using VIEW and WITH RECURSIVE.

The cause of this crash was the same as of the crash reported in mdev-14755.

with condition_pushdown_for_derived=on

This bug is a consequence of the bug mdev-14368 fixed in 5.5.59.

If the specification of a CTE contains a reference to a temporary table
then THD::open_temporary_table() must be called for this reference for
any occurrence of the CTE in the query. By mistake this was done only
for the first occurrences of CTEs.
The patch fixes this problem in With_element::clone_parsed_spec().
It also moves there the call of check_dependencies_in_with_clauses()
to its proper place before the call of check_table_access().
Additionally the patch optimizes the number of calls of the
function check_dependencies_in_with_clauses().

row_sel_get_clust_rec_for_mysql(): Look up the page from the
buffer pool, similar to how MySQL 5.7 does it.

debug_key_management
encrypt_and_grep
innodb_encryption

If real table count is different from what is expected by the test, it
just hangs on waiting to fulfill hardcoded number. And then exits with
**failed** after 10 minutes of wait: quite unfriendly and hard to
figure out what's going on.

Fix and enable some of the tests; some remain disabled.

The tests innodb_gis.rtree_old and innodb_gis.row_format
duplicated some versions of the test main.gis-rtree.
Instead of duplicating, source that test, in a new test
innodb_gis.innodb_gis_rtree.

Introduce innodb_row_format.combinations. Due to this,
ROW_FORMAT=COMPRESSED will not be covered in some tests
where it is covered in MySQL 5.7.

The function rtr_update_mbr_field_in_place() is generating
MLOG_REC_UPDATE_IN_PLACE or MLOG_COMP_REC_UPDATE_IN_PLACE records
on non-leaf pages, even though MLOG_WRITE_STRING would perfectly
suffice for updating a fixed-length data field.

btr_cur_parse_update_in_place(): If flags==7, the record may be
from rtr_update_mbr_field_in_place(), and we must check if the
page is a leaf page. Otherwise, assume that it is.

btr_cur_update_in_place(): Assert that the page is a leaf page.

Other things, mainly to get
create_mysqld_error_find_printf_error tool to work:

- Added protection to not include mysqld_error.h twice
- Include "unireg.h" instead of "mysqld_error.h" in server
- Added protection if ER_XX messages are already defined
- Removed wrong calls to my_error(ER_OUTOFMEMORY) as
  my_malloc() and my_alloc will do this automatically
- Added missing %s to ER_DUP_QUERY_NAME
- Removed old and wrong calls to my_strerror() when using
  MY_ERROR_ON_RENAME (wrong merge)
- Fixed deadlock error message from Galera. Before the extra
  information given to ER_LOCK_DEADLOCK was missing because
  ER_LOCK_DEADLOCK doesn't provide any extra information.

I kept #ifdef mysqld_error_find_printf_error_used in sql_acl.h
to make it easy to do this kind of check again in the future

trx_undo_page_report_modify(): For SPATIAL INDEX, keep logging
updated off-page columns twice, so that
the minimum bounding rectangle (MBR) will be logged.
Avoiding the redundant logging would require larger changes
to the undo log format.

row_build_index_entry_low(): Handle SPATIAL_UNKNOWN more robustly,
by refusing to purge the record from the spatial index.
We can get this code when processing old undo log from 10.2.10 or
10.2.11 (the releases affected by MDEV-14799, which was a regression
from MDEV-14051).

The InnoDB background tasks can modify tables while LOCK TABLES...WRITE
is in effect. The purge of InnoDB history always worked like this in
MariaDB, but in MySQL 5.7 it sometimes yields to LOCK TABLES.

Also, make gcol.innodb_virtual_index run the purge for an UPDATE
before DROP TABLE is executed.

This is to be able to better track where things goes wrong

Make rpl_ctype_latin1 more portable by printing names in hex
Also only run if lower_case_table_names is 0, as this affects the result

up fix for MDEV-12458

trx_undo_rec_get_partial_row(): When the PRIMARY KEY includes a
column prefix of an externally stored column, the already parsed
part of the undo log record may contain a reference to
an off-page column. This is the case in the bug58912 test in
innodb.innodb.

This is a regression caused by MDEV-14051 'Undo log record is too big.'

Purge in the secondary index is wrongly skipped in
row_purge_upd_exist_or_extern() because node->row only does not contain all
indexed columns.

trx_undo_rec_get_partial_row(): Add the parameter for node->update
so that the updated columns will be copied from the initial part
of the undo log record.

Disabled warnings for directory option as this is depending on compilation
options.

Code in QUICK_RANGE_SELECT::init_ror_merged_scan() could theoretically
have caused crashes if this was ever called from an update or delete

This also found a bug in the vcol/range.result. file.

Added a system variabe rocsdb_git_hash to MyRocks which tell us the version of RocksDB being used

Currently explain format=json does not show the order direction of fields used during filesort.
This patch would remove this limitation

             Locked_tables_list::unlock_locked_tables

Similarly to regular DROP TABLE, don't leave locked tables mode if CREATE OR
REPLACE dropped temporary table but failed to cerate new one.

The problem is that there's no track of which temporary table was "locked" by
LOCK TABLES.

Window definitions are resolved during fix fields. Updating used tables
for window functions must be done after all window functions have had a
chance to be resolved.

There was an additional problem with the implementation: expressions that
contained window functions never updated the expression's used tables.
To fix both these issues, make sure to call "update_used_tables" on all
items that contain window functions after we have passed through all
items.