added ability to do SHOW CREATE DATABASE `information_schema`

 Use open_normal_and_derived_tables instead of open_and_lock_tables when reading metadata for a table.
 Add two test cases, one for "USE database" and one for "SHOW COLUMNS FROM table"


mysql-test/r/lock_multi.result:
  Updated test results for test case for bug9998
mysql-test/r/mysql.result:
  Updated results for test case for bug9998
mysql-test/t/lock_multi.test:
  Test that "show columsn from t1" is not locked by another thread having a write lock on t1
mysql-test/t/mysql.test:
  Add test case for BUG9998
    - lock table t1 for write from mysql-test, then execute 'mysql' and call "USE test". This will test exactly what caused the bug.
sql/sql_show.cc:
  Open tables without locking when reading metadata

Ensure we get error if INSERT IGNORE ... SELECT fails
Fixed wrong key_part->key_length usage in index_merge


client/mysql.cc:
  Code cleanups & simply optimizations
mysql-test/r/information_schema.result:
  Safety
mysql-test/t/information_schema.test:
  Safety
sql/ha_ndbcluster.cc:
  Code cleanups
sql/item.cc:
  Code cleanups
sql/item_subselect.cc:
  Code cleanups
sql/item_sum.cc:
  Code cleanups
sql/opt_range.cc:
  Made get_index_only_read_time() static (instad of inline) to increase portability (function was not declared before use)
  Simple optimization
  Fixed wrong key_part->key_length usage in index_merge
  Removed not used variable n_used_covered
  Indentation fixes & comment cleanups
sql/parse_file.cc:
  Code cleanups
sql/sql_base.cc:
  Code cleanups
sql/sql_bitmap.h:
  Added missing return
sql/sql_insert.cc:
  Ensure we get error if INSERT IGNORE ... SELECT fails
sql/sql_select.cc:
  Code cleanups
sql/sql_show.cc:
  Safety fix if a LOT of errors are ignored
sql/sql_update.cc:
  Code cleanups
sql/table.cc:
  Code cleanups
sql/table.h:
  Code cleanups
sql/uniques.cc:
  Code cleanups
strings/decimal.c:
  Simple optimization
  Code cleanups

    information schema table names are always upper case table names

  "REFERENCED_TABLE_NAME", "REFERENCED_COLUMN_NAME" fields into
  KEY_COLUMN_USAGE table

sql/mysql_priv.h:
  Changed last_query_cost status variable from global to thread-local.
sql/mysqld.cc:
  Changed last_query_cost status variable from global to thread-local.
sql/sql_class.cc:
  Changed last_query_cost status variable from global to thread-local.
sql/sql_class.h:
  Changed last_query_cost status variable from global to thread-local.
sql/sql_select.cc:
  Changed last_query_cost status variable from global to thread-local.
sql/sql_show.cc:
  All SHOW_DOUBLE-type status variables are now assumed to be thread-local (there is 
  only one currently, last_query_cost, and this cset makes it thread-local)

  Rename some functions
  more fine-grained sp privileges
  make grant/revoke sp grammar less ambigious


mysql-test/r/sp-security.result:
  change test for new syntax
mysql-test/r/system_mysql_db.result:
  change test for new syntax
mysql-test/t/sp-security.test:
  change test for new syntax
scripts/mysql_create_system_tables.sh:
  now store routine_type for procs_priv
scripts/mysql_fix_privilege_tables.sql:
  now store routine_type for procs_priv
sql/item_func.cc:
  rename of function
sql/mysql_priv.h:
  rename of function
sql/sp_head.cc:
  extra arg for check_some_routine_access
sql/sql_acl.cc:
  rename of function. now handle func/proc acls seperately
sql/sql_acl.h:
  rename of function
sql/sql_parse.cc:
  rename of function
  grants for procs handled distinctly from funcs
sql/sql_show.cc:
  check_some_routine_access extra arg
sql/sql_base.cc:
  fix for build
sql/sql_yacc.yy:
  fix for build

sql/sql_show.cc:
  Use int10_to_str() instead of my_snprintf() for formatting number

  additional fix after review

sql/sql_show.cc:
  operations with grant_tables should be #ifdef-ed

   I_S table columns information is available for any user 


mysql-test/r/information_schema.result:
  Fix for bug#10261: INFORMATION_SCHEMA.COLUMNS, incomplete result for non root user
mysql-test/t/information_schema.test:
  Fix for bug#10261: INFORMATION_SCHEMA.COLUMNS, incomplete result for non root user
sql/sql_show.cc:
  Fix for bug#10261: INFORMATION_SCHEMA.COLUMNS, incomplete result for non root user

slave_skip_errors in SHOW VARIABLES. (Bug #7800)


sql/structs.h:
  Add SHOW_SLAVE_SKIP_ERRORS
sql/set_var.cc:
  Add slave_compressed_protocol, slave_load_tmpdir, slave_skip_errors to list
  of variables.
mysql-test/t/rpl_variables.test:
  Add test for additional slave-related variables
mysql-test/r/rpl_variables.result:
  Update results
sql/sql_show.cc:
  Handle showing slave_skip_errors

sql/sql_acl.cc:
  Fix for bug #9838: INFORMATION_SCHEMA.COLUMNS columns of granted views missing
    -increase grant_version in acl_init, mysql_table_grant
    -table privileges should be taken into account when we calculate column
     grants
sql/sql_show.cc:
  Fix for bug #9838: INFORMATION_SCHEMA.COLUMNS columns of granted views missing
       use 'base_name', 'file_name' because 'tables->db', 'tables->tables' could be invalid in
       case of view(derived tables).

   DEFAULT_COLLTION_NAME field is added to SCHEMATA table


mysql-test/r/information_schema.result:
  Fix for bug#8998: information_schema: Table SCHEMATA should report default collations
sql/sql_show.cc:
  Fix for bug#8998: information_schema: Table SCHEMATA should report default collations

Added option --count to mysqlshow (to show number of rows)
Fixed possible core dump in information schema


client/client_priv.h:
  --count for mysqlshow
client/mysqlshow.c:
  Added option --count to be used when the user want's number of rows per table in the output
  (We shouldn't use count(*) as default as this can be a slow operation)
mysys/my_thr_init.c:
  Correct comment
sql/ha_berkeley.cc:
  Remove not used variable
sql/ha_berkeley.h:
  Remove not used variable
sql/ha_innodb.cc:
  Remove not used function
sql/ha_ndbcluster.cc:
  false -> FALSE
  true -> TRUE
sql/handler.cc:
  Added and fixed comments
  Remove 'strange' code to remove compiler warnings (better to do things like this with attribute)
sql/item.cc:
  false -> FALSE
sql/item_cmpfunc.cc:
  Fixed indentation
sql/item_cmpfunc.h:
  marked BETWEEN as a bool function
sql/item_func.cc:
  Simple optimzation
sql/key.cc:
  Removed wrong code
sql/log.cc:
  Check result from open_index_file()
sql/mysql_priv.h:
  Simplyfy some test of netware
sql/mysqld.cc:
  Fixed indentation
  Check result form open_index_file()
  Simplify code with IF_NETWARE()
sql/opt_range.cc:
  false -> FALSE
  true -> TRUE
  Fixed indentation
sql/opt_sum.cc:
  Fixed comments
sql/sp_head.cc:
  Simple optimzation
  Move variable declarations to begining of blocks
sql/sql_acl.cc:
  Fix long lines
  Rename xx -> column
  Move declaration to beginning of block
sql/sql_parse.cc:
  Removed comment
sql/sql_select.cc:
  Indentation fixes
sql/sql_show.cc:
  Fixed reference outside of array (possible core dump)
sql/sql_table.cc:
  Simplify code
  Combine common code
sql/sql_test.cc:
  false -> FALSE
sql/sql_trigger.cc:
  false -> false
  true -> TRUE
sql/sql_yacc.yy:
  Simpler test
sql/unireg.cc:
  Added comment

Mostly about precision/decimals of the results of the operations


include/decimal.h:
  decimal interface changed a little
sql/field.cc:
  a lot of precision/decimals related changes to the Field_new_decimal
sql/field.h:
  Field_new_decimal interface changed
sql/ha_ndbcluster.cc:
  f->precision should be used here
sql/item.cc:
  precision/decimals counting related changes
sql/item.h:
  precision/decimals counting related changes
sql/item_cmpfunc.cc:
  precision/decimals counting related changes
sql/item_cmpfunc.h:
  precision/decimals counting related changes
sql/item_func.cc:
  precision/decimals counting related changes
sql/item_func.h:
  precision/decimals counting related changes
sql/item_sum.cc:
  precision/decimals counting related changes
sql/item_sum.h:
  precision/decimals counting related changes
sql/my_decimal.cc:
  precision/decimals counting related changes
sql/my_decimal.h:
  precision/decimals counting related changes
sql/mysqld.cc:
  precision/decimals counting related changes
sql/set_var.cc:
  precision/decimals counting related changes
sql/sp_head.cc:
  dbug_decimal_print was replaced with dbug_decimal_as_string
sql/sql_class.h:
  div_precincrement variable added
sql/sql_parse.cc:
  precision/decimals counting related changes
sql/sql_select.cc:
  precision/decimals counting related changes
sql/sql_show.cc:
  Field::representation_length was removed
strings/decimal.c:
  decimal_actual_fraction was introduced
BitKeeper/etc/logging_ok:
  Logging to logging@openlogging.org accepted

Changed type of "Sub_part" column in SHOW KEYS from TINYINT to SMALLINT (as MAX_KEY_LENGTH=1024)
(this is the final cset with proper tests)


mysql-test/r/show_check.result:
  Testcase for BUG#9439
mysql-test/t/show_check.test:
  Testcase for BUG#9439
sql/sql_show.cc:
  Fix for BUG#9439:
  Changed type of "Sub_part" column in SHOW KEYS from TINYINT to SMALLINT (as MAX_KEY_LENGTH=1024)

  Added a test in connection with the fix for
  bug #6106.
view.result, view.test:
  Added test cases for bugs #6106/6107.
sql_show.cc:
  The addition of the case for items of the type REF_ITEM in the
  function uses_only_table_name_fields became necessary after
  the fix for bug #6106.
sql_base.cc:
  The problem was due to the fact that two different column
  references were glued together though one of them belonged to
  a subquery while another to an outer query. This caused
  eventually a wrong calculation of values for the used_tables
  attribute.


sql/sql_base.cc:
  The problem was due to the fact that two different column
  references were glued together though one of them belonged to
  a subquery while another to an outer query. This caused
  eventually a wrong calculation of values for the used_tables
  attribute.
sql/sql_show.cc:
  The addition of the case for items of the type REF_ITEM in the
  function uses_only_table_name_fields became necessary after
  the fix for bug #6106.
mysql-test/t/view.test:
  Added test cases for bugs #6106/6107.
mysql-test/r/view.result:
  Added test cases for bugs #6106/6107.
mysql-test/t/information_schema.test:
  Added a test in connection with the fix for
  bug #6106.
mysql-test/r/information_schema.result:
  Added a test in connection with the fix for
  bug #6106.

mysql-test/r/drop.result:
  Made error message smaller to not get into trouble with clients with smaller error buffers
  The proper way to fix this is to generate a separate warning for each not found table if there was more than one table in the DROP
  but
mysql-test/r/kill.result:
  Portablity fix
mysql-test/t/kill.test:
  Portablity fix
mysys/my_sync.c:
  Fixed possible compiler warning
sql/records.cc:
  Fixed possible compiler warning
sql/share/errmsg.txt:
  Made error message smaller to not get into trouble with clients with smaller error buffers
sql/sql_class.cc:
  Don't put code in comments
sql/sql_show.cc:
  Cleanup

   restore original 'lex->query_tables' table list after
   processing of information schema table
   remove unnecessary operations


mysql-test/r/ps.result:
  Fix for bug#9383: INFORMATION_SCHEMA.COLUMNS, JOIN, Crash, prepared statement
mysql-test/t/ps.test:
  Fix for bug#9383: INFORMATION_SCHEMA.COLUMNS, JOIN, Crash, prepared statement
sql/sql_lex.h:
  Fix for bug#9383: INFORMATION_SCHEMA.COLUMNS, JOIN, Crash, prepared statement
sql/sql_show.cc:
  Fix for bug#9383: INFORMATION_SCHEMA.COLUMNS, JOIN, Crash, prepared statement

    the fix for 'columns' table only
    Rights check for 'routines' table is already done
    (see WL2131 Access control for SHOW ... PROCEDURE|FUNCTION ...)
    rename 'testtets' database to 'mysqltest' to keep number of
      DBs which created by mysql-test as small as possible


mysql-test/r/information_schema.result:
  Fix for bug#7214: information_schema: unauthorized user can see metadata
mysql-test/t/information_schema.test:
  Fix for bug#7214: information_schema: unauthorized user can see metadata
sql/sql_show.cc:
  Fix for bug#7214: information_schema: unauthorized user can see metadata

BitKeeper/etc/logging_ok:
  Logging to logging@openlogging.org accepted

  The columns CHARACTER_MAXIMUM_LENGTH and CHARACTER_OCTET_LENGTH 
  of INFORMATION_SCHEMA.COLUMNS must be NULL for numeric columns


mysql-test/r/information_schema.result:
  Fix for bug #9344 INFORMATION_SCHEMA, wrong content, numeric columns
mysql-test/t/information_schema.test:
  Fix for bug #9344 INFORMATION_SCHEMA, wrong content, numeric columns
sql/sql_show.cc:
  Fix for bug #9344 INFORMATION_SCHEMA, wrong content, numeric columns
BitKeeper/etc/logging_ok:
  Logging to logging@openlogging.org accepted

Small fixes while doing review of new pushed code
More test cases for decimal


mysql-test/r/date_formats.result:
  Warnings added for STR_TO_DATE()
mysql-test/r/ps_1general.result:
  Better description for BLACKHOLE
mysql-test/r/strict.result:
  Added tests for STR_TO_DATE
mysql-test/r/type_decimal.result:
  Test to test ranges of DECIMAL
mysql-test/t/date_formats.test:
  More tests
mysql-test/t/strict.test:
  Added tests for STR_TO_DATE
mysql-test/t/type_decimal.test:
  Test to test ranges of DECIMAL
sql/field.cc:
  Simple optimization
sql/handler.cc:
  Better description for BLACKHOLE
sql/item_timefunc.cc:
  Give warnings if wrong date/time/datetime argument for STR_TO_DATE
sql/log_event.cc:
  Indentation fixes
sql/log_event.h:
  #ifdef-ed not used code
sql/share/errmsg.txt:
  New error message
sql/sql_show.cc:
  Ensure that we do a proper restore in case of error

  -skip the print of 'MAX_ROWS=#' for 'SHOW CRETAE TABLE' with information schema tables 


mysql-test/r/information_schema.result:
  Fix for bug #8941: garbage MAX_ROWS=# from SHOW CREATE TABLE INFORMATION_SCHEMA.*
sql/sql_show.cc:
  Fix for bug #8941: garbage MAX_ROWS=# from SHOW CREATE TABLE INFORMATION_SCHEMA.*

  there was no check of result of 
  table->file->write_row() function. When count of rows was
  more than table max_rows(HEAP table) the following recodrs
  were not inserted to the table
  -'schema_table_store_record' function is added. 
    The function checks result of write_row function
    and convert HEAP table to MyISAM if necessary
  - Result check after write_row is added into all 
    I_S function which store the records to I_S tables  


mysql-test/r/information_schema.result:
  Bug #9317: Wrong count for tables in INFORMATION_SCHEMA
mysql-test/t/information_schema.test:
  Bug #9317: Wrong count for tables in INFORMATION_SCHEMA
sql/sql_select.cc:
  Bug #9317: Wrong count for tables in INFORMATION_SCHEMA
sql/sql_show.cc:
  Bug #9317: Wrong count for tables in INFORMATION_SCHEMA
sql/table.h:
  Bug #9317: Wrong count for tables in INFORMATION_SCHEMA

"After Monty's review" changes to the fix for BUG#8325 "Deadlock in replication thread stops replication":
s/sleep/safe_sleep (thread safe); sleep 0/1/2/3/4/5/5/5 (get slave less late);
no message on error log (deadlock is too common sometimes), a global counter
instead (SHOW STATUS LIKE 'slave_retried_transactions').
Plus a fix for libmysql/Makefile.shared


libmysql/Makefile.shared:
  When we "make clean" in libmysql/ we remove the symlinks there, so we
  need to mark that they have to be recreated later: this is done by removing
  ../linked_libmysql_sources. If we don't do this, 'make' will fail after 'cd libmysql;make clean'.
  This Makefile.shared is used by libmysql_r too.
  No reason to remove linked_client_sources as we don't remove the links in client/.
mysql-test/r/rpl_deadlock.result:
  result fix
mysql-test/t/rpl_deadlock.test:
  small test addition
sql/mysqld.cc:
  if active_mi could not be alloced, die. New SHOW STATUS LIKE "slave_retried_transactions".
sql/slave.cc:
  If slave retries automatically a transaction, no message on error log
  (too common situation); sleep 0 secs at first retry, then 1, 2, 3, 4,
  5, 5, 5... Sleeping 0 is to get the least possible late, as deadlocks
  are usually resolved at first try. New global counter rli->retried_trans
  (for SHOW STATUS: total number of times the slave had to retry
  any transaction). safe_sleep() is thread-safe, sleep() was not.
  I change the rli->trans_retries counter to go from 0 to max instead
  of the other way (better for new sleep()).
sql/slave.h:
  new global counter rli->retried_trans
sql/sql_show.cc:
  SHOW STATUS LIKE "slave_retried_transactions"; needs replication mutexes.
  Can't be a simple SHOW_LONG, because active_mi is unset (not alloced yet)
  when the static global status_vars is created (active_mi is set
  in init_slave()).
sql/structs.h:
  new SHOW_SLAVE_RETRIED_TRANS
BitKeeper/etc/logging_ok:
  Logging to logging@openlogging.org accepted

  New privilege CREATE USER (CREATE_USER_ACL, Create_user_priv) added
grant2.test:
  new tests (mostly backported from jani's patch)
system_mysql_db.result, sp.result, grant2.result, grant.result:
  results updated


mysql-test/r/grant.result:
  results updated
mysql-test/r/grant2.result:
  results updated
mysql-test/r/sp.result:
  results updated
mysql-test/r/system_mysql_db.result:
  results updated
mysql-test/t/grant2.test:
  new tests (mostly backported from jani's patch)
scripts/mysql_create_system_tables.sh:
  Create_user_priv added
scripts/mysql_fix_privilege_tables.sql:
  Create_user_priv added
sql/sql_acl.cc:
  Create_user_priv added
sql/sql_acl.h:
  Create_user_priv added
sql/sql_parse.cc:
  Create_user_priv added
sql/sql_show.cc:
  Create_user_priv added
sql/sql_yacc.yy:
  Create_user_priv added
BitKeeper/etc/logging_ok:
  Logging to logging@openlogging.org accepted

Removed some optional arguments
Fixed portability problem in federated tests


client/sql_string.cc:
  update from sql/sql_string.cc
client/sql_string.h:
  update from sql/sql_string.h
mysql-test/r/federated.result:
  Fixed error message
sql/field.cc:
  Cleanup during review
  Remove const in 'const unsigned int'
sql/field.h:
  Remove const in 'const unsigned int'
sql/ha_federated.cc:
  Better error string.  Add missing argument to error (before 'errno' was picked up from stack)
sql/handler.cc:
  Removed compiler warning
sql/item_func.cc:
  Cleanup during review
sql/item_sum.cc:
  Cleanup during review
sql/lock.cc:
  Remove optional arguments
sql/log_event.cc:
  Remove optional arguments
sql/mysql_priv.h:
  Remove optional arguments
  cahnge preapre_create_fields to use pointers instead of references
sql/opt_range.cc:
  Fix arguments so that return value is last
sql/sql_base.cc:
  Remove optional arguments
sql/sql_delete.cc:
  Remove optional arguments
sql/sql_error.cc:
  Remove optional arguments
sql/sql_help.cc:
  Remove optional arguments
sql/sql_parse.cc:
  Remove optional arguments
sql/sql_prepare.cc:
  Remove optional arguments
sql/sql_rename.cc:
  Remove optional arguments
sql/sql_select.cc:
  Remove optional arguments
sql/sql_show.cc:
  Cleanup during review
sql/sql_string.cc:
  Simple optimization
sql/sql_table.cc:
  Remove optional arguments
  Fixed indentation
sql/sql_update.cc:
  Remove optional arguments
sql/sql_yacc.yy:
  Change references to pointers

sql_show.cc:
  changed strdup to thd->memdup per Serg's advice


sql/sql_show.cc:
  changed strdup to thd->memdup per Serg's advice

This is a modifiction of my previous patch after receiving feedback. This is a better way to fix the problem.  With this patch, data directory and index directory will use only forward slashes (/) when on Windows.

mysqldump.c:
  Removed fixPaths routine.  Was improper fix for bug #6660
sql_show.cc:
  Changed append_directory to convert backslashes to foward slashes when on Windows.


sql/sql_show.cc:
  Changed append_directory to convert backslashes to foward slashes when on Windows.
client/mysqldump.c:
  Removed fixPaths routine.  Was improper fix for bug #6660

sql/mysql_priv.h:
  Added 'const' to some arguments
sql/sp_head.cc:
  Added comments
  code cleanup
  acceess -> access
sql/sql_acl.cc:
  Simply code by making check_routine_level_acl() available also in embedded server
sql/sql_acl.h:
  Added 'const' to some arguments
sql/sql_parse.cc:
  Added 'const' to some arguments
  Fixed the check_rounte_level_acl() is always called
  (old code didn't properly check access privilges if grant_option was not set)
sql/sql_show.cc:
  Simplify usage of get_field()
  Now we can always call check_some_routine_access()
sql/sql_udf.cc:
  Don't give warnings for suspicios UDF's if --log-warnings isn't given
sql/table.cc:
  Simplify usage of get_field()

(Found during build process)


extra/comp_err.c:
  Remove compiler warnings
extra/perror.c:
  Remove compiler warnings
innobase/dict/dict0dict.c:
  Remove compiler warnings
innobase/dict/dict0load.c:
  Remove compiler warnings
innobase/pars/pars0sym.c:
  Remove compiler warnings
innobase/row/row0row.c:
  Remove compiler warnings
innobase/row/row0sel.c:
  Remove compiler warnings
libmysqld/lib_sql.cc:
  Remove not used variables
myisam/mi_key.c:
  Remove compiler warnings
regex/engine.c:
  Added comment
sql/derror.cc:
  Remove not used variables
sql/examples/ha_archive.cc:
  Fixed bug in blob handling
  Removed not used variable
sql/field.cc:
  Remove compiler warnings
  Remove not used variables
sql/filesort.cc:
  Remove compiler warnings
sql/ha_heap.cc:
  Remove not used variable
sql/ha_innodb.cc:
  Remove not used variables
  Remove compiler warnings
sql/handler.cc:
  Remove compiler warnings and remove not used variables
sql/item.cc:
  Remove compiler warnings and remove not used variables
sql/item_subselect.cc:
  Remove compiler warnings
sql/item_sum.cc:
  Remove compiler warnings
sql/item_sum.h:
  Remove compiler warnings and remove not used variables
sql/log.cc:
  Remove compiler warnings and remove not used variables
sql/log_event.cc:
  Remove compiler warnings
sql/mysqld.cc:
  Remove compiler warnings and remove not used variables
sql/opt_range.cc:
  Remove compiler warnings and remove not used variables
sql/slave.cc:
  Remove compiler warnings and remove not used variables
sql/sp_pcontext.cc:
  Remove compiler warnings and remove not used variables
sql/sql_acl.cc:
  Remove compiler warnings and remove not used variables
sql/sql_analyse.cc:
  Remove compiler warnings and remove not used variables
sql/sql_base.cc:
  Remove compiler warnings and remove not used variables
sql/sql_db.cc:
  Remove compiler warnings and remove not used variables
sql/sql_help.cc:
  Remove compiler warnings and remove not used variables
sql/sql_insert.cc:
  Remove compiler warnings and remove not used variables
sql/sql_load.cc:
  Remove compiler warnings and remove not used variables
sql/sql_parse.cc:
  Remove compiler warnings and remove not used variables
sql/sql_prepare.cc:
  Remove compiler warnings and remove not used variables
sql/sql_select.cc:
  Remove compiler warnings and remove not used variables
sql/sql_show.cc:
  Remove compiler warnings and remove not used variables
sql/sql_table.cc:
  Remove compiler warnings
sql/sql_union.cc:
  Remove compiler warnings
sql/sql_update.cc:
  Remove compiler warnings and remove not used variables
sql/sql_yacc.yy:
  Remove compiler warnings and remove not used variables
sql/strfunc.cc:
  Remove compiler warnings and remove not used variables
strings/ctype-ucs2.c:
  Remove compiler warnings
tests/mysql_client_test.c:
  Remove compiler warnings and remove not used variables
tools/mysqlmanager.c:
  Remove compiler warnings and remove not used variables

Previously we only stored the first given error (the error sent to the client)


include/mysys_err.h:
  Moved ER_FILENOTFOUND last to avoid getting 'error 0'
mysql-test/r/backup.result:
  More warnings from in case of errors
mysql-test/r/innodb.result:
  More warnings from in case of errors
mysql-test/r/key_cache.result:
  More warnings from in case of errors
mysql-test/r/myisam.result:
  More warnings from in case of errors
mysql-test/r/preload.result:
  More warnings from in case of errors
mysql-test/r/repair.result:
  More warnings from in case of errors
mysql-test/r/rpl_failed_optimize.result:
  More warnings from in case of errors
mysql-test/r/subselect.result:
  More warnings from in case of errors
mysql-test/t/backup.test:
  Portability fix
mysql-test/t/innodb.test:
  Avoid some warnings
mysql-test/t/myisam.test:
  Portability fix
mysql-test/t/rpl_rotate_logs.test:
  Portablitiy fix
mysys/errors.c:
  Move ER_FILENOTFOUND last to avoid error 0
sql/mysql_priv.h:
  Indentation change
sql/mysqld.cc:
  Save all given errors for SHOW ERRORS
sql/protocol.cc:
  Store new errors for SHOW ERROR
  (Ignore those that are already reported with my_error())
sql/sql_class.cc:
  More errors with SHOW ERROR
sql/sql_class.h:
  More errors with SHOW ERROR
sql/sql_error.cc:
  More errors with SHOW ERROR
sql/sql_parse.cc:
  Temporary fix: Don't give error for the to-be-removed 'pre-open tables code'
sql/sql_show.cc:
  More debugging
  Don't register errors while populating schema tables.
  (If we do, we get a LOT of warnings from information_schema.test)
sql/sql_table.cc:
  Don't register errors from handler in drop table (as we already do it the upper level DROP TABLE code)
  Don't register errors for 'restore_table' as in most cases the table doesn't exists
sql/sql_union.cc:
  Use file->write_row() instead of 'write_record'.
  This gives us easer to handle error messages and is faster than the old version