I have the following perl script, that takes in a parameters' file and stores it into a hash. I want to modify & pass this hash to another perl script that I am calling using the system command:
script1.pl
#!/usr/bin/perl -w
# usage perl script1.pl script1.params
# script1.params file looks like this:
# PROJECTNAME=>project_dir
# FASTALIST=>samples_fastq.csv
use Data::Dumper;
my $paramfile = $ARGV[0];
# open parameter file
open PARAM, $paramfile or die print $!;
# save it in a hash
my %param;
while(<PARAM>)
{
chomp;
#r = split('=>');
$param{$r[0]}=$r[1];
}
# define directories
# add to parameters' hash
$param{'INDIR'} = $param{'PROJECTNAME'}.'/input';
$param{'OUTDIR'} = $param{'PROJECTNAME'}.'/output';
.... do something ...
# #samples is a list of sample names
foreach (#samples)
{
# for each sample, pass the hash values & sample name to a separate script
system('perl script2.pl <hash> $_');
}
script2.pl
#!/usr/bin/perl -w
use Data::Dumper;
## usage <script2.pl> <hash> <samplename>
# something like getting and printing the hash
my #string = $ARGV[0];
print #string;
If you can help me showing how to pass and get the hash object (something simple like printing the hash object in the second script would do), then I'd appreciate your help.
Thanks!
What you're looking for is something called serialisation. It's difficult to directly represent a memory structure in such a way as to pass it between processes, because of all sorts of fun things like pointers and buffers.
So you need to turn your hash into something simple enough to hand over in a single go.
Three key options for this in my opinion:
Storable - a perl core module that lets you freeze and thaw a data structure for this sort of purpose.
JSON - a text based representation of a hash-like structure.
XML - bit like JSON, but with slightly different strengths/weaknesses.
Which you should use depends a little on how big your data structure is.
Storable is probably the simplest, but it's not going to be particularly portable.
There's also Data::Dumper that's an option too, as it prints data structures. Generally though, I'd suggest that has all the downsides of all the above - you still need to parse it like JSON/XML but it's also not portable.
Example using Storable:
use strict;
use warnings;
use Storable qw ( freeze );
use MIME::Base64;
my %test_hash = (
"fish" => "paste",
"apples" => "pears"
);
my $frozen = encode_base64 freeze( \%test_hash );
system( "perl", "some_other_script.pl", $frozen );
Calling:
use strict;
use warnings;
use Storable qw ( thaw );
use Data::Dumper;
use MIME::Base64;
my ($imported_scalar) = #ARGV;
print $imported_scalar;
my $thing = thaw (decode_base64 $imported_scalar ) ;
print Dumper $thing;
Or:
my %param = %{ thaw (decode_base64 $imported_scalar ) };
print Dumper \%param;
This will print:
BAoIMTIzNDU2NzgEBAQIAwIAAAAKBXBhc3RlBAAAAGZpc2gKBXBlYXJzBgAAAGFwcGxlcw==
$VAR1 = {
'apples' => 'pears',
'fish' => 'paste'
};
Doing the same with JSON - which has the advantage of being passed as plain text, and in a general purpose format. (Most languages can parse JSON):
#!/usr/bin/env perl
use strict;
use warnings;
use JSON;
my %test_hash = (
"fish" => "paste",
"apples" => "pears"
);
my $json_text = encode_json ( \%test_hash );
print "Encoded: ",$json_text,"\n";
system( "perl", "some_other_script.pl", quotemeta $json_text );
Calling:
#!/usr/bin/env perl
use strict;
use warnings;
use JSON;
use Data::Dumper;
my ($imported_scalar) = #ARGV;
$imported_scalar =~ s,\\,,g;
print "Got: ",$imported_scalar,"\n";
my $thing = decode_json $imported_scalar ;
print Dumper $thing;
Need the quotemeta and the removal of slashes unfortunately, because the shell interpolates them. This is the common problem if you're trying to do this sort of thing.
Related
%hash = ('abc' => 123, 'def' => [4,5,6]);
how can I store above hash in file using data dumper in Perl
Files can only contain sequences of bytes, so you need to convert the data structure into a sequence of bytes somehow. This process is called serialization.
The possibilities available to you are endless, but a few are worth mentioning:
JSON is a very common choice.
YAML is more flexible.
Storable is specifically made for Perl data structures.
There is also Data::Dumper, as you say.
use Data::Dumper qw( );
sub serialize {
my ($x) = #_;
local $Data::Dumper::Purity = 1; # Required for some data structures.
local $Data::Dumper::Useqq = 1; # Optional. Limits output to ASCII.
local $Data::Dumper::Sortkeys = 1; # Optional. Makes revision control easier.
return Data::Dumper->Dump([$x], ["x"]);
}
print($fh serialize($x));
Data::Dumper isn't a particularly good choice, since there's no existing module to safely deserialize the structure in Perl[1], and there's even less support outside of Perl.
sub deserialize {
my ($s) = #_;
my $x;
eval($s); # XXX Unsafe!
die($#) if $#;
return $x;
}
If you're ok with limiting yourself to data structure JSON can serialize (by setting Purity to 0), then you could use Data::Undump to safely deserialize. But then why not just use JSON?!
use Data::Dumper
open (FL, ">", "file.txt") or die "Cant open file $! ";
print FL Dumper \%hash;
close FL;
"how can I store above hash in file using data dumper in Perl"
Store it as JSON so it can be read back by (almost) anything, using Data::Dumper configured to print JSON.
use strict;
use warnings;
use Data::Dumper;
local $Data::Dumper::Pair = ' : ';
local $Data::Dumper::Quotekeys = 1;
local $Data::Dumper::Useqq = 1;
local $Data::Dumper::Terse = 1;
my %hash = ('abc' => 123, 'def' => [4,5,6]);
open my $file, '>', 'foo.json' or die $!;
print $file Dumper \%hash;
Output:
$ cat foo.json
{
"def" : [
4,
5,
6
],
"abc" : 123
}
(Note: I would of course rather use a dedicated JSON-handling module for this, but you asked ....)
I am storing a hash in a script like so (part of code left out):
use strict;
use warnings;
use utf8;
use Storable qw(nstore_fd);
open( my $fh, ">", 'hash.hash');
my $hash_ref = \%hash;
nstore_fd($hash_ref, $fh);
close $fh;
Which creates a file hash.hash. In another file, then, I try to 'import' the hash and read its values to be used (e.g. looping the hash, reading values, checking existence ...).
I tried something simple, i.e. read the hash, dereference and read its keys, and print those out. However I get the error Not a HASH reference at CreateSql.pl line 12. It is a scalar. But I don't see why. The documentation of the module states that a reference is returned.
To retrieve data stored to disk, use retrieve with a file name. The
objects stored into that file are recreated into memory for you, and a
reference to the root object is returned.
So what did I do wrong, and how do I get access to the stored hash?
use strict;
use warnings;
use utf8;
use Data::Dumper;
use Storable qw(retrieve);
my ($hash_path) = #ARGV;
my $hash_ref = retrieve($hash_path);
my #keys = keys % { $hash_ref }; # throws error
print Dumper(\#keys);
You really must show the code that has given you the problems that you describe, otherwise it becomes very hard to help you. The code in your question doesn't compile. If you want to make changes to the program before you publish it, the test it again to make sure that it at least compiles, and also display the problems you're reporting
The main problem is that you are writing binary data to a file opened as text. If you change open( my $fh, ">", 'hash.hash') to open( my $fh, ">:raw", 'hash.hash') then everything should work
This variant of your own code correctly stores and retrieves some sample data
use strict;
use warnings 'all';
use Storable qw/ nstore_fd retrieve /;
use constant STORAGE => 'hash.hash';
my %hash = ( a => 'b', c => 'd' );
{
open my $fh, '>:raw', STORAGE or die $!;
nstore_fd \%hash, $fh;
}
my $hash_ref = retrieve STORAGE;
use Data::Dump;
dd $hash_ref;
my #keys = keys % { $hash_ref }; # throws error
dd \#keys;
output
{ a => "b", c => "d" }
["a", "c"]
However, you've made some strange choices. There's no need to use an nstore function to store the data in "network order", and you could have avoided the problem altogether if you had let the module open the file itself
Here's a variant that just uses store and retrieve. The output is identical to that of the code above
use strict;
use warnings 'all';
use Storable qw/ store retrieve /;
use constant STORAGE => 'hash.hash';
my %hash = ( a => 'b', c => 'd' );
store \%hash, STORAGE;
my $hash_ref = retrieve STORAGE;
use Data::Dump;
dd $hash_ref;
my #keys = keys % { $hash_ref }; # throws error
dd \#keys;
I want to (ab-)use the global %ENV to store a hash. This seems to work differently for %ENV than for ordinary hashes. in the program below, the $ENV{myhash} still contains 'myhash' => 'HASH(0x7ffad8807380)' and the %ahash is still around. is it possible to convert the hex address back to point at its location, instead of just containing the string? I guess I could serialize and unserialize the hash instead. what is the right way to do this?
#!/usr/bin/perl -w
use common::sense;
use Data::Dumper;
my %QENV = ( nohash => 'noh' );
my %ahash= ( hv1 => 'htext1', hv2 => 'htext2' );
$QENV{'myhash'} = \%ahash;
print "works: ". Dumper(\%QENV)."\n\n\n";
$ENV{'myhash'} = \%ahash;
print "fails: ". Dumper(\%ENV)."\n";
%ENV is a magical hash. It reflects the process's environment. Reading from it reads from the environment, and writing to it changes the environment.
If you can guarantee the referenced variable is still alive (by it still being in scope or by it having its REFCNT increased), you can indeed create a reference to it from the address.
use strict;
use warnings;
use Data::Dumper;
use Inline C => <<'__EOS__';
SV* _newRV(IV iv) {
return newRV((SV*)iv);
}
__EOS__
my %hash = ( a => 1, b => 2 );
my $ref = \%hash;
my $ref_string = "$ref";
my $addr = hex( ( $ref_string =~ /\((.*)\)/ )[0] );
my $ref2 = _newRV($addr);
print(Dumper($ref2));
I have no idea why you'd want to do this. It would not permit another process to access the data since one process can't access the memory of another.
You seem to want to share data. Here's an example I put out there often that shows how to store data in a JSON file, then retrieve it. JSON is cross-language, so the data can be used by many programming languages, not just Perl. Although this example is within a single script, imagine it being two different Perl applications:
use warnings;
use strict;
use JSON;
my $file = 'data.json';
my %h = (
foo => 'bar',
raz => {
ABC => 123,
XYZ => [4, 5, 6],
}
);
my $json = encode_json \%h;
# write the JSON to a file, and close it
open my $fh, '>', $file or die $!;
print $fh $json;
close $fh or die $!;
# open the JSON file, and read it
open my $json_file, '<', $file or die $!;
my $read_json;
{
local $/;
$read_json = <$json_file>;
}
my $perl_hash = decode_json $read_json;
I want to pass a hash reference as an argument from one perl script (script1.pl) to another perl script (script2.pl). This is how my code looks:
----------------------------script1.pl---------------------------------
#!/usr/bin/perl -w
use strict;
use warnings;
my %hash = (
'a' => "Harsha",
'b' => "Manager"
);
my $ref = \%hash;
system "perl script2.pl $ref";
----------------------------script2.pl---------------------------------
#!/usr/bin/perl -w
use strict;
use warnings;
my %hash = %{$ARGV[0]};
my $string = "a";
if (exists($hash{$string})){
print "$string = $hash{$string}\n";
}
And this is the output error:
sh: -c: line 0: syntax error near unexpected token `('
sh: -c: line 0: `perl script2.pl HASH(0x8fbed0)'
I can't figure out the right way to pass the reference.
A hash is an in memory data structure. Processes 'own' their own memory space, and other processes can't just access it. If you think about it, I'm sure you'll spot why quite quickly.
A hash reference is an address of that memory location. Even if the other process could 'understand' it, it still wouldn't be able to access the memory space.
What we're talking about here is actually quite a big concept - Inter Process Communication or IPC - so much so there's a whole chapter of the documentation about it, called perlipc.
The long and short of it is this - you can't do what you're trying to do. Sharing memory between processes is much more difficult than you imagine.
What you can do is transfer the data back and forth - not by reference, but the actual information contained.
I would suggest that for your example, the tool for the job is JSON, because then you can encode and decode your hash:
#!/usr/bin/perl -w
use strict;
use warnings;
use JSON;
my %hash = (
'a' => "Harsha",
'b' => "Manager"
);
my $json_string = to_json( \%hash );
print $json_string;
This gives:
{"b":"Manager","a":"Harsha"}
Then your can 'pass' your $json_string - either on the command line, although bear in mind that any spaces in it confuses #ARGV a bit if you're not careful - or via STDIN.
And then decode in your sub process:
use strict;
use warnings;
use JSON;
my $json_string = '{"b":"Manager","a":"Harsha"}';
my $json = from_json ( $json_string );
my $string = "a";
if (exists($json -> {$string} )){
print "$string = ",$json -> {$string},"\n";
}
(You can make it more similar to your code by doing:
my $json = from_json ( $json_string );
my %hash = %$json;
Other options would be:
use Storable - either freezing and thawing ( memory) or storing and retrieving (disk)
use IPC::Open2 and send data on STDIN.
There's a variety of options really - have a look at perlipc. But it's not as simple a matter as 'just passing a reference' unfortunately.
Use Storable to store data in first script and retrieve it from other.
firstscript.pl
store (\%hash, "/home/chankey/secondscript.$$") or die "could not store";
system("perl", "secondscript.pl", $$) == 0 or die "error";
secondscript.pl
my $parentpid = shift;
my $ref = retrieve("/home/chankey/secondscript.$parentpid") or die "couldn't retrieve";
print Dumper $ref;
You've received the %hash in $ref. Now use it the way you want.
You can't pass a reference from one script to another - that reference only has meaning within the currently running instance of perl.
You would need to "serialise" the data in the first script, and then "deserialise" it in the second.
Your way of calling perl file is wrong.
Just change the way of calling it and you are done.
Script1.pl
---------------------------------
#!/usr/bin/perl -w
use strict;
use warnings;
my %hash = (
'a' => "Harsha",
'b' => "Manager"
);
system("perl","script2.pl",%hash);
Use this %hash in another perl script as shown below.
Script2.pl
----------------------------------
#!/usr/bin/perl -w
use strict;
use warnings;
my %hash = #ARGV;
my $string = "a";
if (exists($hash{$string})){
print "$string = $hash{$string}\n";
}
OutPut is
a = Harsha
I have a main setup script which sets up the test env and stores data in some variables:
package main;
use Test::Harness;
our $foo, $xyz, $pqr;
($foo, $xyz, $pqr) = &subroutinesetup();
# ^ here
#test_files = glob "t/*";
print "Executing test #test\n";
runtests(#test_files);
In the test folder I have a testsuite (t/testsuite1.t, testsuite2.t etc.).
How can I access the value of $foo inside the testsuite1.t?
package main;
use Test::More;
$actual = getActual();
is($foo, $actual, passfoor);
# ^ here
done_testing();
Use Storable to store data in first script and retrieve it from other.
main.pl
($foo, $xyz, $pqr) = &subroutinesetup();
store ($foo, "/home/chankey/testsuite.$$") or die "could not store";
system("perl", "testsuite.pl", $$) == 0 or die "error";
testsuite.pl
my $parentpid = shift;
my $ref = retrieve("/home/chankey/testsuite.$parentpid") or die "couldn't retrieve";
print Dumper $ref;
You've received the $foo in $ref. Now use it the way you want.
You can't share a variable directly, because a new Perl process is started for each test file.
As noted in the documentation of Test::Harness, you should switch to TAP::Harness. It's more flexible: for example, it provides the test_args mechanism to pass arguments to test scripts.
$ cat 1.pl
#!/usr/bin/perl
use warnings;
use strict;
use TAP::Harness;
my $harness = 'TAP::Harness'->new({
test_args => [ qw( propagate secret ) ]
});
$harness->runtests('1.t');
__END__
$ cat 1.t
#!/usr/bin/perl
use warnings;
use strict;
use Test::More;
my %args = #ARGV;
is($args{propagate}, 'secret', 'propagated');
done_testing();