With this change, both 'SET @my_id = (SELECT LAST_INSERT_ID())' and
'SET @my_id = (SELECT @@LAST_INSERT_ID)' are classified as
QUERY_TYPE_MASTER_READ|QUERY_TYPE_SESSION_WRITE|QUERY_TYPE_USERVAR_WRITE
Earlier @@last_insert_id had the QUERY_TYPE_READ bit set as well, which
indirectly caused the problems of MXS-2898.
DIV and MOD are now also accepted instead of / and % respectively.
MOD is a keyword but (in principle incorrectly) decays into an id
when used in some other context. That is so that it will be
parser by the general function rule ("id ( ... )"). If used
incorrectly, the server will later reject.
It appears that the function db_my_assert may be used but not
built into the library.
With 10.4.12 there seem to occasionally be superfluous empty
table names.
The mon_ping_or_connect_to_db resets the MYSQL handle which caused the
loss of the error message. Returning a new enumeration value for
authentication errors solves this problem.
The library is now named "maxtest". The related include-files are, for
now, usable without designating the full include path. This may be changed
later, but would require modifying every test.
Documented readconnroute and schemarouter router diagnostics. The
duplicated fields that are already in the general service diagnostic
output are not documented with the hopes that they can be eventually
removed.
The code in 2.3 was changed to allow "empty" SSL definitions now that the
system CA is used. The code in 2.4 did not allow this which caused non-TLS
connections to be created when only ssl=true was defined.
The code relied on last_read for the idle time calculation which caused
the pings that were written to not reset the idle time. This increased the
chance of multiple COM_PING packets being sent to a backend before a reply
was received.
The use of the server state is not transactional across multiple uses of
the function. This means that any assertions on the target state can fail
if the monitor updates the state between target selection and the
assertion.
The parameter can be modified at runtime, as is evident from the unit
tests that test it, and should not be listed in the set of unmodifiable
parameters.
The backend didn't expect AuthSwitchRequest packets in response to the
handshake response packets. This is allowed by the protocol and appears to
happen with at least MySQL 8.0.
