var_dump() is debugging functionality, so it should print
floating-point numbers accurately. We do this by switching
to serialize_precision, which (by default) will print with
as much precision as necessary to preserve the exact value
of the float.
This also affects debug_zval_dump().
Closes GH-5172.
The primary motivation to have each test worker running its own console
is to allow the windows_mb_path tests to run in parallel. A nice side
effect is that this also prevents changing the code page of the
tester's console window (which can even cause its font to be changed).
To be able to do so, we introduce the `create_new_console` option for
`proc_open()`, which might occasionally be useful for other purposes
than testing.
- When valgrind is used, communication overhead is relatively small,
so just use a batch size of 1.
- If this is running a small enough number of tests,
reduce the batch size to give batches to more workers.
(Previously, if there were 90 tests and -j8,
only 3 of 8 workers would get a batch of size 32 or less.
After this change, the batch size is 12 or less)
Closes GH-5098
If there's only 2 files to test, then only start 2 workers instead of N.
If there's only 1 file, then avoid parallelism entirely.
A separate option such as `--force-parallel` could be added
if this turns out to be something developers would want to do
when debugging test failures.
Currently mail related tests are split for *nix and Windows (if there
are even Windows versions). The basic difference is that the *nix
variants set the INI directive sendmail_path to just write the email to
disk, while the Windows tests use ext/imap. The latter tests are way
more verbose, and such duplicated tests are generally a pain point.
Furthermore, the Windows tests are much slower, and could not be run
without ext/imap being available.
We therefore introduce a small fakemail application, which basically
works like `tee <path> >/dev/null`, and which will be shipped with the
Windows tests packs. fakemail.exe would also need to be added to the
PHP binary SDK, so these tests could be run during developments.
To cater to the remaining differences, we also introduce support for
`{MAIL:<path>}` placeholders in the INI sections to run-tests.php. How
to use this can be seen in mail_basic.phpt, which is currently the only
modified test case, because these tests are yet supposed to fail on
Windows, due to the missing fakemail.exe in the PHP SDK.
Now that we're displaying errors in skipif, suppress this particular
category. Otherwise we get warnings in the SKIPIF of tests with
EXTENSION where the extension does not exist, and there's no way
to suppress them.
Otherwise we would try to access an array element of `false`, which
issues a notice as of PHP 7.4.0. This would happen, for instance, for
bug63447_001.phpt if CGI is not available.
- when $status is boolean, E_NOTICE appears and tests fail
- opcache is never enabled on these tests anyway.
- Add opcache.revalidate_freq=0 to testing script
Instead of checking this in the tests files it can be used directly in
the main testing script. Also this fixes failed tests.
- Use opcache_invalidate() instead of sleep() delay in some tests. Some
delays are still used since the opcache_invalidate seems to be buggy
and not working as expected.
Closes GH-4392
The classic Windows console only supports the Unicode BMP, so it can't
handle the LIGHTNING MOOD character (U+1F5F2). It also seems that
common fonts do not support the LIGHTNING (U+2607) character, so a
general removal is appropriate.
However, Windows is already "ready for the future"[1], but PHP for
Windows is not (yet).
[1] <https://devblogs.microsoft.com/commandline/windows-command-line-unicode-and-utf-8-output-text-buffer/>
Formerly, there was at most a single OPcache instance per user and the
so called system ID (which is determined from the PHP version).
Sometimes multiple OPcaches might be desired, though, particularly for
unrelated CLI scripts, which may even be necessary (e.g. for our test
suite in parallel mode).
We therefore introduce a new INI directive `opcache.cache_id` which
allows to configure independent OPcache instances for the same user.
We also use `GetUserNameW()` instead of `php_win32_get_username()`,
because the latter retrieves the user name encoded in the
`default_charset`, which can obviously yield different results for
different charsets, leading to OPcache "incompatibilities". Slightly
worse, some characters may not even be encodeable in the
`default_charset` and would be replaced by question marks, which could
result in different users sharing the same OPcache.
We also refactor, and re-use existing APIs to avoid duplicated code.
Also adds an --asan flag to run-tests.php to setup all the necessary
environment variables. Some tests are marked as skipped because they
are incompatible with asan or too slow.
I'm basing this on the DEBUG_ZTS build, which seems to give us the
most mileage.
If we're testing ini settings that cause startup failures, we'll
never get to that SKIPIF block...
Also change settings2param to return a value instead of modifying
(to a different type no less!) in place.
