We do not need an explicit INDEX column for ADDITIONAL to work.
If the ADDITIONAL option is set the constraint should be passed
into the virtual table context.
This is to have better separation between them and the python tests.
BUCK files have been updated accordingly.
Changes to the internal generateCopyFileTarget function were needed.
The function now supports a base folder to be set so that the regex or file path
is appended to that base path, without having it included in the destination.
It will also not use a library target anymore, but a custom one so
that's possible to set properties with custom names.
Use only one copy target to copy test config files instead of having
multiple that overlaps.
It may give issues when copying and tests in the end will need all of
them.
* Summary:
This adds tests for the windows and darwin conversions strings helper
classes. Darwin currently has tests, but they are not hooked up and
being run. Windows does not have tests, this adds two simple unit tests
against the logic for our string conversions.
Test Plan:
Ran the tests.
Recent changes in INDEX enforcement broke extensions.
This fix adds a compatibility flag that defaults to the old behavior of INDEX on all columns.
This only affects extension virtual tables, all built in tables have index constraints enforced.
While in the end the osqueryd binary will require almost all libraries
built and linked, having each intermediate library depend on the whole
set of AWS libraries might result in higher compiling time.
Especially since on some platforms not all libraries are used.
This should speed up the compilation speed on Windows.
This is an API change for the format of logged events.
The top-level log field "logNumericsAsNumbers" was introduced to help
migrate from string-encoded numerics to JSON numbers.
This change updates the field to be "numerics", updates the flag to
conform to flag naming conventions, and documents the expectation.
When testing on IPv6-only hosts, with an IPv4 lo stack, the resolver may
not return an IPv4 address. This seems to be an issue with the
deprecated overload of async_resolve. Upon inspection, the previous
overload would only return 1 result_type.
Migrating to the recommended overload works as expected.
Read the analysis within #5952.
There are multiple sources suggesting an optimal value is 0. Our
implementation previously hardcoded the value to 1. It is now 0 and can
be configured to a different value using the audit_backlog_wait_time
option or command line flag.
If a Thrift server is requested to stop before everything is
properly initialized a race condition could happen.
Specifically if stopping during a listen() a race condition will happen
accessing childInterruptSockWriter_, which is a socket initialized by listen(),
in a non thread-safe way, and used by stop() to notify other threads
that they should shutdown.
This was determined to be the wrong approach to adding simple
killswitches. The intent was to quickly flip on/off features. It was
not widely adopted due to the dependencies killswitching has.
A different approach is to implement the same functions with something
simple like filesystem flags.
There is a bug in RapidJSON 1.1.0 that results in wild pointer reads
when parsing exponents. We want to prevent regressing.
This requires updating the Buck/pre-built libraries.
When a thread different from the main requests a shutdown
through Initializer::requestShutdown, it should not call
waitForShutdown; there's no reason to wait, moreover the function
doesn't only wait, but also actually stops other components and then
finally calls exit().
Since the main thread is already inside the waitForShutdown call
waiting on Dispatcher::joinServices or inside the shutdown() callable
on Windows, having a secondary thread do
the same work potentially at the same time is wrong.
Moreover calling exit() from a secondary thread is most of the time
incorrect.
The waitForShutdown function has been renamed to waitThenShutdown
to better represent what it's actually doing.
