The default sql mode must now be provided explicitly when the query
classifier is setup. This is in preparation for "sql_mode" becoming
a global configuration parameter of MaxScale.
That's where it belongs as it is only the mysql client protocol that
will use it. It's a bit unfortunate that the qc test program compare
now needs to include a file from a protocol module directory, but
the fact is that the query classifier implementation and the test
programs should actually be *under* the mysql client protocol module.
The current sql mode will be tracked in that variable. Since one
thread will handle multiple sessions, we need to track the sql mode
of a session, so that the sql mode of the query classifier can be
set whenever we handle a request of a particular session.
As the sql-mode will be managed by the client protocol, the data
can be private to it as well.
The parser is needed not only by mysql_client but also by the test
program of the qc-plugins, so it need to be available someplace
common.
Not that nice that both a qc-component and mysql_client will include
stuff from server/core/maxscale, but at least temporarily ok.
EXPLAIN statements are no longer parsed completely as doing so makes
it hard to modify the grammar for the needs or Oracle SQL.
Consequently, for an EXPLAIN statement you now bascially only get the
type and the operation (the newly added QUERY_OP_EXPLAIN and QUERY_OP_SHOW).
The other information is not interesting and is related to
information_schema and similar tables.
Several filters use a match-or-ignore logic with two regular
expressions when filtering queries. This commit adds a convenience
function for this task. Also adds a convenience function for reading
several regular expression parameters at once, compiling them and
saving the code while checking for errors.
Also, use the new functions in QLA and CCR filters.
The test created a DCB of an unexpected type in an unexpected state with
an invalid file descriptor. This caused the test to fail due to a debug
assertion where the the validity of the file descriptor was asserted in
relation to the internal state of the DCB.
The two cases where the case-sensitive parser functions were used don't
appear to hold any special meaning. The case-insensitive function should
be used as it implements a superset of functionality compared to the
case-sensitive version.
If a prepared statement response was collected into one buffer, it doesn't
need to be processed again. By jumping directly to the routing of the
collected result, we prevent the unnecessary splitting of buffers that
appears to happend with continuous preparations of prepared statements.
Providing helper functions for the commonly used parts of the server makes
code easier to read. It also removes any possibility for formatting
problems by moving the URI and name string handling inside the Backend
class.
Added helper functions that check various server states. This makes the
readwritesplit code easier to read as the function names convey the
intention better than the macro invokations.
As the session commands are always appended to the end of the list, the
name should reflect that action. For this reason, the function was renamed
to append_session_command.
Readwritesplit supports replacement of slave servers by storing all
executed session commands in a list. To make the copying of this list a
bit cleaner, an overload for a list of session commands was added. This
will allow relatively smooth addition of server replacement to all router
modules that use the Backend class.
The class now allows simpler construction of session commands by
overloading the add_session_command with a version that accepts a const
reference to a shared pointer. This removes the need to copy the
references to the source buffer by calling gwbuf_clone.
Exposed the first session command as a const reference to allow
interaction with it. Currently, it is planned to be used to get the
session command position of each backend.
This is the first step to taking the Backend class into use. It is now
used in rwsplit_select_backends.cc and readwritesplit.hh. The module is
not yet functional and doesn't even compile.
Added some helper functions to the Backend class to get easier access to
the server referenced by the SERVER_REF and to check the state of the
backend.
The states are now internal to the Backend class. This simplifies the use
of the class by moving the burder of state tracking to the class
itself.
Refactored the way the schemarouter uses the Backend class.
Also fixed a memory leak in the schemarouter when `ignore_databases_regex`
was used..
Using the same implementation of the Backend and SessionCommand classes in
both schemarouter and readwritesplit will prevent duplication of code.
This commit only splits the generic parts of the class to a Backend class
which the schemarouter then extends. The session commands for both routers
are similar so they require no special handling.
That allows the version to be updated and read atomically. If
major/minor/patch are stored as separate variables, you can get an
inconsistent set. Now it may be out of date by the time it is used,
but it will never be internally inconsistent.
The behaviour of the query classifier needs to change depending
on the version of the servers of a service. With this function
"some" version of the service can be obtained.
The behaviour of the query classifier needs to be different
depending on the actual version of the server. There is already
a human readable string, but for programmatic use it needs to
be in a format that can easily be parsed.
If no regex type parameters are defined and a compiled pattern is
requested, the function will return a NULL value. This allows code that
directly calls the config_get_compiled_regex function to work without
doing additional checks for the presence of optional parameters.
With this change, it is no longer possible that the server version
is deallocated at the very moment it is read. There is still a race,
but it's mostly harmless.
MXS-1282. The flags "-g=arglist" or "--debug=arglist" activate
the debug settings specified in the comma-separated list arglist.
Currently, the setting "disable-module-unloading", which prevents
modules (.so-files) from being unloaded at exit, is supported. This
allows Valgrind to produce line numbers for leak reports when the
memory was allocated in a module. The setting "enable-module-unloading"
is also available, but this is activated by default.
The debug arguments and their actions are defined in a structure
array, so adding more is straightforward.
The function now returns the compiled code or null or error. Also,
instead of capcount, it writes the ovector size. The ovector pointer
can be null. Removed the other config_get_compiled_regex()-function
and replaced it with the refactored function since they are almost
identical.
The admin interface now uses HTTP BA authentication by default. This will
prevent unrestricted access to the REST API but the authentication is by
no means secure and the HTTPS mode for the REST API should be enabled for
all production systems.
The module commands can now produce JSON formatted output which is passed
to the caller. The output should conform to the JSON API as closely as
possible.
Currently, the REST API wraps all JSON produced by module commands inside
a meta-object of the following type:
{
"meta": <output of module command>
}
This allows the output to be JSON API conformant without modifying the
modules and allows incremental updates to code.
This is for backwards compatibility to allow current modules to use
the automatically compiled regex parameters. The enforcement should
be restored in a later version. A warning is output for every non-
enclosed regex.
Similar changes are also applied for QUOTED_STRING.
