| 1 |
package WebPAC::Input; |
| 2 |
|
| 3 |
use warnings; |
| 4 |
use strict; |
| 5 |
|
| 6 |
=head1 NAME |
| 7 |
|
| 8 |
WebPAC::Input - core module for input file format |
| 9 |
|
| 10 |
=head1 VERSION |
| 11 |
|
| 12 |
Version 0.01 |
| 13 |
|
| 14 |
=cut |
| 15 |
|
| 16 |
our $VERSION = '0.01'; |
| 17 |
|
| 18 |
=head1 SYNOPSIS |
| 19 |
|
| 20 |
This module will load particular loader module and execute it's functions. |
| 21 |
|
| 22 |
Perhaps a little code snippet. |
| 23 |
|
| 24 |
use WebPAC::Input; |
| 25 |
|
| 26 |
my $db = WebPAC::Input->new( |
| 27 |
format => 'NULL', |
| 28 |
config => $config, |
| 29 |
lookup => $lookup_obj, |
| 30 |
); |
| 31 |
|
| 32 |
$db->open('/path/to/database'); |
| 33 |
print "database size: ",$db->size,"\n"; |
| 34 |
while (my $row = $db->fetch) { |
| 35 |
... |
| 36 |
} |
| 37 |
$db->close; |
| 38 |
|
| 39 |
=head1 FUNCTIONS |
| 40 |
|
| 41 |
=head2 new |
| 42 |
|
| 43 |
Create new input database object. |
| 44 |
|
| 45 |
my $db = new WebPAC::Input( |
| 46 |
format => 'NULL' |
| 47 |
code_page => 'ISO-8859-2', |
| 48 |
); |
| 49 |
|
| 50 |
Optional parametar C<code_page> specify application code page (which will be |
| 51 |
used internally). This should probably be your terminal encoding, and by |
| 52 |
default, it C<ISO-8859-2>. |
| 53 |
|
| 54 |
=cut |
| 55 |
|
| 56 |
sub new { |
| 57 |
my $class = shift; |
| 58 |
my $self = {@_}; |
| 59 |
bless($self, $class); |
| 60 |
|
| 61 |
$self->{'code_page'} ||= 'ISO-8859-2'; |
| 62 |
|
| 63 |
$self ? return $self : return undef; |
| 64 |
} |
| 65 |
|
| 66 |
=head1 MEMORY USAGE |
| 67 |
|
| 68 |
C<low_mem> options is double-edged sword. If enabled, WebPAC |
| 69 |
will run on memory constraint machines (which doesn't have enough |
| 70 |
physical RAM to create memory structure for whole source database). |
| 71 |
|
| 72 |
If your machine has 512Mb or more of RAM and database is around 10000 records, |
| 73 |
memory shouldn't be an issue. If you don't have enough physical RAM, you |
| 74 |
might consider using virtual memory (if your operating system is handling it |
| 75 |
well, like on FreeBSD or Linux) instead of dropping to L<DBM::Deep> to handle |
| 76 |
parsed structure of ISIS database (this is what C<low_mem> option does). |
| 77 |
|
| 78 |
Hitting swap at end of reading source database is probably o.k. However, |
| 79 |
hitting swap before 90% will dramatically decrease performance and you will |
| 80 |
be better off with C<low_mem> and using rest of availble memory for |
| 81 |
operating system disk cache (Linux is particuallary good about this). |
| 82 |
However, every access to database record will require disk access, so |
| 83 |
generation phase will be slower 10-100 times. |
| 84 |
|
| 85 |
Parsed structures are essential - you just have option to trade RAM memory |
| 86 |
(which is fast) for disk space (which is slow). Be sure to have planty of |
| 87 |
disk space if you are using C<low_mem> and thus L<DBM::Deep>. |
| 88 |
|
| 89 |
However, when WebPAC is running on desktop machines (or laptops :-), it's |
| 90 |
highly undesireable for system to start swapping. Using C<low_mem> option can |
| 91 |
reduce WecPAC memory usage to around 64Mb for same database with lookup |
| 92 |
fields and sorted indexes which stay in RAM. Performance will suffer, but |
| 93 |
memory usage will really be minimal. It might be also more confortable to |
| 94 |
run WebPAC reniced on those machines. |
| 95 |
|
| 96 |
|
| 97 |
=head1 AUTHOR |
| 98 |
|
| 99 |
Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >> |
| 100 |
|
| 101 |
=head1 COPYRIGHT & LICENSE |
| 102 |
|
| 103 |
Copyright 2005 Dobrica Pavlinusic, All Rights Reserved. |
| 104 |
|
| 105 |
This program is free software; you can redistribute it and/or modify it |
| 106 |
under the same terms as Perl itself. |
| 107 |
|
| 108 |
=cut |
| 109 |
|
| 110 |
1; # End of WebPAC::Input |
Parent Directory
| 
Revision Log