Koha/C4/BackgroundJob.pm
Galen Charlton a869ed0435 MARC import: part 5 of large file support
Staging job now gets moved to background so that
it does not get killed if it takes longer
than the Apache timeout.

Added AJAX monitoring of job status.

Signed-off-by: Joshua Ferraro <jmf@liblime.com>
2007-11-26 20:05:34 -06:00

314 lines
5.6 KiB
Perl

package C4::BackgroundJob;
# Copyright (C) 2007 LibLime
# Galen Charlton <galen.charlton@liblime.com>
#
# This file is part of Koha.
#
# Koha is free software; you can redistribute it and/or modify it under the
# terms of the GNU General Public License as published by the Free Software
# Foundation; either version 2 of the License, or (at your option) any later
# version.
#
# Koha is distributed in the hope that it will be useful, but WITHOUT ANY
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
# A PARTICULAR PURPOSE. See the GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along with
# Koha; if not, write to the Free Software Foundation, Inc., 59 Temple Place,
# Suite 330, Boston, MA 02111-1307 USA
use strict;
use C4::Context;
use C4::Auth qw/get_session/;
use Digest::MD5;
use vars qw($VERSION);
# set the version for version checking
$VERSION = 3.00;
=head1 NAME
C4::BackgroundJob - manage long-running jobs
initiated from the web staff interface
=head1 SYNOPSIS
=over 4
# start tracking a job
my $job = C4::BackgroundJob->new($sessionID, $job_name, $job_invoker, $num_work_units);
my $jobID = $job->id();
$job->progress($work_units_processed);
$job->finish($job_result_hashref);
# get status and results of a job
my $job = C4::BackgroundJob->fetch($sessionID, $jobID);
my $max_work_units = $job->size();
my $work_units_processed = $job->progress();
my $job_status = $job->status();
my $job_name = $job->name();
my $job_invoker = $job->invoker();
my $results_hashref = $job->results();
=back
This module manages tracking the progress and results
of (potentially) long-running jobs initiated from
the staff user interface. Such jobs can include
batch MARC and patron record imports.
=cut
=head1 METHODS
=cut
=head2 new
=over 4
my $job = C4::BackgroundJob->new($sessionID, $job_name, $job_invoker, $num_work_units);
=back
Create a new job object and set its status to 'running'. C<$num_work_units>
should be a number representing the size of the job; the units of the
job size are up to the caller and could be number of records,
number of bytes, etc.
=cut
sub new {
my $class = shift;
my ($sessionID, $job_name, $job_invoker, $num_work_units) = @_;
my $self = {};
$self->{'sessionID'} = $sessionID;
$self->{'name'} = $job_name;
$self->{'invoker'} = $job_invoker;
$self->{'size'} = $num_work_units;
$self->{'progress'} = 0;
$self->{'status'} = "running";
$self->{'jobID'} = Digest::MD5::md5_hex(Digest::MD5::md5_hex(time().{}.rand().{}.$$));
bless $self, $class;
$self->_serialize();
return $self;
}
# store object in CGI session
sub _serialize {
my $self = shift;
my $prefix = "job_" . $self->{'jobID'};
my $session = get_session($self->{'sessionID'});
$session->param($prefix, $self);
$session->flush();
}
=head2 id
=over 4
my $jobID = $job->id();
=back
Read-only accessor for job ID.
=cut
sub id {
my $self = shift;
return $self->{'jobID'};
}
=head2 name
=over 4
my $name = $job->name();
$job->name($name);
=back
Read/write accessor for job name.
=cut
sub name {
my $self = shift;
if (@_) {
$self->{'name'} = shift;
$self->_serialize();
} else {
return $self->{'name'};
}
}
=head2 invoker
=over 4
my $invoker = $job->invoker();
$job->invoker($invoker);
=back
Read/write accessor for job invoker.
=cut
sub invoker {
my $self = shift;
if (@_) {
$self->{'invoker'} = shift;
$self->_serialize();
} else {
return $self->{'invoker'};
}
}
=head2 progress
=over 4
my $progress = $job->progress();
$job->progress($progress);
=back
Read/write accessor for job progress.
=cut
sub progress {
my $self = shift;
if (@_) {
$self->{'progress'} = shift;
$self->_serialize();
} else {
return $self->{'progress'};
}
}
=head2 status
=over 4
my $status = $job->status();
=back
Read-only accessor for job status.
=cut
sub status {
my $self = shift;
return $self->{'status'};
}
=head2 size
=over 4
my $size = $job->size();
$job->size($size);
=back
Read/write accessor for job size.
=cut
sub size {
my $self = shift;
if (@_) {
$self->{'size'} = shift;
$self->_serialize();
} else {
return $self->{'size'};
}
}
=head2 finish
=over 4
$job->finish($results_hashref);
=back
Mark the job as finished, setting its status to 'completed'.
C<$results_hashref> should be a reference to a hash containing
the results of the job.
=cut
sub finish {
my $self = shift;
my $results_hashref = shift;
$self->{'status'} = 'completed';
$self->{'results'} = $results_hashref;
$self->_serialize();
}
=head2 results
=over 4
my $results_hashref = $job->results();
=back
Retrieve the results of the current job. Returns undef
if the job status is not 'completed'.
=cut
sub results {
my $self = shift;
return undef unless $self->{'status'} eq 'completed';
return $self->{'results'};
}
=head2 fetch
=over 4
my $job = C4::BackgroundJob->fetch($sessionID, $jobID);
=back
Retrieve a job that has been serialized to the database.
Returns C<undef> if the job does not exist in the current
session.
=cut
sub fetch {
my $class = shift;
my $sessionID = shift;
my $jobID = shift;
my $session = get_session($sessionID);
my $prefix = "job_$jobID";
unless (defined $session->param($prefix)) {
return undef;
}
my $self = $session->param($prefix);
bless $self, $class;
return $self;
}
=head1 AUTHOR
Koha Development Team <info@koha.org>
Galen Charlton <galen.charlton@liblime.com>
=cut