NAME
Data::StreamSerializer - non-blocking serializer.
SYNOPSIS
use Data::StreamSerializer;
my $sr = new Data::StreamSerializer('You data');
while(defined(my $part = $sr->next)) {
print $socket $part;
}DESCRIPTION
Sometimes You need to serialize a lot of data. If You use 'Dumper' it can take You for much time. If Your code is executed in event machine it can be inadmissible. So using the module You can serialize Your data progressively and do something between serialization itearions.
This module works slower than Data::Dumper, but it can serialize object progressively and You can do something else between serialization iterations.
Recognized types.
HASH
ARRAY
REF
Regexp
SCALAR
METHODS
new
Constructor. All arguments will be serialized.
next
Returns next part of serialized string or undef if all data were serialized.
block_size
Block size for one iteration. Too small value allows You to spend less time for each iteration, but in this case total serialization time will grow. Nod bad choice to set the value between 200 - 2000 bytes (default value is 512). See BENCHMARKS to make a decision.
recursion_depth
If serialized object has recursive references, they will be replaced by empty objects. But if this value is higher than 1 recursion will be reserialized until the value is reached.
Example:
my $t = { a => 'b' };
$t->{c} = $t;This example will be serialized into string:
{"c",{"c",{},"a","b"},"a","b"}and if You increment recursion_depth, this example will be serialized into string: {"c",{"c",{"c",{},"a","b"},"a","b"},"a","b"}
etc.
recursion_detected
Returns TRUE if a recursion was detected.
is_eof
Returns TRUE if eof is reached. If it is TRUE the following next will return undef.
SEE ALSO
BENCHMARKS
You can try a few scripts in benchmark/ directory. There are a few test arrays in this directory.
Here are a few test results of my system.
Array which contains 100 hashes:
$ perl benchmark/vs_dumper.pl -n 1000 -b 512 benchmark/tests/01_100x10
38296 bytes were read
First serializing by eval... done
First serializing by Data::StreamSerializer... done
Starting 1000 iterations for Dumper... done (40.376 seconds)
Starting 1000 iterations for Data::StreamSerializer... done (137.960 seconds)
Dumper statistic:
1000 iterations were done
maximum serialization time: 0.0867 seconds
minimum serialization time: 0.0396 seconds
average serialization time: 0.0404 seconds
Data::StreamSerializer statistic:
1000 iterations were done
58000 SUBiterations were done
maximum serialization time: 0.1585 seconds
minimum serialization time: 0.1356 seconds
average serialization time: 0.1380 seconds
average subiteration time: 0.00238 secondsArray which contains 1000 hashes:
$ perl benchmark/vs_dumper.pl -n 1000 -b 512 benchmark/tests/02_1000x10
355623 bytes were read
First serializing by eval... done
First serializing by Data::StreamSerializer... done
Starting 1000 iterations for Dumper... done (405.334 seconds)
Starting 1000 iterations for Data::StreamSerializer... done (1407.899 seconds)
Dumper statistic:
1000 iterations were done
maximum serialization time: 0.4564 seconds
minimum serialization time: 0.4018 seconds
average serialization time: 0.4053 seconds
Data::StreamSerializer statistic:
1000 iterations were done
520000 SUBiterations were done
maximum serialization time: 2.0050 seconds
minimum serialization time: 1.3862 seconds
average serialization time: 1.4079 seconds
average subiteration time: 0.00271 secondsYou can see that in any cases one iteration gets the same time.
AUTHOR
Dmitry E. Oboukhov, <unera@debian.org>
COPYRIGHT AND LICENSE
Copyright (C) 2011 by Dmitry E. Oboukhov
This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself, either Perl version 5.10.1 or, at your option, any later version of Perl 5 you may have available.
VCS
The project is placed in my git repo. See here: http://git.uvw.ru/?p=data-stream-serializer;a=summary