Author: paul
Date: 2007-02-12 02:45:58 +0100 (Mon, 12 Feb 2007)
New Revision: 4895
Log:
r16114@frost: paul | 2007-02-11 19:45:00 -0600
C API binary protocol:
Additions to/corrections of information about type codes.
Modified:
trunk/refman-4.1/apis-c.xml
trunk/refman-5.0/apis-c.xml
trunk/refman-5.1/apis-c.xml
Property changes on: trunk
___________________________________________________________________
Name: svk:merge
- 4767c598-dc10-0410-bea0-d01b485662eb:/mysqldoc-local/mysqldoc/trunk:19666
7d8d2c4e-af1d-0410-ab9f-b038ce55645b:/mysqldoc-local/mysqldoc:16111
b5ec3a16-e900-0410-9ad2-d183a3acac99:/mysqldoc-local/mysqldoc/trunk:14218
bf112a9c-6c03-0410-a055-ad865cd57414:/mysqldoc-local/mysqldoc/trunk:13520
+ 4767c598-dc10-0410-bea0-d01b485662eb:/mysqldoc-local/mysqldoc/trunk:19666
7d8d2c4e-af1d-0410-ab9f-b038ce55645b:/mysqldoc-local/mysqldoc:16114
b5ec3a16-e900-0410-9ad2-d183a3acac99:/mysqldoc-local/mysqldoc/trunk:14218
bf112a9c-6c03-0410-a055-ad865cd57414:/mysqldoc-local/mysqldoc/trunk:13520
Modified: trunk/refman-4.1/apis-c.xml
===================================================================
--- trunk/refman-4.1/apis-c.xml 2007-02-12 00:25:53 UTC (rev 4894)
+++ trunk/refman-4.1/apis-c.xml 2007-02-12 01:45:58 UTC (rev 4895)
Changed blocks: 12, Lines Added: 163, Lines Deleted: 68; 19446 bytes
@@ -714,12 +714,13 @@
<para>
To distinguish between binary and non-binary data for string
data types, check whether the <literal>charsetnr</literal>
- value is 63. If so, the character set is
- <literal>binary</literal>, which indicates binary rather than
- non-binary data. This is how to distinguish between
- <literal>BINARY</literal> and <literal>CHAR</literal>,
- <literal>VARBINARY</literal> and <literal>VARCHAR</literal>,
- and <literal>BLOB</literal> and <literal>TEXT</literal>.
+ value is 63 (see <xref linkend="c-api-datatypes"/>). If so,
+ the character set is <literal>binary</literal>, which
+ indicates binary rather than non-binary data. This is how to
+ distinguish between <literal>BINARY</literal> and
+ <literal>CHAR</literal>, <literal>VARBINARY</literal> and
+ <literal>VARCHAR</literal>, and <literal>BLOB</literal> and
+ <literal>TEXT</literal>.
</para>
</listitem>
@@ -7992,13 +7993,15 @@
</para>
<para>
- The type of the buffer. The allowable
+ The type of the buffer. This member indicates the data
+ type of the C language variable that you are binding to
+ the statement parameter. The allowable
<literal>buffer_type</literal> values are listed later in
this section. For input, <literal>buffer_type</literal>
- indicates the data type of the C language variable that
- you are binding to a statement parameter. For output, it
- indicates the SQL data type of the value that you expect
- to receive in a result buffer.
+ indicates the type of the variable containing the value
+ that you will send to the server. For output, it indicates
+ the type of the variable into which you want a value
+ received from the server to be stored.
</para>
</listitem>
@@ -8008,36 +8011,49 @@
</para>
<para>
+ A pointer to the buffer to be used for data transfer. This
+ is the address of a variable.
+ </para>
+
+ <para>
For input, <literal>buffer</literal> is a pointer to the
- buffer in which a statement parameter's data value is
+ variable in which a statement parameter's data value is
stored. When you call
<literal>mysql_stmt_execute()</literal>, MySQL takes the
- value that you have stored in the buffer and uses it in
+ value that you have stored in the variable and uses it in
place of the corresponding parameter marker in the
statement.
</para>
<para>
For output, <literal>buffer</literal> is a pointer to the
- buffer in which to return a result set column value. When
- you call <literal>mysql_stmt_fetch()</literal>, MySQL
- returns a column value and stores it in this buffer. You
+ variable in which to return a result set column value.
+ When you call <literal>mysql_stmt_fetch()</literal>, MySQL
+ returns a column value and stores it in this variable. You
can access the value when the call returns.
</para>
<para>
- For character (non-binary) and binary string data types,
- <literal>buffer</literal> should point to a character
- buffer. For numeric data types, <literal>buffer</literal>
- should point to a variable of the proper numeric C type.
- (For <literal>char</literal> or integer variables, you
- should also indicate whether the variable has the
+ To minimize the need for MySQL to perform type conversions
+ between C language values on the client side and SQL
+ values on the server side, use variables that have types
+ similar to those of the corresponding SQL values. For
+ numeric data types, <literal>buffer</literal> should point
+ to a variable of the proper numeric C type. (For
+ <literal>char</literal> or integer variables, you should
+ also indicate whether the variable has the
<literal>unsigned</literal> attribute by setting the
<literal>is_unsigned</literal> member, described later in
- this list.) For date and time data types,
+ this list.) For character (non-binary) and binary string
+ data types, <literal>buffer</literal> should point to a
+ character buffer. For date and time data types,
<literal>buffer</literal> should point to a
<literal>MYSQL_TIME</literal> structure.
</para>
+
+ <para>
+ See the notes about type conversions later in the section.
+ </para>
</listitem>
<listitem>
@@ -8132,9 +8148,11 @@
</para>
<para>
- <literal>is_null</literal> is a pointer to a boolean
- rather than a boolean scalar to provide flexibility in how
- you specify <literal>NULL</literal> values:
+ The reason that <literal>is_null</literal> is not a
+ boolean scalar but is instead a
+ <emphasis>pointer</emphasis> to a boolean scalar is to
+ provide flexibility in how you specify
+ <literal>NULL</literal> values:
</para>
<itemizedlist>
@@ -8161,7 +8179,7 @@
<listitem>
<para>
In all other cases, set the other members
- appriopriate, and set <literal>is_null</literal> to
+ appriopriately, and set <literal>is_null</literal> to
the address of a <literal>my_bool</literal> variable.
Set that variable's value to true or false
appropriately between executions to indicate whether
@@ -8190,22 +8208,23 @@
can be <literal>unsigned</literal>
(<literal>char</literal>, <literal>short int</literal>,
<literal>int</literal>, <literal>long long int</literal>).
- integer types. Set <literal>is_unsigned</literal> to true
- if the variable pointed to by <literal>buffer</literal> is
+ Set <literal>is_unsigned</literal> to true if the variable
+ pointed to by <literal>buffer</literal> is
<literal>unsigned</literal> and false otherwise. For
example, if you bind a <literal>char</literal> variable to
<literal>buffer</literal>, specify a type code of
<literal>MYSQL_TYPE_TINY</literal> and set
<literal>is_unsigned</literal> to false. If you bind an
<literal>unsigned char</literal> instead, the type code is
- the same, but <literal>is_unsigned</literal> should be
+ the same but <literal>is_unsigned</literal> should be
true.
</para>
<para>
- <literal>is_unsigned</literal> indicates nothing about the
- signedness of the SQL value that you are sending or
- receiving. For example, if you use an
+ <literal>is_unsigned</literal> applies only to the C
+ language variable on the client side. It indicates nothing
+ about the signedness of the corresponding SQL value on the
+ server side. For example, if you use an
<literal>int</literal> variable to supply a value for a
<literal>BIGINT UNSIGNED</literal> column,
<literal>is_unsigned</literal> should be false because
@@ -8330,69 +8349,87 @@
value should be chosen according to the data type of the C
language variable that you are binding. If the variable is
<literal>unsigned</literal>, you should also set the
- <literal>is_unsigned</literal> member to true.
+ <literal>is_unsigned</literal> member to true. The table shows the
+ C variable types that you can use, the corresponding type codes,
+ and the SQL data types for which the the supplied value can be
+ used without conversion.
</para>
<informaltable>
- <tgroup cols="2">
+ <tgroup cols="3">
<colspec colwidth="30*"/>
- <colspec colwidth="70*"/>
+ <colspec colwidth="40*"/>
+ <colspec colwidth="30*"/>
<tbody>
<row>
- <entry><emphasis role="bold">C Variable Type</emphasis></entry>
+ <entry><emphasis role="bold">Input Variable C Type</emphasis></entry>
<entry><literal>buffer_type</literal> <emphasis role="bold">Value</emphasis></entry>
+ <entry><emphasis role="bold">SQL Type of Destination Value</emphasis></entry>
</row>
<row>
<entry><literal>char</literal></entry>
<entry><literal>MYSQL_TYPE_TINY</literal></entry>
+ <entry><literal>TINYINT</literal></entry>
</row>
<row>
<entry><literal>short int</literal></entry>
<entry><literal>MYSQL_TYPE_SHORT</literal></entry>
+ <entry><literal>SMALLINT</literal></entry>
</row>
<row>
<entry><literal>int</literal></entry>
<entry><literal>MYSQL_TYPE_LONG</literal></entry>
+ <entry><literal>INT</literal></entry>
</row>
<row>
<entry><literal>long long int</literal></entry>
<entry><literal>MYSQL_TYPE_LONGLONG</literal></entry>
+ <entry><literal>BIGINT</literal></entry>
</row>
<row>
<entry><literal>float</literal></entry>
<entry><literal>MYSQL_TYPE_FLOAT</literal></entry>
+ <entry><literal>FLOAT</literal></entry>
</row>
<row>
<entry><literal>double</literal></entry>
<entry><literal>MYSQL_TYPE_DOUBLE</literal></entry>
+ <entry><literal>DOUBLE</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_TIME</literal></entry>
+ <entry><literal>TIME</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_DATE</literal></entry>
+ <entry><literal>DATE</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_DATETIME</literal></entry>
+ <entry><literal>DATETIME</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_TIMESTAMP</literal></entry>
+ <entry><literal>TIMESTAMP</literal></entry>
</row>
<row>
<entry><literal>char[]</literal></entry>
<entry><literal>MYSQL_TYPE_STRING</literal> (for non-binary data)</entry>
+ <entry><literal>TEXT, CHAR, VARCHAR</literal></entry>
</row>
<row>
<entry><literal>char[]</literal></entry>
<entry><literal>MYSQL_TYPE_BLOB</literal> (for binary data)</entry>
+ <entry><literal>BLOB, BINARY, VARBINARY</literal></entry>
</row>
<row>
<entry/>
<entry><literal>MYSQL_TYPE_NULL</literal></entry>
+ <entry><literal>NULL</literal></entry>
</row>
</tbody>
</tgroup>
@@ -8407,13 +8444,16 @@
The following table shows the allowable values that may be
specified in the <literal>buffer_type</literal> member of
<literal>MYSQL_BIND</literal> structures for output values. The
- value should be chosen according to the SQL data type of the value
- that you receive from the server. The table also shows the
- recommended type of C language variable to bind to the
- <literal>MYSQL_BIND</literal> structure for each SQL type. If you
- use an <literal>unsigned</literal> variable, you should also set
- the <literal>is_unsigned</literal> member to true. Note that
- <literal>DECIMAL</literal> values are returned as strings.
+ value should be chosen according to the data type of the C
+ language variable that you are binding. If the variable is
+ <literal>unsigned</literal>, you should also set the
+ <literal>is_unsigned</literal> member to true. The table shows the
+ SQL types of recieved values, the corresponding type code that
+ such values have in result set metadata, and the recommended C
+ language data types to bind to the <literal>MYSQL_BIND</literal>
+ structure to receive the SQL values without conversion. Note that
+ <literal>DECIMAL</literal> values are returned as strings, which
+ is why the corresponding C type is <literal>char[]</literal>.
</para>
<informaltable>
@@ -8423,9 +8463,9 @@
<colspec colwidth="35*"/>
<tbody>
<row>
- <entry><emphasis role="bold">SQL Type</emphasis></entry>
+ <entry><emphasis role="bold">SQL Type of Received Value</emphasis></entry>
<entry><literal>buffer_type</literal> <emphasis role="bold">Value</emphasis></entry>
- <entry><emphasis role="bold">Recommended C Type</emphasis></entry>
+ <entry><emphasis role="bold">Output Variable C Type</emphasis></entry>
</row>
<row>
<entry><literal>TINYINT</literal></entry>
@@ -8458,6 +8498,11 @@
<entry><literal>double</literal></entry>
</row>
<row>
+ <entry><literal>YEAR</literal></entry>
+ <entry><literal>MYSQL_TYPE_SHORT</literal></entry>
+ <entry><literal>short int</literal></entry>
+ </row>
+ <row>
<entry><literal>TIME</literal></entry>
<entry><literal>MYSQL_TYPE_TIME</literal></entry>
<entry><literal>MYSQL_TIME</literal></entry>
@@ -8512,23 +8557,73 @@
</informaltable>
<para>
- The C language variable types are <quote>recommended</quote>
- because implicit type conversion may be performed in both
- directions. The <literal>buffer_type</literal> value controls the
- conversion that will be performed. For example, to fetch a SQL
- <literal>MEDIUMINT</literal> column value, you can specify a
- <literal>buffer_type</literal> value of
- <literal>MYSQL_TYPE_LONG</literal> and use a C variable of type
- <literal>int</literal> as the destination buffer. If you fetch a
- numeric column with a value of 255 into a
- <literal>char[4]</literal> character array, specify a
- <literal>buffer_type</literal> value of
- <literal>MYSQL_TYPE_STRING</literal> and the resulting value in
- the array will be a 4-byte string containing
- <literal>'255\0'</literal>.
+ The C language variable types are those recommended if you want to
+ avoid type conversions. If there is a mismatch between the C
+ variable type on the client side and the corresponding SQL value
+ on the server side, MySQL performs implicit type conversions in
+ both directions.
</para>
<para>
+ MySQL knows the type code for the SQL value on the server side.
+ The <literal>buffer_type</literal> value indicates the MySQL the
+ type code of the C variable that holds the value on the client
+ side. The two codes together tell MySQL what conversion must be
+ performed, if any. Here are some examples:
+ </para>
+
+ <itemizedlist>
+
+ <listitem>
+ <para>
+ If you use <literal>MYSQL_TYPE_LONG</literal> with an
+ <literal>int</literal> variable to pass an integer value to
+ the server that is to be stored into a
+ <literal>FLOAT</literal> column, MySQL converts the value to
+ floating-point format before storing it.
+ </para>
+ </listitem>
+
+ <listitem>
+ <para>
+ To fetch a SQL <literal>MEDIUMINT</literal> column value, you
+ can specify a <literal>buffer_type</literal> value of
+ <literal>MYSQL_TYPE_LONG</literal> and use a C variable of
+ type <literal>int</literal> as the destination buffer. MySQL
+ will convert the <literal>MEDIUMINT</literal> value (a 3-byte
+ value) for storage into the <literal>int</literal> (a 4-byte
+ variable).
+ </para>
+ </listitem>
+
+ <listitem>
+ <para>
+ If you fetch a numeric column with a value of 255 into a
+ <literal>char[4]</literal> character array, specify a
+ <literal>buffer_type</literal> value of
+ <literal>MYSQL_TYPE_STRING</literal> and the resulting value
+ in the array will be a 4-byte string containing
+ <literal>'255\0'</literal>.
+ </para>
+ </listitem>
+
+ </itemizedlist>
+
+ <para>
+ Before binding variables to the <literal>MYSQL_BIND</literal>
+ structures that are to be used for fetching the column values, you
+ can check the type codes for the columns of a result set. This
+ might be desirable if you want to determine which variable types
+ would be best to use to avoid type conversions. To get the type
+ codes, call <literal>mysql_stmt_result_metadata()</literal> after
+ executing the prepared statement with
+ <literal>mysql_stmt_execute()</literal>. The metadata provides
+ access to the type codes for the result set as described in
+ <xref linkend="mysql-stmt-result-metadata"/>, and
+ <xref linkend="c-api-datatypes"/>.
+ </para>
+
+ <para>
For input character (non-binary) string data (indicated by
<literal>MYSQL_TYPE_STRING</literal>), the value is assumed to be
in the character set indicated by the
@@ -8542,15 +8637,15 @@
</para>
<para>
- To determine whether string values in a result set returned from
- the server contain binary or non-binary data, check whether the
- <literal>charsetnr</literal> value of the result set metadata is
- 63. If so, the character set is <literal>binary</literal>, which
- indicates binary rather than non-binary data. This enables you to
- distinguish between <literal>BINARY</literal> and
+ To determine whether output string values in a result set returned
+ from the server contain binary or non-binary data, check whether
+ the <literal>charsetnr</literal> value of the result set metadata
+ is 63. If so, the character set is <literal>binary</literal>,
+ which indicates binary rather than non-binary data. This enables
+ you to distinguish between <literal>BINARY</literal> and
<literal>CHAR</literal>, <literal>VARBINARY</literal> and
- <literal>VARCHAR</literal>, and <literal>BLOB</literal> and
- <literal>TEXT</literal>.
+ <literal>VARCHAR</literal>, and the <literal>BLOB</literal> and
+ <literal>TEXT</literal> types.
</para>
</section>
Modified: trunk/refman-5.0/apis-c.xml
===================================================================
--- trunk/refman-5.0/apis-c.xml 2007-02-12 00:25:53 UTC (rev 4894)
+++ trunk/refman-5.0/apis-c.xml 2007-02-12 01:45:58 UTC (rev 4895)
Changed blocks: 12, Lines Added: 163, Lines Deleted: 68; 19442 bytes
@@ -713,12 +713,13 @@
<para>
To distinguish between binary and non-binary data for string
data types, check whether the <literal>charsetnr</literal>
- value is 63. If so, the character set is
- <literal>binary</literal>, which indicates binary rather than
- non-binary data. This is how to distinguish between
- <literal>BINARY</literal> and <literal>CHAR</literal>,
- <literal>VARBINARY</literal> and <literal>VARCHAR</literal>,
- and <literal>BLOB</literal> and <literal>TEXT</literal>.
+ value is 63 (see <xref linkend="c-api-datatypes"/>). If so,
+ the character set is <literal>binary</literal>, which
+ indicates binary rather than non-binary data. This is how to
+ distinguish between <literal>BINARY</literal> and
+ <literal>CHAR</literal>, <literal>VARBINARY</literal> and
+ <literal>VARCHAR</literal>, and <literal>BLOB</literal> and
+ <literal>TEXT</literal>.
</para>
</listitem>
@@ -8083,13 +8084,15 @@
</para>
<para>
- The type of the buffer. The allowable
+ The type of the buffer. This member indicates the data
+ type of the C language variable that you are binding to
+ the statement parameter. The allowable
<literal>buffer_type</literal> values are listed later in
this section. For input, <literal>buffer_type</literal>
- indicates the data type of the C language variable that
- you are binding to a statement parameter. For output, it
- indicates the SQL data type of the value that you expect
- to receive in a result buffer.
+ indicates the type of the variable containing the value
+ that you will send to the server. For output, it indicates
+ the type of the variable into which you want a value
+ received from the server to be stored.
</para>
</listitem>
@@ -8099,36 +8102,49 @@
</para>
<para>
+ A pointer to the buffer to be used for data transfer. This
+ is the address of a variable.
+ </para>
+
+ <para>
For input, <literal>buffer</literal> is a pointer to the
- buffer in which a statement parameter's data value is
+ variable in which a statement parameter's data value is
stored. When you call
<literal>mysql_stmt_execute()</literal>, MySQL takes the
- value that you have stored in the buffer and uses it in
+ value that you have stored in the variable and uses it in
place of the corresponding parameter marker in the
statement.
</para>
<para>
For output, <literal>buffer</literal> is a pointer to the
- buffer in which to return a result set column value. When
- you call <literal>mysql_stmt_fetch()</literal>, MySQL
- returns a column value and stores it in this buffer. You
+ variable in which to return a result set column value.
+ When you call <literal>mysql_stmt_fetch()</literal>, MySQL
+ returns a column value and stores it in this variable. You
can access the value when the call returns.
</para>
<para>
- For character (non-binary) and binary string data types,
- <literal>buffer</literal> should point to a character
- buffer. For numeric data types, <literal>buffer</literal>
- should point to a variable of the proper numeric C type.
- (For <literal>char</literal> or integer variables, you
- should also indicate whether the variable has the
+ To minimize the need for MySQL to perform type conversions
+ between C language values on the client side and SQL
+ values on the server side, use variables that have types
+ similar to those of the corresponding SQL values. For
+ numeric data types, <literal>buffer</literal> should point
+ to a variable of the proper numeric C type. (For
+ <literal>char</literal> or integer variables, you should
+ also indicate whether the variable has the
<literal>unsigned</literal> attribute by setting the
<literal>is_unsigned</literal> member, described later in
- this list.) For date and time data types,
+ this list.) For character (non-binary) and binary string
+ data types, <literal>buffer</literal> should point to a
+ character buffer. For date and time data types,
<literal>buffer</literal> should point to a
<literal>MYSQL_TIME</literal> structure.
</para>
+
+ <para>
+ See the notes about type conversions later in the section.
+ </para>
</listitem>
<listitem>
@@ -8223,9 +8239,11 @@
</para>
<para>
- <literal>is_null</literal> is a pointer to a boolean
- rather than a boolean scalar to provide flexibility in how
- you specify <literal>NULL</literal> values:
+ The reason that <literal>is_null</literal> is not a
+ boolean scalar but is instead a
+ <emphasis>pointer</emphasis> to a boolean scalar is to
+ provide flexibility in how you specify
+ <literal>NULL</literal> values:
</para>
<itemizedlist>
@@ -8252,7 +8270,7 @@
<listitem>
<para>
In all other cases, set the other members
- appriopriate, and set <literal>is_null</literal> to
+ appriopriately, and set <literal>is_null</literal> to
the address of a <literal>my_bool</literal> variable.
Set that variable's value to true or false
appropriately between executions to indicate whether
@@ -8281,22 +8299,23 @@
can be <literal>unsigned</literal>
(<literal>char</literal>, <literal>short int</literal>,
<literal>int</literal>, <literal>long long int</literal>).
- integer types. Set <literal>is_unsigned</literal> to true
- if the variable pointed to by <literal>buffer</literal> is
+ Set <literal>is_unsigned</literal> to true if the variable
+ pointed to by <literal>buffer</literal> is
<literal>unsigned</literal> and false otherwise. For
example, if you bind a <literal>char</literal> variable to
<literal>buffer</literal>, specify a type code of
<literal>MYSQL_TYPE_TINY</literal> and set
<literal>is_unsigned</literal> to false. If you bind an
<literal>unsigned char</literal> instead, the type code is
- the same, but <literal>is_unsigned</literal> should be
+ the same but <literal>is_unsigned</literal> should be
true.
</para>
<para>
- <literal>is_unsigned</literal> indicates nothing about the
- signedness of the SQL value that you are sending or
- receiving. For example, if you use an
+ <literal>is_unsigned</literal> applies only to the C
+ language variable on the client side. It indicates nothing
+ about the signedness of the corresponding SQL value on the
+ server side. For example, if you use an
<literal>int</literal> variable to supply a value for a
<literal>BIGINT UNSIGNED</literal> column,
<literal>is_unsigned</literal> should be false because
@@ -8446,69 +8465,87 @@
value should be chosen according to the data type of the C
language variable that you are binding. If the variable is
<literal>unsigned</literal>, you should also set the
- <literal>is_unsigned</literal> member to true.
+ <literal>is_unsigned</literal> member to true. The table shows the
+ C variable types that you can use, the corresponding type codes,
+ and the SQL data types for which the the supplied value can be
+ used without conversion.
</para>
<informaltable>
- <tgroup cols="2">
+ <tgroup cols="3">
<colspec colwidth="30*"/>
- <colspec colwidth="70*"/>
+ <colspec colwidth="40*"/>
+ <colspec colwidth="30*"/>
<tbody>
<row>
- <entry><emphasis role="bold">C Variable Type</emphasis></entry>
+ <entry><emphasis role="bold">Input Variable C Type</emphasis></entry>
<entry><literal>buffer_type</literal> <emphasis role="bold">Value</emphasis></entry>
+ <entry><emphasis role="bold">SQL Type of Destination Value</emphasis></entry>
</row>
<row>
<entry><literal>char</literal></entry>
<entry><literal>MYSQL_TYPE_TINY</literal></entry>
+ <entry><literal>TINYINT</literal></entry>
</row>
<row>
<entry><literal>short int</literal></entry>
<entry><literal>MYSQL_TYPE_SHORT</literal></entry>
+ <entry><literal>SMALLINT</literal></entry>
</row>
<row>
<entry><literal>int</literal></entry>
<entry><literal>MYSQL_TYPE_LONG</literal></entry>
+ <entry><literal>INT</literal></entry>
</row>
<row>
<entry><literal>long long int</literal></entry>
<entry><literal>MYSQL_TYPE_LONGLONG</literal></entry>
+ <entry><literal>BIGINT</literal></entry>
</row>
<row>
<entry><literal>float</literal></entry>
<entry><literal>MYSQL_TYPE_FLOAT</literal></entry>
+ <entry><literal>FLOAT</literal></entry>
</row>
<row>
<entry><literal>double</literal></entry>
<entry><literal>MYSQL_TYPE_DOUBLE</literal></entry>
+ <entry><literal>DOUBLE</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_TIME</literal></entry>
+ <entry><literal>TIME</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_DATE</literal></entry>
+ <entry><literal>DATE</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_DATETIME</literal></entry>
+ <entry><literal>DATETIME</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_TIMESTAMP</literal></entry>
+ <entry><literal>TIMESTAMP</literal></entry>
</row>
<row>
<entry><literal>char[]</literal></entry>
<entry><literal>MYSQL_TYPE_STRING</literal> (for non-binary data)</entry>
+ <entry><literal>TEXT, CHAR, VARCHAR</literal></entry>
</row>
<row>
<entry><literal>char[]</literal></entry>
<entry><literal>MYSQL_TYPE_BLOB</literal> (for binary data)</entry>
+ <entry><literal>BLOB, BINARY, VARBINARY</literal></entry>
</row>
<row>
<entry/>
<entry><literal>MYSQL_TYPE_NULL</literal></entry>
+ <entry><literal>NULL</literal></entry>
</row>
</tbody>
</tgroup>
@@ -8523,13 +8560,16 @@
The following table shows the allowable values that may be
specified in the <literal>buffer_type</literal> member of
<literal>MYSQL_BIND</literal> structures for output values. The
- value should be chosen according to the SQL data type of the value
- that you receive from the server. The table also shows the
- recommended type of C language variable to bind to the
- <literal>MYSQL_BIND</literal> structure for each SQL type. If you
- use an <literal>unsigned</literal> variable, you should also set
- the <literal>is_unsigned</literal> member to true. Note that
- <literal>DECIMAL</literal> values are returned as strings.
+ value should be chosen according to the data type of the C
+ language variable that you are binding. If the variable is
+ <literal>unsigned</literal>, you should also set the
+ <literal>is_unsigned</literal> member to true. The table shows the
+ SQL types of recieved values, the corresponding type code that
+ such values have in result set metadata, and the recommended C
+ language data types to bind to the <literal>MYSQL_BIND</literal>
+ structure to receive the SQL values without conversion. Note that
+ <literal>DECIMAL</literal> values are returned as strings, which
+ is why the corresponding C type is <literal>char[]</literal>.
</para>
<informaltable>
@@ -8539,9 +8579,9 @@
<colspec colwidth="35*"/>
<tbody>
<row>
- <entry><emphasis role="bold">SQL Type</emphasis></entry>
+ <entry><emphasis role="bold">SQL Type of Received Value</emphasis></entry>
<entry><literal>buffer_type</literal> <emphasis role="bold">Value</emphasis></entry>
- <entry><emphasis role="bold">Recommended C Type</emphasis></entry>
+ <entry><emphasis role="bold">Output Variable C Type</emphasis></entry>
</row>
<row>
<entry><literal>BIT</literal></entry>
@@ -8584,6 +8624,11 @@
<entry><literal>char[]</literal></entry>
</row>
<row>
+ <entry><literal>YEAR</literal></entry>
+ <entry><literal>MYSQL_TYPE_SHORT</literal></entry>
+ <entry><literal>short int</literal></entry>
+ </row>
+ <row>
<entry><literal>TIME</literal></entry>
<entry><literal>MYSQL_TYPE_TIME</literal></entry>
<entry><literal>MYSQL_TIME</literal></entry>
@@ -8638,23 +8683,73 @@
</informaltable>
<para>
- The C language variable types are <quote>recommended</quote>
- because implicit type conversion may be performed in both
- directions. The <literal>buffer_type</literal> value controls the
- conversion that will be performed. For example, to fetch a SQL
- <literal>MEDIUMINT</literal> column value, you can specify a
- <literal>buffer_type</literal> value of
- <literal>MYSQL_TYPE_LONG</literal> and use a C variable of type
- <literal>int</literal> as the destination buffer. If you fetch a
- numeric column with a value of 255 into a
- <literal>char[4]</literal> character array, specify a
- <literal>buffer_type</literal> value of
- <literal>MYSQL_TYPE_STRING</literal> and the resulting value in
- the array will be a 4-byte string containing
- <literal>'255\0'</literal>.
+ The C language variable types are those recommended if you want to
+ avoid type conversions. If there is a mismatch between the C
+ variable type on the client side and the corresponding SQL value
+ on the server side, MySQL performs implicit type conversions in
+ both directions.
</para>
<para>
+ MySQL knows the type code for the SQL value on the server side.
+ The <literal>buffer_type</literal> value indicates the MySQL the
+ type code of the C variable that holds the value on the client
+ side. The two codes together tell MySQL what conversion must be
+ performed, if any. Here are some examples:
+ </para>
+
+ <itemizedlist>
+
+ <listitem>
+ <para>
+ If you use <literal>MYSQL_TYPE_LONG</literal> with an
+ <literal>int</literal> variable to pass an integer value to
+ the server that is to be stored into a
+ <literal>FLOAT</literal> column, MySQL converts the value to
+ floating-point format before storing it.
+ </para>
+ </listitem>
+
+ <listitem>
+ <para>
+ To fetch a SQL <literal>MEDIUMINT</literal> column value, you
+ can specify a <literal>buffer_type</literal> value of
+ <literal>MYSQL_TYPE_LONG</literal> and use a C variable of
+ type <literal>int</literal> as the destination buffer. MySQL
+ will convert the <literal>MEDIUMINT</literal> value (a 3-byte
+ value) for storage into the <literal>int</literal> (a 4-byte
+ variable).
+ </para>
+ </listitem>
+
+ <listitem>
+ <para>
+ If you fetch a numeric column with a value of 255 into a
+ <literal>char[4]</literal> character array, specify a
+ <literal>buffer_type</literal> value of
+ <literal>MYSQL_TYPE_STRING</literal> and the resulting value
+ in the array will be a 4-byte string containing
+ <literal>'255\0'</literal>.
+ </para>
+ </listitem>
+
+ </itemizedlist>
+
+ <para>
+ Before binding variables to the <literal>MYSQL_BIND</literal>
+ structures that are to be used for fetching the column values, you
+ can check the type codes for the columns of a result set. This
+ might be desirable if you want to determine which variable types
+ would be best to use to avoid type conversions. To get the type
+ codes, call <literal>mysql_stmt_result_metadata()</literal> after
+ executing the prepared statement with
+ <literal>mysql_stmt_execute()</literal>. The metadata provides
+ access to the type codes for the result set as described in
+ <xref linkend="mysql-stmt-result-metadata"/>, and
+ <xref linkend="c-api-datatypes"/>.
+ </para>
+
+ <para>
For input character (non-binary) string data (indicated by
<literal>MYSQL_TYPE_STRING</literal>), the value is assumed to be
in the character set indicated by the
@@ -8668,15 +8763,15 @@
</para>
<para>
- To determine whether string values in a result set returned from
- the server contain binary or non-binary data, check whether the
- <literal>charsetnr</literal> value of the result set metadata is
- 63. If so, the character set is <literal>binary</literal>, which
- indicates binary rather than non-binary data. This enables you to
- distinguish between <literal>BINARY</literal> and
+ To determine whether output string values in a result set returned
+ from the server contain binary or non-binary data, check whether
+ the <literal>charsetnr</literal> value of the result set metadata
+ is 63. If so, the character set is <literal>binary</literal>,
+ which indicates binary rather than non-binary data. This enables
+ you to distinguish between <literal>BINARY</literal> and
<literal>CHAR</literal>, <literal>VARBINARY</literal> and
- <literal>VARCHAR</literal>, and <literal>BLOB</literal> and
- <literal>TEXT</literal>.
+ <literal>VARCHAR</literal>, and the <literal>BLOB</literal> and
+ <literal>TEXT</literal> types.
</para>
</section>
Modified: trunk/refman-5.1/apis-c.xml
===================================================================
--- trunk/refman-5.1/apis-c.xml 2007-02-12 00:25:53 UTC (rev 4894)
+++ trunk/refman-5.1/apis-c.xml 2007-02-12 01:45:58 UTC (rev 4895)
Changed blocks: 12, Lines Added: 163, Lines Deleted: 68; 19442 bytes
@@ -712,12 +712,13 @@
<para>
To distinguish between binary and non-binary data for string
data types, check whether the <literal>charsetnr</literal>
- value is 63. If so, the character set is
- <literal>binary</literal>, which indicates binary rather than
- non-binary data. This is how to distinguish between
- <literal>BINARY</literal> and <literal>CHAR</literal>,
- <literal>VARBINARY</literal> and <literal>VARCHAR</literal>,
- and <literal>BLOB</literal> and <literal>TEXT</literal>.
+ value is 63 (see <xref linkend="c-api-datatypes"/>). If so,
+ the character set is <literal>binary</literal>, which
+ indicates binary rather than non-binary data. This is how to
+ distinguish between <literal>BINARY</literal> and
+ <literal>CHAR</literal>, <literal>VARBINARY</literal> and
+ <literal>VARCHAR</literal>, and <literal>BLOB</literal> and
+ <literal>TEXT</literal>.
</para>
</listitem>
@@ -8202,13 +8203,15 @@
</para>
<para>
- The type of the buffer. The allowable
+ The type of the buffer. This member indicates the data
+ type of the C language variable that you are binding to
+ the statement parameter. The allowable
<literal>buffer_type</literal> values are listed later in
this section. For input, <literal>buffer_type</literal>
- indicates the data type of the C language variable that
- you are binding to a statement parameter. For output, it
- indicates the SQL data type of the value that you expect
- to receive in a result buffer.
+ indicates the type of the variable containing the value
+ that you will send to the server. For output, it indicates
+ the type of the variable into which you want a value
+ received from the server to be stored.
</para>
</listitem>
@@ -8218,36 +8221,49 @@
</para>
<para>
+ A pointer to the buffer to be used for data transfer. This
+ is the address of a variable.
+ </para>
+
+ <para>
For input, <literal>buffer</literal> is a pointer to the
- buffer in which a statement parameter's data value is
+ variable in which a statement parameter's data value is
stored. When you call
<literal>mysql_stmt_execute()</literal>, MySQL takes the
- value that you have stored in the buffer and uses it in
+ value that you have stored in the variable and uses it in
place of the corresponding parameter marker in the
statement.
</para>
<para>
For output, <literal>buffer</literal> is a pointer to the
- buffer in which to return a result set column value. When
- you call <literal>mysql_stmt_fetch()</literal>, MySQL
- returns a column value and stores it in this buffer. You
+ variable in which to return a result set column value.
+ When you call <literal>mysql_stmt_fetch()</literal>, MySQL
+ returns a column value and stores it in this variable. You
can access the value when the call returns.
</para>
<para>
- For character (non-binary) and binary string data types,
- <literal>buffer</literal> should point to a character
- buffer. For numeric data types, <literal>buffer</literal>
- should point to a variable of the proper numeric C type.
- (For <literal>char</literal> or integer variables, you
- should also indicate whether the variable has the
+ To minimize the need for MySQL to perform type conversions
+ between C language values on the client side and SQL
+ values on the server side, use variables that have types
+ similar to those of the corresponding SQL values. For
+ numeric data types, <literal>buffer</literal> should point
+ to a variable of the proper numeric C type. (For
+ <literal>char</literal> or integer variables, you should
+ also indicate whether the variable has the
<literal>unsigned</literal> attribute by setting the
<literal>is_unsigned</literal> member, described later in
- this list.) For date and time data types,
+ this list.) For character (non-binary) and binary string
+ data types, <literal>buffer</literal> should point to a
+ character buffer. For date and time data types,
<literal>buffer</literal> should point to a
<literal>MYSQL_TIME</literal> structure.
</para>
+
+ <para>
+ See the notes about type conversions later in the section.
+ </para>
</listitem>
<listitem>
@@ -8342,9 +8358,11 @@
</para>
<para>
- <literal>is_null</literal> is a pointer to a boolean
- rather than a boolean scalar to provide flexibility in how
- you specify <literal>NULL</literal> values:
+ The reason that <literal>is_null</literal> is not a
+ boolean scalar but is instead a
+ <emphasis>pointer</emphasis> to a boolean scalar is to
+ provide flexibility in how you specify
+ <literal>NULL</literal> values:
</para>
<itemizedlist>
@@ -8371,7 +8389,7 @@
<listitem>
<para>
In all other cases, set the other members
- appriopriate, and set <literal>is_null</literal> to
+ appriopriately, and set <literal>is_null</literal> to
the address of a <literal>my_bool</literal> variable.
Set that variable's value to true or false
appropriately between executions to indicate whether
@@ -8400,22 +8418,23 @@
can be <literal>unsigned</literal>
(<literal>char</literal>, <literal>short int</literal>,
<literal>int</literal>, <literal>long long int</literal>).
- integer types. Set <literal>is_unsigned</literal> to true
- if the variable pointed to by <literal>buffer</literal> is
+ Set <literal>is_unsigned</literal> to true if the variable
+ pointed to by <literal>buffer</literal> is
<literal>unsigned</literal> and false otherwise. For
example, if you bind a <literal>char</literal> variable to
<literal>buffer</literal>, specify a type code of
<literal>MYSQL_TYPE_TINY</literal> and set
<literal>is_unsigned</literal> to false. If you bind an
<literal>unsigned char</literal> instead, the type code is
- the same, but <literal>is_unsigned</literal> should be
+ the same but <literal>is_unsigned</literal> should be
true.
</para>
<para>
- <literal>is_unsigned</literal> indicates nothing about the
- signedness of the SQL value that you are sending or
- receiving. For example, if you use an
+ <literal>is_unsigned</literal> applies only to the C
+ language variable on the client side. It indicates nothing
+ about the signedness of the corresponding SQL value on the
+ server side. For example, if you use an
<literal>int</literal> variable to supply a value for a
<literal>BIGINT UNSIGNED</literal> column,
<literal>is_unsigned</literal> should be false because
@@ -8564,69 +8583,87 @@
value should be chosen according to the data type of the C
language variable that you are binding. If the variable is
<literal>unsigned</literal>, you should also set the
- <literal>is_unsigned</literal> member to true.
+ <literal>is_unsigned</literal> member to true. The table shows the
+ C variable types that you can use, the corresponding type codes,
+ and the SQL data types for which the the supplied value can be
+ used without conversion.
</para>
<informaltable>
- <tgroup cols="2">
+ <tgroup cols="3">
<colspec colwidth="30*"/>
- <colspec colwidth="70*"/>
+ <colspec colwidth="40*"/>
+ <colspec colwidth="30*"/>
<tbody>
<row>
- <entry><emphasis role="bold">C Variable Type</emphasis></entry>
+ <entry><emphasis role="bold">Input Variable C Type</emphasis></entry>
<entry><literal>buffer_type</literal> <emphasis role="bold">Value</emphasis></entry>
+ <entry><emphasis role="bold">SQL Type of Destination Value</emphasis></entry>
</row>
<row>
<entry><literal>char</literal></entry>
<entry><literal>MYSQL_TYPE_TINY</literal></entry>
+ <entry><literal>TINYINT</literal></entry>
</row>
<row>
<entry><literal>short int</literal></entry>
<entry><literal>MYSQL_TYPE_SHORT</literal></entry>
+ <entry><literal>SMALLINT</literal></entry>
</row>
<row>
<entry><literal>int</literal></entry>
<entry><literal>MYSQL_TYPE_LONG</literal></entry>
+ <entry><literal>INT</literal></entry>
</row>
<row>
<entry><literal>long long int</literal></entry>
<entry><literal>MYSQL_TYPE_LONGLONG</literal></entry>
+ <entry><literal>BIGINT</literal></entry>
</row>
<row>
<entry><literal>float</literal></entry>
<entry><literal>MYSQL_TYPE_FLOAT</literal></entry>
+ <entry><literal>FLOAT</literal></entry>
</row>
<row>
<entry><literal>double</literal></entry>
<entry><literal>MYSQL_TYPE_DOUBLE</literal></entry>
+ <entry><literal>DOUBLE</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_TIME</literal></entry>
+ <entry><literal>TIME</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_DATE</literal></entry>
+ <entry><literal>DATE</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_DATETIME</literal></entry>
+ <entry><literal>DATETIME</literal></entry>
</row>
<row>
<entry><literal>MYSQL_TIME</literal></entry>
<entry><literal>MYSQL_TYPE_TIMESTAMP</literal></entry>
+ <entry><literal>TIMESTAMP</literal></entry>
</row>
<row>
<entry><literal>char[]</literal></entry>
<entry><literal>MYSQL_TYPE_STRING</literal> (for non-binary data)</entry>
+ <entry><literal>TEXT, CHAR, VARCHAR</literal></entry>
</row>
<row>
<entry><literal>char[]</literal></entry>
<entry><literal>MYSQL_TYPE_BLOB</literal> (for binary data)</entry>
+ <entry><literal>BLOB, BINARY, VARBINARY</literal></entry>
</row>
<row>
<entry/>
<entry><literal>MYSQL_TYPE_NULL</literal></entry>
+ <entry><literal>NULL</literal></entry>
</row>
</tbody>
</tgroup>
@@ -8641,13 +8678,16 @@
The following table shows the allowable values that may be
specified in the <literal>buffer_type</literal> member of
<literal>MYSQL_BIND</literal> structures for output values. The
- value should be chosen according to the SQL data type of the value
- that you receive from the server. The table also shows the
- recommended type of C language variable to bind to the
- <literal>MYSQL_BIND</literal> structure for each SQL type. If you
- use an <literal>unsigned</literal> variable, you should also set
- the <literal>is_unsigned</literal> member to true. Note that
- <literal>DECIMAL</literal> values are returned as strings.
+ value should be chosen according to the data type of the C
+ language variable that you are binding. If the variable is
+ <literal>unsigned</literal>, you should also set the
+ <literal>is_unsigned</literal> member to true. The table shows the
+ SQL types of recieved values, the corresponding type code that
+ such values have in result set metadata, and the recommended C
+ language data types to bind to the <literal>MYSQL_BIND</literal>
+ structure to receive the SQL values without conversion. Note that
+ <literal>DECIMAL</literal> values are returned as strings, which
+ is why the corresponding C type is <literal>char[]</literal>.
</para>
<informaltable>
@@ -8657,9 +8697,9 @@
<colspec colwidth="35*"/>
<tbody>
<row>
- <entry><emphasis role="bold">SQL Type</emphasis></entry>
+ <entry><emphasis role="bold">SQL Type of Received Value</emphasis></entry>
<entry><literal>buffer_type</literal> <emphasis role="bold">Value</emphasis></entry>
- <entry><emphasis role="bold">Recommended C Type</emphasis></entry>
+ <entry><emphasis role="bold">Output Variable C Type</emphasis></entry>
</row>
<row>
<entry><literal>BIT</literal></entry>
@@ -8702,6 +8742,11 @@
<entry><literal>char[]</literal></entry>
</row>
<row>
+ <entry><literal>YEAR</literal></entry>
+ <entry><literal>MYSQL_TYPE_SHORT</literal></entry>
+ <entry><literal>short int</literal></entry>
+ </row>
+ <row>
<entry><literal>TIME</literal></entry>
<entry><literal>MYSQL_TYPE_TIME</literal></entry>
<entry><literal>MYSQL_TIME</literal></entry>
@@ -8756,23 +8801,73 @@
</informaltable>
<para>
- The C language variable types are <quote>recommended</quote>
- because implicit type conversion may be performed in both
- directions. The <literal>buffer_type</literal> value controls the
- conversion that will be performed. For example, to fetch a SQL
- <literal>MEDIUMINT</literal> column value, you can specify a
- <literal>buffer_type</literal> value of
- <literal>MYSQL_TYPE_LONG</literal> and use a C variable of type
- <literal>int</literal> as the destination buffer. If you fetch a
- numeric column with a value of 255 into a
- <literal>char[4]</literal> character array, specify a
- <literal>buffer_type</literal> value of
- <literal>MYSQL_TYPE_STRING</literal> and the resulting value in
- the array will be a 4-byte string containing
- <literal>'255\0'</literal>.
+ The C language variable types are those recommended if you want to
+ avoid type conversions. If there is a mismatch between the C
+ variable type on the client side and the corresponding SQL value
+ on the server side, MySQL performs implicit type conversions in
+ both directions.
</para>
<para>
+ MySQL knows the type code for the SQL value on the server side.
+ The <literal>buffer_type</literal> value indicates the MySQL the
+ type code of the C variable that holds the value on the client
+ side. The two codes together tell MySQL what conversion must be
+ performed, if any. Here are some examples:
+ </para>
+
+ <itemizedlist>
+
+ <listitem>
+ <para>
+ If you use <literal>MYSQL_TYPE_LONG</literal> with an
+ <literal>int</literal> variable to pass an integer value to
+ the server that is to be stored into a
+ <literal>FLOAT</literal> column, MySQL converts the value to
+ floating-point format before storing it.
+ </para>
+ </listitem>
+
+ <listitem>
+ <para>
+ To fetch a SQL <literal>MEDIUMINT</literal> column value, you
+ can specify a <literal>buffer_type</literal> value of
+ <literal>MYSQL_TYPE_LONG</literal> and use a C variable of
+ type <literal>int</literal> as the destination buffer. MySQL
+ will convert the <literal>MEDIUMINT</literal> value (a 3-byte
+ value) for storage into the <literal>int</literal> (a 4-byte
+ variable).
+ </para>
+ </listitem>
+
+ <listitem>
+ <para>
+ If you fetch a numeric column with a value of 255 into a
+ <literal>char[4]</literal> character array, specify a
+ <literal>buffer_type</literal> value of
+ <literal>MYSQL_TYPE_STRING</literal> and the resulting value
+ in the array will be a 4-byte string containing
+ <literal>'255\0'</literal>.
+ </para>
+ </listitem>
+
+ </itemizedlist>
+
+ <para>
+ Before binding variables to the <literal>MYSQL_BIND</literal>
+ structures that are to be used for fetching the column values, you
+ can check the type codes for the columns of a result set. This
+ might be desirable if you want to determine which variable types
+ would be best to use to avoid type conversions. To get the type
+ codes, call <literal>mysql_stmt_result_metadata()</literal> after
+ executing the prepared statement with
+ <literal>mysql_stmt_execute()</literal>. The metadata provides
+ access to the type codes for the result set as described in
+ <xref linkend="mysql-stmt-result-metadata"/>, and
+ <xref linkend="c-api-datatypes"/>.
+ </para>
+
+ <para>
For input character (non-binary) string data (indicated by
<literal>MYSQL_TYPE_STRING</literal>), the value is assumed to be
in the character set indicated by the
@@ -8786,15 +8881,15 @@
</para>
<para>
- To determine whether string values in a result set returned from
- the server contain binary or non-binary data, check whether the
- <literal>charsetnr</literal> value of the result set metadata is
- 63. If so, the character set is <literal>binary</literal>, which
- indicates binary rather than non-binary data. This enables you to
- distinguish between <literal>BINARY</literal> and
+ To determine whether output string values in a result set returned
+ from the server contain binary or non-binary data, check whether
+ the <literal>charsetnr</literal> value of the result set metadata
+ is 63. If so, the character set is <literal>binary</literal>,
+ which indicates binary rather than non-binary data. This enables
+ you to distinguish between <literal>BINARY</literal> and
<literal>CHAR</literal>, <literal>VARBINARY</literal> and
- <literal>VARCHAR</literal>, and <literal>BLOB</literal> and
- <literal>TEXT</literal>.
+ <literal>VARCHAR</literal>, and the <literal>BLOB</literal> and
+ <literal>TEXT</literal> types.
</para>
</section>
| Thread |
|---|
| • svn commit - mysqldoc@docsrva: r4895 - in trunk: . refman-4.1 refman-5.0 refman-5.1 | paul | 12 Feb |
