If you only have access to MariaDB, I guess it's a matter of setting up test tables, test data, and test operations in such a way that you can confirm what the return values actually correspond to (in some set of scenarios relevant to your application).

If it really matters to you how this compares to a Mysql server, but you don't have access to try that out, perhaps you could post your experimental tables, data and operations here, and someone who does have Mysql could try it out and tell you what they get.

Okay, good idea, graff. Here's what I did:

use strict;
use warnings;
use DBI;
use Try::Tiny;

$|=1;

my $redo;
my $dbh;
print 'mysql username: ';
chomp(my $username=<STDIN>);
print 'mysql password: ';
system('stty','-echo');chomp(my $pw=<STDIN>);system('stty','echo');
print $/;
sub reconn {
  $dbh = DBI->connect('dbi:mysql:scratch;mysql_enable_utf8=1',$usernam
+e,$pw,{mysql_auto_reconnect => 1}) or die $DBI::errstr;
}
reconn;
my $sth;

# MariaDB [scratch]> create table foo(bar int,baz int,primary key(bar)
+);
# Query OK, 0 rows affected (0.02 sec)
# 
# MariaDB [scratch]> insert foo(bar)value(1);
# Query OK, 1 row affected (0.01 sec)
# 
# MariaDB [scratch]> desc foo;
# +-------+---------+------+-----+---------+-------+
# | Field | Type    | Null | Key | Default | Extra |
# +-------+---------+------+-----+---------+-------+
# | bar   | int(11) | NO   | PRI | 0       |       |
# | baz   | int(11) | YES  |     | NULL    |       |
# +-------+---------+------+-----+---------+-------+
# 2 rows in set (0.00 sec)
# 
# # MariaDB [scratch]> select * from foo;
# +-----+------+
# | bar | baz  |
# +-----+------+
# |   1 | NULL |
# +-----+------+
# 1 row in set (0.00 sec)

$\=$/;

my $rows;
$dbh->ping or reconn;
$sth = $dbh->prepare("insert foo(bar)value(2);");
$redo = 1; while ($redo) {$redo=0; try {$rows=$sth->execute();} catch 
+{$redo = /deadlock found/i;}}
print "new: $rows";

undef $rows;
$dbh->ping or reconn;
$sth = $dbh->prepare("insert ignore foo(bar)value(1);");
$redo = 1; while ($redo) {$redo=0; try {$rows=$sth->execute();} catch 
+{$redo = /deadlock found/i;}}
print "old ignore: $rows";

undef $rows;
$dbh->ping or reconn;
$sth = $dbh->prepare("insert foo(bar)value(1) on duplicate key update 
+baz=1;");
$redo = 1; while ($redo) {$redo=0; try {$rows=$sth->execute();} catch 
+{$redo = /deadlock found/i;}}
print "old on duplicate key update: $rows";
[download]

And the results were:

new: 1
old ignore: 0E0
old on duplicate key update: 2
[download]

And row_count gives the same numbers (in these cases):

# MariaDB [scratch]> select * from foo;
# +-----+------+
# | bar | baz  |
# +-----+------+
# |   1 | NULL |
# +-----+------+
# 1 row in set (0.00 sec)
# 
# MariaDB [scratch]> insert foo(bar)value(2);
# Query OK, 1 row affected (0.00 sec)
# 
# MariaDB [scratch]> select row_count();
# +-------------+
# | row_count() |
# +-------------+
# |           1 |
# +-------------+
# 1 row in set (0.00 sec)
# 
# MariaDB [scratch]> insert ignore foo(bar)value(1);
# Query OK, 0 rows affected, 1 warning (0.00 sec)
# 
# MariaDB [scratch]> select row_count();
# +-------------+
# | row_count() |
# +-------------+
# |           0 |
# +-------------+
# 1 row in set (0.01 sec)
# 
# MariaDB [scratch]> insert foo(bar)value(1) on duplicate key update b
+az=1;
# Query OK, 2 rows affected (0.00 sec)
# 
# MariaDB [scratch]> select row_count();
# +-------------+
# | row_count() |
# +-------------+
# |           2 |
# +-------------+
# 1 row in set (0.00 sec)
[download]

$_="msh210";$"=$\;@_=@{[split//,uc]}[2,0];$_="@_$\1";$\=$/;++$_[0]for$...1;print lc substr crypt($_,"@_"),1,6

Ah, but now I've tried an on duplicate key update where the update didn't actually change anything (the value to update to was there already), and select row_count() gave 0, but execute returned 1. So they're not always the same.

(The query was insert foo(bar)value(1)on duplicate key update baz=1; and the table already had 1 in each column.)

$_="msh210";$"=$\;@_=@{[split//,uc]}[2,0];$_="@_$\1";$\=$/;++$_[0]for$...1;print lc substr crypt($_,"@_"),1,6