line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package Schedule::LongSteps::Storage::DBIxClass; |
2
|
|
|
|
|
|
|
$Schedule::LongSteps::Storage::DBIxClass::VERSION = '0.023'; |
3
|
2
|
|
|
2
|
|
68122
|
use Moose; |
|
2
|
|
|
|
|
379625
|
|
|
2
|
|
|
|
|
16
|
|
4
|
|
|
|
|
|
|
extends qw/Schedule::LongSteps::Storage/; |
5
|
|
|
|
|
|
|
|
6
|
2
|
|
|
2
|
|
13983
|
use DateTime; |
|
2
|
|
|
|
|
753188
|
|
|
2
|
|
|
|
|
100
|
|
7
|
2
|
|
|
2
|
|
383
|
use Log::Any qw/$log/; |
|
2
|
|
|
|
|
5811
|
|
|
2
|
|
|
|
|
14
|
|
8
|
2
|
|
|
2
|
|
2320
|
use Scope::Guard; |
|
2
|
|
|
|
|
699
|
|
|
2
|
|
|
|
|
77
|
|
9
|
2
|
|
|
2
|
|
534
|
use Action::Retry; |
|
2
|
|
|
|
|
64576
|
|
|
2
|
|
|
|
|
1210
|
|
10
|
|
|
|
|
|
|
|
11
|
|
|
|
|
|
|
has 'schema' => ( is => 'ro', isa => 'DBIx::Class::Schema', required => 1); |
12
|
|
|
|
|
|
|
has 'resultset_name' => ( is => 'ro', isa => 'Str', required => 1); |
13
|
|
|
|
|
|
|
|
14
|
|
|
|
|
|
|
has 'limit_per_tick' => ( is => 'ro', isa => 'Int', default => 50 ); |
15
|
|
|
|
|
|
|
|
16
|
|
|
|
|
|
|
sub _get_resultset{ |
17
|
0
|
|
|
0
|
|
|
my ($self) = @_; |
18
|
0
|
|
|
|
|
|
return $self->schema()->resultset($self->resultset_name()); |
19
|
|
|
|
|
|
|
} |
20
|
|
|
|
|
|
|
|
21
|
|
|
|
|
|
|
around [ 'prepare_due_processes', 'create_process' ] => sub{ |
22
|
|
|
|
|
|
|
my ($orig, $self, @rest ) = @_; |
23
|
|
|
|
|
|
|
|
24
|
|
|
|
|
|
|
# Transfer the current autocommit nature of the DBH |
25
|
|
|
|
|
|
|
# as a transation might have been created on this DBH outside |
26
|
|
|
|
|
|
|
# of this schema. A transaction on DBI sets AutoCommit to false |
27
|
|
|
|
|
|
|
# on the DBH. transaction_depth is just a boolean on the storage. |
28
|
|
|
|
|
|
|
|
29
|
|
|
|
|
|
|
# First restore transaction depth as it was. |
30
|
|
|
|
|
|
|
my $pre_transaction_depth = $self->schema()->storage()->transaction_depth(); |
31
|
|
|
|
|
|
|
my $guard = Scope::Guard->new( |
32
|
|
|
|
|
|
|
sub{ |
33
|
|
|
|
|
|
|
$log->trace("Restoring transaction_depth = $pre_transaction_depth"); |
34
|
|
|
|
|
|
|
$self->schema()->storage()->transaction_depth( $pre_transaction_depth ); |
35
|
|
|
|
|
|
|
}); |
36
|
|
|
|
|
|
|
|
37
|
|
|
|
|
|
|
my $current_transaction_depth = $self->schema()->storage()->dbh()->{AutoCommit} ? 0 : 1; |
38
|
|
|
|
|
|
|
$log->trace("Setting transaction_depth as NOT dbh AutoCommit = ".$current_transaction_depth); |
39
|
|
|
|
|
|
|
$self->schema()->storage()->transaction_depth( $current_transaction_depth ); |
40
|
|
|
|
|
|
|
return $self->$orig( @rest ); |
41
|
|
|
|
|
|
|
}; |
42
|
|
|
|
|
|
|
|
43
|
|
|
|
|
|
|
|
44
|
|
|
|
|
|
|
=head1 NAME |
45
|
|
|
|
|
|
|
|
46
|
|
|
|
|
|
|
Schedule::LongSteps::Storage::DBIxClass - DBIx::Class based storage. |
47
|
|
|
|
|
|
|
|
48
|
|
|
|
|
|
|
=head1 SYNOPSIS |
49
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
|
First instantiate a storage with your L<DBIx::Class::Schema> and the name |
51
|
|
|
|
|
|
|
of the resultset that represent the stored process: |
52
|
|
|
|
|
|
|
|
53
|
|
|
|
|
|
|
my $storage = Schedule::LongSteps::Storage::DBIxClass->new({ |
54
|
|
|
|
|
|
|
schema => $dbic_schema, |
55
|
|
|
|
|
|
|
resultset_name => 'LongstepsProcess' |
56
|
|
|
|
|
|
|
}); |
57
|
|
|
|
|
|
|
|
58
|
|
|
|
|
|
|
Then build and use a L<Schedule::LongSteps> object: |
59
|
|
|
|
|
|
|
|
60
|
|
|
|
|
|
|
my $long_steps = Schedule::LongSteps->new({ storage => $storage }); |
61
|
|
|
|
|
|
|
|
62
|
|
|
|
|
|
|
... |
63
|
|
|
|
|
|
|
|
64
|
|
|
|
|
|
|
=head1 ATTRIBUTES |
65
|
|
|
|
|
|
|
|
66
|
|
|
|
|
|
|
=over |
67
|
|
|
|
|
|
|
|
68
|
|
|
|
|
|
|
=item schema |
69
|
|
|
|
|
|
|
|
70
|
|
|
|
|
|
|
You DBIx::Class::Schema. Mandatory. |
71
|
|
|
|
|
|
|
|
72
|
|
|
|
|
|
|
=item resultset_name |
73
|
|
|
|
|
|
|
|
74
|
|
|
|
|
|
|
The name of the resultset holding the processes in your Schema. See section 'RESULTSET REQUIREMENTS'. Mandatory. |
75
|
|
|
|
|
|
|
|
76
|
|
|
|
|
|
|
=item limit_per_tick |
77
|
|
|
|
|
|
|
|
78
|
|
|
|
|
|
|
The maximum number of processes that will actually run each time you |
79
|
|
|
|
|
|
|
call $longsteps->run_due_processes(). Use that to control how long it takes to run |
80
|
|
|
|
|
|
|
a single call to $longsteps->run_due_processes(). |
81
|
|
|
|
|
|
|
|
82
|
|
|
|
|
|
|
Note that you can have an arbitrary number of processes all doing $longsteps->run_due_processes() AT THE SAME TIME. |
83
|
|
|
|
|
|
|
|
84
|
|
|
|
|
|
|
This will ensure that no process step is run more than one time. |
85
|
|
|
|
|
|
|
|
86
|
|
|
|
|
|
|
Default to 50. |
87
|
|
|
|
|
|
|
|
88
|
|
|
|
|
|
|
=back |
89
|
|
|
|
|
|
|
|
90
|
|
|
|
|
|
|
=head1 RESULTSET REQUIREMENTS |
91
|
|
|
|
|
|
|
|
92
|
|
|
|
|
|
|
The resultset to use with this storage MUST contain the following columns, constraints and indices: |
93
|
|
|
|
|
|
|
|
94
|
|
|
|
|
|
|
=over |
95
|
|
|
|
|
|
|
|
96
|
|
|
|
|
|
|
=item id |
97
|
|
|
|
|
|
|
|
98
|
|
|
|
|
|
|
A unique primary key auto incrementable identifier |
99
|
|
|
|
|
|
|
|
100
|
|
|
|
|
|
|
=item process_class |
101
|
|
|
|
|
|
|
|
102
|
|
|
|
|
|
|
A VARCHAR long enough to hold your L<Schedule::LongSteps::Process> class names. NOT NULL. |
103
|
|
|
|
|
|
|
|
104
|
|
|
|
|
|
|
=item what |
105
|
|
|
|
|
|
|
|
106
|
|
|
|
|
|
|
A VARCHAR long enough to hold the name of one of your steps. Can be NULL. |
107
|
|
|
|
|
|
|
|
108
|
|
|
|
|
|
|
=item status |
109
|
|
|
|
|
|
|
|
110
|
|
|
|
|
|
|
A VARCHAR(50) NOT NULL, defaults to 'pending' |
111
|
|
|
|
|
|
|
|
112
|
|
|
|
|
|
|
=item run_at |
113
|
|
|
|
|
|
|
|
114
|
|
|
|
|
|
|
A Datetime (or timestamp with timezone in PgSQL). Will hold a UTC Timezoned date of the next run. Default to NULL. |
115
|
|
|
|
|
|
|
|
116
|
|
|
|
|
|
|
Please index this so it is fast to select a range. |
117
|
|
|
|
|
|
|
|
118
|
|
|
|
|
|
|
=item run_id |
119
|
|
|
|
|
|
|
|
120
|
|
|
|
|
|
|
A CHAR or VARCHAR (at least 36). Default to NULL. |
121
|
|
|
|
|
|
|
|
122
|
|
|
|
|
|
|
Please index this so it is fast to select rows with a matching run_id |
123
|
|
|
|
|
|
|
|
124
|
|
|
|
|
|
|
=item state |
125
|
|
|
|
|
|
|
|
126
|
|
|
|
|
|
|
A Reasonably long TEXT field (or JSON field in supporting databases) capable of holding |
127
|
|
|
|
|
|
|
a JSON dump of pure Perl data. NOT NULL. |
128
|
|
|
|
|
|
|
|
129
|
|
|
|
|
|
|
You HAVE to implement inflating and deflating yourself. See L<DBIx::Class::InflateColumn::Serializer::JSON> |
130
|
|
|
|
|
|
|
or similar techniques. |
131
|
|
|
|
|
|
|
|
132
|
|
|
|
|
|
|
See t/fullblown.t for a full blown working example. |
133
|
|
|
|
|
|
|
|
134
|
|
|
|
|
|
|
=item error |
135
|
|
|
|
|
|
|
|
136
|
|
|
|
|
|
|
A reasonably long TEXT field capable of holding a full stack trace in case something goes wrong. Defaults to NULL. |
137
|
|
|
|
|
|
|
|
138
|
|
|
|
|
|
|
=back |
139
|
|
|
|
|
|
|
|
140
|
|
|
|
|
|
|
=cut |
141
|
|
|
|
|
|
|
|
142
|
|
|
|
|
|
|
=head2 prepare_due_processes |
143
|
|
|
|
|
|
|
|
144
|
|
|
|
|
|
|
See L<Schedule::LongSteps::Storage::DBIxClass> |
145
|
|
|
|
|
|
|
|
146
|
|
|
|
|
|
|
=cut |
147
|
|
|
|
|
|
|
|
148
|
|
|
|
|
|
|
sub prepare_due_processes{ |
149
|
|
|
|
|
|
|
my ($self) = @_; |
150
|
|
|
|
|
|
|
|
151
|
|
|
|
|
|
|
my $now = DateTime->now(); |
152
|
|
|
|
|
|
|
my $rs = $self->_get_resultset(); |
153
|
|
|
|
|
|
|
my $dtf = $self->schema()->storage()->datetime_parser(); |
154
|
|
|
|
|
|
|
|
155
|
|
|
|
|
|
|
my $uuid = $self->uuid()->create_str(); |
156
|
|
|
|
|
|
|
$log->info("Creating batch ID $uuid"); |
157
|
|
|
|
|
|
|
|
158
|
|
|
|
|
|
|
|
159
|
|
|
|
|
|
|
# Note that we do not use the SELECT FOR UPDATE technique here. |
160
|
|
|
|
|
|
|
# Instead this generates a single UPDATE statement like this one: |
161
|
|
|
|
|
|
|
# UPDATE longsteps_process SET run_id = ?, status = ? WHERE ( id IN ( SELECT me.id FROM longsteps_process me WHERE ( ( run_at <= ? AND run_id IS NULL ) ) LIMIT ? ) ) |
162
|
|
|
|
|
|
|
my $stuff = sub{ |
163
|
|
|
|
|
|
|
$rs->search({ |
164
|
|
|
|
|
|
|
run_at => { '<=' => $dtf->format_datetime( $now ) }, |
165
|
|
|
|
|
|
|
run_id => undef, |
166
|
|
|
|
|
|
|
}, { |
167
|
|
|
|
|
|
|
rows => $self->limit_per_tick(), |
168
|
|
|
|
|
|
|
} ) |
169
|
|
|
|
|
|
|
->update({ |
170
|
|
|
|
|
|
|
run_id => $uuid, |
171
|
|
|
|
|
|
|
status => 'running' |
172
|
|
|
|
|
|
|
}); |
173
|
|
|
|
|
|
|
}; |
174
|
|
|
|
|
|
|
$stuff->(); |
175
|
|
|
|
|
|
|
|
176
|
|
|
|
|
|
|
# And return them as individual results. |
177
|
|
|
|
|
|
|
return $rs->search({ |
178
|
|
|
|
|
|
|
run_id => $uuid, |
179
|
|
|
|
|
|
|
})->all(); |
180
|
|
|
|
|
|
|
} |
181
|
|
|
|
|
|
|
|
182
|
|
|
|
|
|
|
=head2 create_process |
183
|
|
|
|
|
|
|
|
184
|
|
|
|
|
|
|
See L<Schedule::LongSteps::Storage> |
185
|
|
|
|
|
|
|
|
186
|
|
|
|
|
|
|
This override adds retrying in case of deadlock detection. |
187
|
|
|
|
|
|
|
|
188
|
|
|
|
|
|
|
=cut |
189
|
|
|
|
|
|
|
|
190
|
|
|
|
|
|
|
sub create_process{ |
191
|
|
|
|
|
|
|
my ($self, $process_properties) = @_; |
192
|
|
|
|
|
|
|
return $self->_retry_transaction( |
193
|
|
|
|
|
|
|
sub{ |
194
|
|
|
|
|
|
|
return $self->_get_resultset()->create($process_properties); |
195
|
|
|
|
|
|
|
}); |
196
|
|
|
|
|
|
|
} |
197
|
|
|
|
|
|
|
|
198
|
|
|
|
|
|
|
=head2 find_process |
199
|
|
|
|
|
|
|
|
200
|
|
|
|
|
|
|
See L<Schedule::LongSteps::Storage> |
201
|
|
|
|
|
|
|
|
202
|
|
|
|
|
|
|
=cut |
203
|
|
|
|
|
|
|
|
204
|
|
|
|
|
|
|
sub find_process{ |
205
|
0
|
|
|
0
|
1
|
|
my ($self, $process_id) = @_; |
206
|
0
|
|
|
|
|
|
return $self->_get_resultset()->find({ id => $process_id }); |
207
|
|
|
|
|
|
|
} |
208
|
|
|
|
|
|
|
|
209
|
|
|
|
|
|
|
=head2 update_process |
210
|
|
|
|
|
|
|
|
211
|
|
|
|
|
|
|
Overrides L<Schedule::LongSteps::Storage#update_process> to add |
212
|
|
|
|
|
|
|
some retrying in case of DB deadlock detection. |
213
|
|
|
|
|
|
|
|
214
|
|
|
|
|
|
|
=cut |
215
|
|
|
|
|
|
|
|
216
|
|
|
|
|
|
|
override 'update_process' => sub{ |
217
|
|
|
|
|
|
|
my ($self, $process, $properties) = @_; |
218
|
|
|
|
|
|
|
return $self->_retry_transaction( sub{ |
219
|
|
|
|
|
|
|
$log->trace("Attempting to update process ".$process->id()); |
220
|
|
|
|
|
|
|
$process->update( $properties ); |
221
|
|
|
|
|
|
|
} ); |
222
|
|
|
|
|
|
|
}; |
223
|
|
|
|
|
|
|
|
224
|
|
|
|
|
|
|
sub _retry_transaction{ |
225
|
0
|
|
|
0
|
|
|
my ($self, $code) = @_; |
226
|
|
|
|
|
|
|
|
227
|
|
|
|
|
|
|
my $retry = Action::Retry->new( |
228
|
|
|
|
|
|
|
attempt_code => $code, |
229
|
|
|
|
|
|
|
retry_if_code => sub{ |
230
|
0
|
|
|
0
|
|
|
my $exception = $_[0]; |
231
|
|
|
|
|
|
|
# The driver tells us to retry the transaction. |
232
|
|
|
|
|
|
|
# For instance: https://dev.mysql.com/doc/refman/5.6/en/innodb-deadlocks-handling.html |
233
|
|
|
|
|
|
|
|
234
|
|
|
|
|
|
|
# Note that if this is false, then the Action::Retry code |
235
|
|
|
|
|
|
|
# will set $@ to the last error and return whatever the code has returned (most |
236
|
|
|
|
|
|
|
# probably undef in this case. |
237
|
|
|
|
|
|
|
# This is managed by the error testing after the call to 'run' |
238
|
0
|
|
0
|
|
|
|
return !! ( ( $exception || '' ) =~ m/try restarting transaction/ ); |
239
|
|
|
|
|
|
|
}, |
240
|
0
|
|
|
|
|
|
strategy => 'Fibonacci', |
241
|
|
|
|
|
|
|
); |
242
|
0
|
|
|
|
|
|
my $ret = $retry->run(); |
243
|
0
|
0
|
|
|
|
|
if( my $err = $@ ){ |
244
|
0
|
|
|
|
|
|
confess($err); |
245
|
|
|
|
|
|
|
} |
246
|
0
|
|
|
|
|
|
return $ret; |
247
|
|
|
|
|
|
|
} |
248
|
|
|
|
|
|
|
|
249
|
|
|
|
|
|
|
__PACKAGE__->meta->make_immutable(); |