operating systems.

The plugin gets the sid of the client process and considers the user
authenticated if the given username matches the username of this sid.

Problem was that insert-order (enforced by the optimizer) did not handle
the case where the bitmap changed to a new one.

Fixed by remembering the last bitmap page used and to force usage of this when inserting new rows

Account for timeout of select returning a 0 error code.

        The metadata creation part of the mysqld_shww_create separated
        to be used on the mysqld_stmt_prepare stage.

Removing the "conv_charset" member and using collation.collation
instead, as they duplicated each other.

fix compiler warnings

Debian tests are run in 2017, update the test to
use 2027 as a "future".

Don't compare "field == table->next_number_field" because the field
can be special nullable field copy created by the trigger.
Compare field_index values instead.

Cannot do password validation in fix_lex_user(), we don't know
there what "GRANT ... TO user" means - creating a new user with
an empty password (need validation) or granting privileges
to an existing user (no validation needed).

Move validation down into replace_user_table(). And copy into
check_change_password().

increase the thread stack a bit for main.signal_demo3
and sys_vars.max_sp_recursion_depth_func to stop failing

Was:
* LEX_USER::password was storing sometimes
  plaintext password and sometimes password hash
* LEX_USER::auth was storing sometimes password hash and
  sometimes plugin authentication string

Now:
* LEX_USER::pwtext stores the password in plain-text
* LEX_USER::pwhash stores the password hash
* LEX_USER::auth stores the plugin authentication string

as this is used quite often

add /usr/share/pkgconfig to the rpm ignored list

allow more characters in a valid user/group name:
* POSIX allows dashes '-' and dots '.'
* also the name may end with a dollar sign '$'

for our purposes it's enough to allow [-.$] anywhere in the name

different fix. remove old ones, wait for THD to be fully
initialized before continuing with the server startup process.

This reverts commits db2e21bf, 13615c5e, 3f515a09, 70113ee1.

Conflicts:
	plugin/server_audit/server_audit.c

        Now both are generated by the cmake from the
        scripts/maria_add_gis_sp.sql.in

Undo the change in test_if_skip_sort_order() that set ref_key=-1 when
a variant of index_merge is used (was made in fix for MDEV-9021).

It turned out that test_if_cheaper_ordering() call below assumes that
ref_key=-1 means "no index is used", that is, "an inefficient full table
scan is done".
This is not the same as index_merge, index_merge can actually be quite
efficient. So, ref_key=MAX_KEY denotes the fact that some index is used,
not any given index.

Do not include BLOB fields by default.

Fix spelling of my name

innodbchecksum: add fclose and handle errors

This reverts commit 727f92fe.

This reverts commit df32495c.

ISO C90 does not allow mixed declarations and code.