manual.texi Some HANDLER section cleanup; section is still confusing, though.

We are not afraid to add extensions to SQL or support for non-SQL

features if this greatly increase the usability of MySQL Server for a big

part of our users.  (The new @code{HANDLER} interface in MySQL Server 4.0

is an example of this strategy. @xref{HANDLER}.)

is an example of this strategy. @xref{HANDLER, , @code{HANDLER}}.)

We will continue to support transactional and not transactional

databases to satisfy both heavy web/logging usage and mission critical

columns. However, columns @strong{missing} from the incoming data feed are not

touched.

@item

For tables tables with primary keys that are missing some part of the key

For tables with primary keys that are missing some part of the key

in the incoming data stream, or that have no primary key, the feed is

treated as a @code{LOAD DATA INFILE ... REPLACE INTO} is now.

@end itemize

MySQL needs at least Linux Version 2.0.

@strong{WARNING}:

We have have reports from some MySQL users that they have got serious

We have reports from some MySQL users that they have got serious

stability problems with MySQL with Linux kernel 2.2.14.  If you are

using this kernel you should upgrade to 2.2.19 (or newer) or to a 2.4

kernel.  If you have a multi-cpu box, then you should seriously consider

@code{--socket} option to specify the name of the pipe.

Note that starting from 3.23.50, named pipes are only enabled if mysqld is

started with with @code{--enable-named-pipe}.  This is because some users

started with @code{--enable-named-pipe}.  This is because some users

have experienced problems shutting down the MySQL server when one uses

named pipes.

of @code{DEFAULT}, @code{BACKUP}, @code{FORCE} or @code{QUICK}.  You can

also set this explicitly to @code{""} if you want to disable this

option. If this option is used, @code{mysqld} will on open check if the

table is marked as crashed or if if the table wasn't closed properly.

table is marked as crashed or if the table wasn't closed properly.

(The last option only works if you are running with

@code{--skip-locking}.)  If this is the case @code{mysqld} will run

check on the table. If the table was corrupted, @code{mysqld} will

statement, a dedicated table object is allocated for the thread.

This table object is not shared by other threads an will not be closed

until the thread calls @code{HANDLER table_name CLOSE} or the thread dies.

@xref{HANDLER}.  When this happens, the table is put back in the table_cache

(if it isn't full).

@xref{HANDLER, , @code{HANDLER}}.  When this happens, the table is put

back in the table_cache (if it isn't full).

You can check if your table cache is too small by checking the mysqld

variable @code{Opened_tables}.  If this is quite big, even if you

haven't done a lot of @code{FLUSH TABLES}, you should increase your table

cache. @xref{SHOW STATUS}.

cache. @xref{SHOW STATUS, , @code{SHOW STATUS}}.

@node Creating many tables,  , Table cache, Optimising Database Structure

@end itemize

If you don't use the keyword @code{ALL} for the @code{UNION}, all

returned rows will be unique, as if if you had done a @code{DISTINCT} for

returned rows will be unique, as if you had done a @code{DISTINCT} for

the total result set.  If you specify @code{ALL}, then you will get all

matching rows from all the used @code{SELECT} statements.

@subsection @code{HANDLER} Syntax

@example

HANDLER table OPEN [ AS alias ]

HANDLER table READ index @{ = | >= | <= | < @} (value1, value2, ... )

HANDLER tbl_name OPEN [ AS alias ]

HANDLER tbl_name READ index_name @{ = | >= | <= | < @} (value1,value2,...)

    [ WHERE ... ] [LIMIT ... ]

HANDLER table READ index @{ FIRST | NEXT | PREV | LAST @}

HANDLER tbl_name READ index_name @{ FIRST | NEXT | PREV | LAST @}

    [ WHERE ... ] [LIMIT ... ]

HANDLER table READ @{ FIRST | NEXT @}  [ WHERE ... ] [LIMIT ... ]

HANDLER table CLOSE

HANDLER tbl_name READ @{ FIRST | NEXT @}

    [ WHERE ... ] [LIMIT ... ]

HANDLER tbl_name CLOSE

@end example

The @code{HANDLER} statement provides direct access to the @code{MyISAM} table

@code{SELECT}.

The first form of @code{HANDLER} statement opens a table, making

in accessible via the following @code{HANDLER ... READ} routines.

This table object is not shared by other threads an will not be closed

until the thread calls @code{HANDLER table_name CLOSE} or the thread dies.

it accessible via subsequent @code{HANDLER ... READ} statements.

This table object is not shared by other threads and will not be closed

until the thread calls @code{HANDLER tbl_name CLOSE} or the thread dies.

The second form fetches one (or, specified by @code{LIMIT} clause) row

where the index specified complies to the condition and @code{WHERE}

The fourth form (without index specification) fetches one (or, specified

by @code{LIMIT} clause) row from the table in natural row order (as stored

in data file) matching @code{WHERE} condition. It is faster than

@code{HANDLER table READ index} when full table scan is desired.

@code{HANDLER tbl_name READ index_name} when a full table scan is desired.

The last form closes the table, opened with @code{HANDLER ... OPEN}.

@code{HANDLER ... CLOSE} closes a table that was opened with

@code{HANDLER ... OPEN}.

@code{HANDLER} is somewhat of a low-level statement, for example it does

not provide consistency,  that is @code{HANDLER ... OPEN} does @strong{NOT}

@code{HANDLER} is a somewhat low-level statement.  For example, it does

not provide consistency.  That is, @code{HANDLER ... OPEN} does @strong{NOT}

take a snapshot of the table, and does @strong{NOT} lock the table. This

means that after a @code{HANDLER ... OPEN} is issued, table data can be

modified (by this or an other thread) and these modifications may only

partially appear in @code{HANDLER ... NEXT} or @code{HANDLER ... PREV} scans.

modified (by this or any other thread) and these modifications may appear

only partially in @code{HANDLER ... NEXT} or @code{HANDLER ... PREV} scans.

@node INSERT, INSERT DELAYED, HANDLER, Data Manipulation

Thus user A sees the row inserted by B only when B has committed the

insert, and A has committed his own transaction so that the timepoint

is advanced past the the commit of B.

is advanced past the commit of B.

If you want to see the 'freshest' state of the database, you should use

a locking read:

If you want to see the ``freshest'' state of the database, you should

use a locking read:

@example

SELECT * FROM t LOCK IN SHARE MODE;

where an auto-increment column has been defined, and the user does not provide

an explicit value for the column, then InnoDB executes @code{SELECT

MAX(auto-inc-column) FROM T}, and assigns that value incremented

by one to the the column and the auto-increment counter of the table.

by one to the column and the auto-increment counter of the table.

We say that

the auto-increment counter for table @code{T} has been initialised.

on your operating system.

If the insertions to and index are always ascending and

records are deleted only from the end, then the the file space management

records are deleted only from the end, then the file space management

algorithm of InnoDB guarantees that fragmentation in the index will

not occur.

@multitable @columnfractions .25 .65

@item @strong{Function} @tab @strong{When to call}

@item @code{mysql_server_init()} @tab Should be called before any other other MySQL function is called, preferably early in the @code{main()} function.

@item @code{mysql_server_init()} @tab Should be called before any other MySQL function is called, preferably early in the @code{main()} function.

@item @code{mysql_server_end()} @tab Should be called before your program exits.

@item @code{mysql_thread_init()} @tab Should be called in each thread you create that will access MySQL.

@item @code{mysql_thread_end()} @tab Should be called before calling @code{pthread_exit()}

@item

The TCP/IP connection thread handles all connection requests and

creates a new dedicated thread to handle the authentication and

and SQL query processing for each connection.

SQL query processing for each connection.

@item

On Windows NT there is a named pipe handler thread that does the same work as