16 |
|
|
17 |
=head1 VERSION |
=head1 VERSION |
18 |
|
|
19 |
Version 0.13 |
Version 0.14 |
20 |
|
|
21 |
=cut |
=cut |
22 |
|
|
23 |
our $VERSION = '0.13'; |
our $VERSION = '0.14'; |
24 |
|
|
25 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
26 |
|
|
43 |
|
|
44 |
my $db = WebPAC::Input->new( |
my $db = WebPAC::Input->new( |
45 |
module => 'WebPAC::Input::ISIS', |
module => 'WebPAC::Input::ISIS', |
|
low_mem => 1, |
|
46 |
); |
); |
47 |
|
|
48 |
$db->open( path => '/path/to/database' ); |
$db->open( path => '/path/to/database' ); |
62 |
my $db = new WebPAC::Input( |
my $db = new WebPAC::Input( |
63 |
module => 'WebPAC::Input::MARC', |
module => 'WebPAC::Input::MARC', |
64 |
encoding => 'ISO-8859-2', |
encoding => 'ISO-8859-2', |
|
low_mem => 1, |
|
65 |
recode => 'char pairs', |
recode => 'char pairs', |
66 |
no_progress_bar => 1, |
no_progress_bar => 1, |
67 |
); |
); |
73 |
used internally). This should probably be your terminal encoding, and by |
used internally). This should probably be your terminal encoding, and by |
74 |
default, it C<ISO-8859-2>. |
default, it C<ISO-8859-2>. |
75 |
|
|
|
Default is not to use C<low_mem> options (see L<MEMORY USAGE> below). |
|
|
|
|
76 |
C<recode> is optional string constisting of character or words pairs that |
C<recode> is optional string constisting of character or words pairs that |
77 |
should be replaced in input stream. |
should be replaced in input stream. |
78 |
|
|
92 |
|
|
93 |
$log->logconfess("code_page argument is not suppored any more. change it to encoding") if ($self->{lookup}); |
$log->logconfess("code_page argument is not suppored any more. change it to encoding") if ($self->{lookup}); |
94 |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if ($self->{lookup}); |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if ($self->{lookup}); |
95 |
|
$log->logconfess("low_mem argument is not suppored any more. rewrite it to load_row and save_row") if ($self->{low_mem}); |
96 |
|
|
97 |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
98 |
my $module_path = $self->{module}; |
my $module_path = $self->{module}; |
116 |
|
|
117 |
This function will read whole database in memory and produce lookups. |
This function will read whole database in memory and produce lookups. |
118 |
|
|
119 |
|
my $store; # simple in-memory hash |
120 |
|
|
121 |
$input->open( |
$input->open( |
122 |
path => '/path/to/database/file', |
path => '/path/to/database/file', |
123 |
code_page => 'cp852', |
code_page => 'cp852', |
133 |
901 => { '*' => { '^b' => ' ; ' } }, |
901 => { '*' => { '^b' => ' ; ' } }, |
134 |
}, |
}, |
135 |
modify_file => 'conf/modify/mapping.map', |
modify_file => 'conf/modify/mapping.map', |
136 |
|
save_row => sub { |
137 |
|
my $a = shift; |
138 |
|
$store->{ $a->{id} } = $a->{row}; |
139 |
|
}, |
140 |
|
load_row => sub { |
141 |
|
my $a = shift; |
142 |
|
return defined($store->{ $a->{id} }) && |
143 |
|
$store->{ $a->{id} }; |
144 |
|
}, |
145 |
|
|
146 |
); |
); |
147 |
|
|
148 |
By default, C<code_page> is assumed to be C<cp852>. |
By default, C<code_page> is assumed to be C<cp852>. |
163 |
(hopefully) simplier sintax than YAML or perl (see L</modify_file_regex>). This option |
(hopefully) simplier sintax than YAML or perl (see L</modify_file_regex>). This option |
164 |
overrides C<modify_records> if both exists for same input. |
overrides C<modify_records> if both exists for same input. |
165 |
|
|
166 |
|
C<save_row> and C<load_row> are low-level implementation of store engine. Calling convention |
167 |
|
is documented in example above. |
168 |
|
|
169 |
Returns size of database, regardless of C<offset> and C<limit> |
Returns size of database, regardless of C<offset> and C<limit> |
170 |
parametars, see also C<size>. |
parametars, see also C<size>. |
171 |
|
|
192 |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
193 |
} |
} |
194 |
|
|
195 |
|
if ($arg->{load_row} || $arg->{save_row}) { |
196 |
|
$log->logconfess("save_row and load_row must be defined in pair and be CODE") unless ( |
197 |
|
ref($arg->{load_row}) eq 'CODE' && |
198 |
|
ref($arg->{save_row}) eq 'CODE' |
199 |
|
); |
200 |
|
$self->{load_row} = $arg->{load_row}; |
201 |
|
$self->{save_row} = $arg->{save_row}; |
202 |
|
$log->debug("using load_row and save_row instead of in-memory hash"); |
203 |
|
} |
204 |
|
|
205 |
my $filter_ref; |
my $filter_ref; |
206 |
my $recode_regex; |
my $recode_regex; |
207 |
my $recode_map; |
my $recode_map; |
281 |
|
|
282 |
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{encoding}", $self->{stats} ? ' [stats]' : ''); |
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{encoding}", $self->{stats} ? ' [stats]' : ''); |
283 |
|
|
|
# turn on low_mem for databases with more than 100000 records! |
|
|
if (! $self->{low_mem} && $size > 100000) { |
|
|
$log->warn("Using on-disk storage instead of memory for input data. This will affect performance."); |
|
|
$self->{low_mem}++; |
|
|
} |
|
|
|
|
|
# running with low_mem flag? well, use DBM::Deep then. |
|
|
if ($self->{'low_mem'}) { |
|
|
$log->info("running with low_mem which impacts performance (<32 Mb memory usage)"); |
|
|
|
|
|
my $db_file = "data.db"; |
|
|
|
|
|
if (-e $db_file) { |
|
|
unlink $db_file or $log->logdie("can't remove '$db_file' from last run"); |
|
|
$log->debug("removed '$db_file' from last run"); |
|
|
} |
|
|
|
|
|
require DBM::Deep; |
|
|
|
|
|
my $db = new DBM::Deep $db_file; |
|
|
|
|
|
$log->logdie("DBM::Deep error: $!") unless ($db); |
|
|
|
|
|
if ($db->error()) { |
|
|
$log->logdie("can't open '$db_file' under low_mem: ",$db->error()); |
|
|
} else { |
|
|
$log->debug("using file '$db_file' for DBM::Deep"); |
|
|
} |
|
|
|
|
|
$self->{'db'} = $db; |
|
|
} |
|
|
|
|
284 |
# read database |
# read database |
285 |
for (my $pos = $from_rec; $pos <= $to_rec; $pos++) { |
for (my $pos = $from_rec; $pos <= $to_rec; $pos++) { |
286 |
|
|
323 |
} |
} |
324 |
|
|
325 |
# store |
# store |
326 |
if ($self->{low_mem}) { |
if ($self->{save_row}) { |
327 |
$self->{db}->put($pos, $rec); |
$self->{save_row}->({ |
328 |
|
id => $pos, |
329 |
|
row => $rec, |
330 |
|
}); |
331 |
} else { |
} else { |
332 |
$self->{data}->{$pos} = $rec; |
$self->{data}->{$pos} = $rec; |
333 |
} |
} |
418 |
|
|
419 |
my $rec; |
my $rec; |
420 |
|
|
421 |
if ($self->{low_mem}) { |
if ($self->{load_row}) { |
422 |
$rec = $self->{db}->get($mfn); |
$rec = $self->{load_row}->({ id => $mfn }); |
423 |
} else { |
} else { |
424 |
$rec = $self->{data}->{$mfn}; |
$rec = $self->{data}->{$mfn}; |
425 |
} |
} |
664 |
return $regexpes; |
return $regexpes; |
665 |
} |
} |
666 |
|
|
|
=head1 MEMORY USAGE |
|
|
|
|
|
C<low_mem> options is double-edged sword. If enabled, WebPAC |
|
|
will run on memory constraint machines (which doesn't have enough |
|
|
physical RAM to create memory structure for whole source database). |
|
|
|
|
|
If your machine has 512Mb or more of RAM and database is around 10000 records, |
|
|
memory shouldn't be an issue. If you don't have enough physical RAM, you |
|
|
might consider using virtual memory (if your operating system is handling it |
|
|
well, like on FreeBSD or Linux) instead of dropping to L<DBM::Deep> to handle |
|
|
parsed structure of ISIS database (this is what C<low_mem> option does). |
|
|
|
|
|
Hitting swap at end of reading source database is probably o.k. However, |
|
|
hitting swap before 90% will dramatically decrease performance and you will |
|
|
be better off with C<low_mem> and using rest of availble memory for |
|
|
operating system disk cache (Linux is particuallary good about this). |
|
|
However, every access to database record will require disk access, so |
|
|
generation phase will be slower 10-100 times. |
|
|
|
|
|
Parsed structures are essential - you just have option to trade RAM memory |
|
|
(which is fast) for disk space (which is slow). Be sure to have planty of |
|
|
disk space if you are using C<low_mem> and thus L<DBM::Deep>. |
|
|
|
|
|
However, when WebPAC is running on desktop machines (or laptops :-), it's |
|
|
highly undesireable for system to start swapping. Using C<low_mem> option can |
|
|
reduce WecPAC memory usage to around 64Mb for same database with lookup |
|
|
fields and sorted indexes which stay in RAM. Performance will suffer, but |
|
|
memory usage will really be minimal. It might be also more confortable to |
|
|
run WebPAC reniced on those machines. |
|
|
|
|
|
|
|
667 |
=head1 AUTHOR |
=head1 AUTHOR |
668 |
|
|
669 |
Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >> |
Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >> |