1 |
NAME |
2 |
AnyEvent::DBI - asynchronous DBI access |
3 |
|
4 |
SYNOPSIS |
5 |
use AnyEvent::DBI; |
6 |
|
7 |
my $cv = AnyEvent->condvar; |
8 |
|
9 |
my $dbh = new AnyEvent::DBI "DBI:SQLite:dbname=test.db", "", ""; |
10 |
|
11 |
$dbh->exec ("select * from test where num=?", 10, sub { |
12 |
my ($dbh, $rows, $rv) = @_; |
13 |
|
14 |
$#_ or die "failure: $@"; |
15 |
|
16 |
print "@$_\n" |
17 |
for @$rows; |
18 |
|
19 |
$cv->broadcast; |
20 |
}); |
21 |
|
22 |
# asynchronously do sth. else here |
23 |
|
24 |
$cv->wait; |
25 |
|
26 |
DESCRIPTION |
27 |
This module is an AnyEvent user, you need to make sure that you use and |
28 |
run a supported event loop. |
29 |
|
30 |
This module implements asynchronous DBI access by forking or executing |
31 |
separate "DBI-Server" processes and sending them requests. |
32 |
|
33 |
It means that you can run DBI requests in parallel to other tasks. |
34 |
|
35 |
With DBD::mysql, the overhead for very simple statements ("select 0") is |
36 |
somewhere around 50% compared to an explicit |
37 |
prepare_cached/execute/fetchrow_arrayref/finish combination. With |
38 |
DBD::SQlite3, it's more like a factor of 8 for this trivial statement. |
39 |
|
40 |
ERROR HANDLING |
41 |
This module defines a number of functions that accept a callback |
42 |
argument. All callbacks used by this module get their AnyEvent::DBI |
43 |
handle object passed as first argument. |
44 |
|
45 |
If the request was successful, then there will be more arguments, |
46 |
otherwise there will only be the $dbh argument and $@ contains an error |
47 |
message. |
48 |
|
49 |
A convenient way to check whether an error occurred is to check $#_ - if |
50 |
that is true, then the function was successful, otherwise there was an |
51 |
error. |
52 |
|
53 |
METHODS |
54 |
$dbh = new AnyEvent::DBI $database, $user, $pass, [key => value]... |
55 |
Returns a database handle for the given database. Each database |
56 |
handle has an associated server process that executes statements in |
57 |
order. If you want to run more than one statement in parallel, you |
58 |
need to create additional database handles. |
59 |
|
60 |
The advantage of this approach is that transactions work as state is |
61 |
preserved. |
62 |
|
63 |
Example: |
64 |
|
65 |
$dbh = new AnyEvent::DBI |
66 |
"DBI:mysql:test;mysql_read_default_file=/root/.my.cnf", "", ""; |
67 |
|
68 |
Additional key-value pairs can be used to adjust behaviour: |
69 |
|
70 |
on_error => $callback->($dbh, $filename, $line, $fatal) |
71 |
When an error occurs, then this callback will be invoked. On |
72 |
entry, $@ is set to the error message. $filename and $line is |
73 |
where the original request was submitted. |
74 |
|
75 |
If the fatal argument is true then the database connection is |
76 |
shut down and your database handle became invalid. In addition |
77 |
to invoking the "on_error" callback, all of your queued request |
78 |
callbacks are called without only the $dbh argument. |
79 |
|
80 |
If omitted, then "die" will be called on any errors, fatal or |
81 |
not. |
82 |
|
83 |
on_connect => $callback->($dbh[, $success]) |
84 |
If you supply an "on_connect" callback, then this callback will |
85 |
be invoked after the database connect attempt. If the connection |
86 |
succeeds, $success is true, otherwise it is missing and $@ |
87 |
contains the $DBI::errstr. |
88 |
|
89 |
Regardless of whether "on_connect" is supplied, connect errors |
90 |
will result in "on_error" being called. However, if no |
91 |
"on_connect" callback is supplied, then connection errors are |
92 |
considered fatal. The client will "die" and the "on_error" |
93 |
callback will be called with $fatal true. |
94 |
|
95 |
When on_connect is supplied, connect error are not fatal and |
96 |
AnyEvent::DBI will not "die". You still cannot, however, use the |
97 |
$dbh object you received from "new" to make requests. |
98 |
|
99 |
fork_template => $AnyEvent::Fork-object |
100 |
"AnyEvent::DBI" uses "AnyEvent::Fork->new" to create the |
101 |
database slave, which in turn either "exec"'s a new process |
102 |
(similar to the old "exec_server" constructor argument) or uses |
103 |
a process forked early (see AnyEvent::Fork::Early). |
104 |
|
105 |
With this argument you can provide your own fork template. This |
106 |
can be useful if you create a lot of "AnyEvent::DBI" handles and |
107 |
want to save memory (And speed up startup) by not having to load |
108 |
"AnyEvent::DBI" again and again into your child processes: |
109 |
|
110 |
my $template = AnyEvent::Fork |
111 |
->new # create new template |
112 |
->require ("AnyEvent::DBI::Slave"); # preload AnyEvent::DBI::Slave module |
113 |
|
114 |
for (...) { |
115 |
$dbh = new AnyEvent::DBI ... |
116 |
fork_template => $template; |
117 |
|
118 |
timeout => seconds |
119 |
If you supply a timeout parameter (fractional values are |
120 |
supported), then a timer is started any time the DBI handle |
121 |
expects a response from the server. This includes connection |
122 |
setup as well as requests made to the backend. The timeout spans |
123 |
the duration from the moment the first data is written (or |
124 |
queued to be written) until all expected responses are returned, |
125 |
but is postponed for "timeout" seconds each time more data is |
126 |
returned from the server. If the timer ever goes off then a |
127 |
fatal error is generated. If you have an "on_error" handler |
128 |
installed, then it will be called, otherwise your program will |
129 |
die(). |
130 |
|
131 |
When altering your databases with timeouts it is wise to use |
132 |
transactions. If you quit due to timeout while performing |
133 |
insert, update or schema-altering commands you can end up not |
134 |
knowing if the action was submitted to the database, |
135 |
complicating recovery. |
136 |
|
137 |
Timeout errors are always fatal. |
138 |
|
139 |
Any additional key-value pairs will be rolled into a hash reference |
140 |
and passed as the final argument to the "DBI->connect (...)" call. |
141 |
For example, to suppress errors on STDERR and send them instead to |
142 |
an AnyEvent::Handle you could do: |
143 |
|
144 |
$dbh = new AnyEvent::DBI |
145 |
"DBI:mysql:test;mysql_read_default_file=/root/.my.cnf", "", "", |
146 |
PrintError => 0, |
147 |
on_error => sub { |
148 |
$log_handle->push_write ("DBI Error: $@ at $_[1]:$_[2]\n"); |
149 |
}; |
150 |
|
151 |
$dbh->on_error ($cb->($dbh, $filename, $line, $fatal)) |
152 |
Sets (or clears, with "undef") the "on_error" handler. |
153 |
|
154 |
$dbh->timeout ($seconds) |
155 |
Sets (or clears, with "undef") the database timeout. Useful to |
156 |
extend the timeout when you are about to make a really long query. |
157 |
|
158 |
$dbh->attr ($attr_name[, $attr_value], $cb->($dbh, $new_value)) |
159 |
An accessor for the database handle attributes, such as |
160 |
"AutoCommit", "RaiseError", "PrintError" and so on. If you provide |
161 |
an $attr_value (which might be "undef"), then the given attribute |
162 |
will be set to that value. |
163 |
|
164 |
The callback will be passed the database handle and the attribute's |
165 |
value if successful. |
166 |
|
167 |
If an error occurs and the "on_error" callback returns, then only |
168 |
$dbh will be passed and $@ contains the error message. |
169 |
|
170 |
$dbh->exec ("statement", @args, $cb->($dbh, \@rows, $rv)) |
171 |
Executes the given SQL statement with placeholders replaced by |
172 |
@args. The statement will be prepared and cached on the server side, |
173 |
so using placeholders is extremely important. |
174 |
|
175 |
The callback will be called with a weakened AnyEvent::DBI object as |
176 |
the first argument and the result of "fetchall_arrayref" as (or |
177 |
"undef" if the statement wasn't a select statement) as the second |
178 |
argument. |
179 |
|
180 |
Third argument is the return value from the "DBI->execute" method |
181 |
call. |
182 |
|
183 |
If an error occurs and the "on_error" callback returns, then only |
184 |
$dbh will be passed and $@ contains the error message. |
185 |
|
186 |
$dbh->stattr ($attr_name, $cb->($dbh, $value)) |
187 |
An accessor for the statement attributes of the most recently |
188 |
executed statement, such as "NAME" or "TYPE". |
189 |
|
190 |
The callback will be passed the database handle and the attribute's |
191 |
value if successful. |
192 |
|
193 |
If an error occurs and the "on_error" callback returns, then only |
194 |
$dbh will be passed and $@ contains the error message. |
195 |
|
196 |
$dbh->begin_work ($cb->($dbh[, $rc])) |
197 |
$dbh->commit ($cb->($dbh[, $rc])) |
198 |
$dbh->rollback ($cb->($dbh[, $rc])) |
199 |
The begin_work, commit, and rollback methods expose the equivalent |
200 |
transaction control method of the DBI driver. On success, $rc is |
201 |
true. |
202 |
|
203 |
If an error occurs and the "on_error" callback returns, then only |
204 |
$dbh will be passed and $@ contains the error message. |
205 |
|
206 |
$dbh->func ('string_which_yields_args_when_evaled', $func_name, |
207 |
$cb->($dbh, $rc, $dbi_err, $dbi_errstr)) |
208 |
This gives access to database driver private methods. Because they |
209 |
are not standard you cannot always depend on the value of $rc or |
210 |
$dbi_err. Check the documentation for your specific driver/function |
211 |
combination to see what it returns. |
212 |
|
213 |
Note that the first argument will be eval'ed to produce the argument |
214 |
list to the func() method. This must be done because the |
215 |
serialization protocol between the AnyEvent::DBI server process and |
216 |
your program does not support the passage of closures. |
217 |
|
218 |
Here's an example to extend the query language in SQLite so it |
219 |
supports an intstr() function: |
220 |
|
221 |
$cv = AnyEvent->condvar; |
222 |
$dbh->func ( |
223 |
q{ |
224 |
instr => 2, sub { |
225 |
my ($string, $search) = @_; |
226 |
return index $string, $search; |
227 |
}, |
228 |
}, |
229 |
create_function => sub { |
230 |
return $cv->send ($@) |
231 |
unless $#_; |
232 |
$cv->send (undef, @_[1,2,3]); |
233 |
} |
234 |
); |
235 |
|
236 |
my ($err,$rc,$errcode,$errstr) = $cv->recv; |
237 |
|
238 |
die $err if defined $err; |
239 |
die "EVAL failed: $errstr" |
240 |
if $errcode; |
241 |
|
242 |
# otherwise, we can ignore $rc and $errcode for this particular func |
243 |
|
244 |
SEE ALSO |
245 |
AnyEvent, DBI, Coro::Mysql. |
246 |
|
247 |
AUTHOR AND CONTACT |
248 |
Marc Lehmann <schmorp@schmorp.de> (current maintainer) |
249 |
http://home.schmorp.de/ |
250 |
|
251 |
Adam Rosenstein <adam@redcondor.com> |
252 |
http://www.redcondor.com/ |
253 |
|