While at it, remove an incorrect #endif comment, that
looks like a leftover from some older implementation.
PiperOrigin-RevId: 446108391
Change-Id: I118eacf6e86a41d26cb81a130f7c54cccc0c5665
This provides a type-safe way for an action to express that it wants to be
called only once, or to capture move-only objects. It is a generalization of
the type system-evading hack in ByMove, with the improvement that it works for
_any_ action (including user-defined ones), and correctly expresses that the
action can only be used with WillOnce. I'll make existing actions benefit in a
future commit.
PiperOrigin-RevId: 440496139
Change-Id: I4145d191cca5655995ef41360bb126c123cb41d3
Avoid instantiating functions like std::get<index> for an out of range index
when doing SFINAE on the invocability of the action itself.
PiperOrigin-RevId: 439415110
Change-Id: Ifc20285a6d526c34830870cd1910c2b2b92e1e81
Currently, the "[ DISABLED ]" banner is printed for every test in a suite.
When iterating on a single test gtest_filter this is very noisy.
PiperOrigin-RevId: 436489088
Change-Id: If337087a7a0986b073fabf2b0a55d26485eb5c37
A few tests are examining code locations and looking af the resulting line
numbers to verify that GoogleTest shows those to users correctly. Some of those
locations change when clang-format is run. For those locations, I've wrapped
portions in:
// clang-format off
...
// clang-format on
There may be other locations that are currently not tickled by running
clang-format.
PiperOrigin-RevId: 434844712
Change-Id: I3a9f0a6f39eff741c576b6de389bef9b1d11139d
This catches when a client creates an action and discards it, thinking that the action has actually been applied to something.
This will help people who make the mistake of defining, for example, both `void Use(Foo*, Bar)` and `ACTION_P(Use, bar) { Use(arg, bar); }` for later application to a Foo. With such an overload, a client may then write `Use(bar);`, selecting the Action<> overload and being confused why nothing happens.
This also catches when a client defines their own action in terms of an ACTION_P()-generated one, invokes the Action<>'s builder, and then fails to invoke the resulting action, thinking it's operating on the outer action's parameters.
PiperOrigin-RevId: 433197479
Change-Id: I98e4389150d01a5e753230113016d9fc38b1d260
Abseil Team