PHP 8.3.0 RC 6 available for testing

mysql_info

(PHP 4 >= 4.3.0, PHP 5)

mysql_info获取最近查询的有关信息

警告

本扩展自 PHP 5.5.0 起已废弃,并在自 PHP 7.0.0 开始被移除。应使用 MySQLiPDO_MySQL 扩展来替换之。参见 MySQL:选择 API 指南来获取更多信息。用以替代本函数的有:

说明

mysql_info(resource $link_identifier = NULL): string

返回最近一条查询的详细信息。

参数

link_identifier

MySQL 连接。如不指定连接标识,则使用由 mysql_connect() 最近打开的连接。如果没有找到该连接,会尝试不带参数调用 mysql_connect() 来创建。如没有找到连接或无法建立连接,则会生成 E_WARNING 级别的错误。

返回值

在成功时返回语句信息,在失败时返回 false。请查看下面的示例,了解哪些语句提供信息,以及返回值可能是什么样子。未列出的语句将返回 false

示例

示例 #1 相关的 MySQL 语句

返回字符串值的语句。这些数字仅用于说明目的;它们的值将与查询相对应。

INSERT INTO ... SELECT ...
String format: Records: 23 Duplicates: 0 Warnings: 0 
INSERT INTO ... VALUES (...),(...),(...)...
String format: Records: 37 Duplicates: 0 Warnings: 0 
LOAD DATA INFILE ...
String format: Records: 42 Deleted: 0 Skipped: 0 Warnings: 0 
ALTER TABLE
String format: Records: 60 Duplicates: 0 Warnings: 0 
UPDATE
String format: Rows matched: 65 Changed: 65 Warnings: 0

注释

注意:

mysql_info() 对于 INSERT ... VALUES 语句仅在该语句中列出了多个值的情况下返回非 false 的值。

参见

add a note

User Contributed Notes 5 notes

up
7
info at granville dot nl
18 years ago
Imade a quick conversion of eric's function just to count matched or affected rows from a query.

/**GD gdf_db_count_query_v1: returns the amount of rows matched or affected by the last query. Must be used immediately after the concerned query.
*/

function gdf_db_count_query($link = 'dbh') {

$info_str = mysql_info($$link);

if (ereg("Records: ([0-9]*)", $info_str, $count) == false) {
ereg("Rows matched: ([0-9]*)", $info_str, $count);
}

return $count;

}
up
5
eric at projectsatellite dot com
20 years ago
I agree that this is a useful function to use when trying to check on whether an update query matched a particular row. I created a simple function that returns an associative array with the values delineated in the returned string.

function get_mysql_info($linkid = null){
$linkid? $strInfo = mysql_info($linkid) : $strInfo = mysql_info();

$return = array();
ereg("Records: ([0-9]*)", $strInfo, $records);
ereg("Duplicates: ([0-9]*)", $strInfo, $dupes);
ereg("Warnings: ([0-9]*)", $strInfo, $warnings);
ereg("Deleted: ([0-9]*)", $strInfo, $deleted);
ereg("Skipped: ([0-9]*)", $strInfo, $skipped);
ereg("Rows matched: ([0-9]*)", $strInfo, $rows_matched);
ereg("Changed: ([0-9]*)", $strInfo, $changed);

$return['records'] = $records[1];
$return['duplicates'] = $dupes[1];
$return['warnings'] = $warnings[1];
$return['deleted'] = $deleted[1];
$return['skipped'] = $skipped[1];
$return['rows_matched'] = $rows_matched[1];
$return['changed'] = $changed[1];

return $return;
}

After trying to update a row that may or may not exist, you can use the above function like so:

$vals = get_mysql_info($linkid);
if($vals['rows_matched'] == 0){
mysql_query("INSERT INTO table values('val1','val2', 'valetc')", $linkid);
}
up
4
tomas at matfyz dot cz
14 years ago
Please note that the information on warning count cannot be taken from the mysql_info() due to mysql bugs #41283 and #41285:

http://bugs.mysql.com/?id=41283
http://bugs.mysql.com/?id=41285
up
1
bdobrica at gmail dot com
17 years ago
As a solution to the problem pointed in the post reffering to mysql_affected_rows() returning 0 when you are making an update query and the fields are not modified although the query is valid, i'm posting the following function. It is very simple and based on a previous post.

function mysql_modified_rows () {
$info_str = mysql_info();
$a_rows = mysql_affected_rows();
ereg("Rows matched: ([0-9]*)", $info_str, $r_matched);
return ($a_rows < 1)?($r_matched[1]?$r_matched[1]:0):$a_rows;
}

Hope you'll find it usefull.
up
-8
carl at NOSPAMthep dot lu dot se
21 years ago
This function can be used as a workaround for a misfeature of MySQL: on an UPDATE, rows that aren't updated _solely because they looked the same before_ will not be seen in mysql_affected_rows(). This causes problems when you want to use the result of the update to determine if there's need to do an INSERT. With MySQL you can do an INSERT IGNORE if there's no risk of if failing because of a duplicate key other than the one used in the UPDATE. However, if this isn't the case or you want a bit of RDBMS independence, there's no easy/pretty workaround. I think I'll resort to doing a SELECT to determine the primary key before doing the update/insert, as using the CVS version of PHP isn't an option for me.
To Top