NAME
PAR - Perl Archive Toolkit
SYNOPSIS
(If you want to make an executable that contains all module, scripts and data files, please consult the pp utility instead. pp used to be part of the PAR distribution but is now shipped as part of the PAR::Packer distribution instead.)
Following examples assume a foo.par file in Zip format.
To use Hello.pm from ./foo.par:
% perl -MPAR=./foo.par -MHello
% perl -MPAR=./foo -MHello
# the .par part is optional
Same thing, but search foo.par in @INC
:
% perl -MPAR -Ifoo.par -MHello
% perl -MPAR -Ifoo -MHello
# ditto
Following paths inside the PAR file are searched:
/lib/
/arch/
/i386-freebsd/
# i.e. $Config{archname}
/5.8.0/
# i.e. $Config{version}
/5.8.0/i386-freebsd/
# both of the above
/
PAR files may also (recursively) contain other PAR files. All files under following paths will be considered as PAR files and searched as well:
/par/i386-freebsd/
# i.e. $Config{archname}
/par/5.8.0/
# i.e. $Config{version}
/par/5.8.0/i386-freebsd/
# both of the above
/par/
Run script/test.pl or test.pl from foo.par:
% perl -MPAR foo.par test.pl
# only when $0 ends in '.par'
However, if the .par archive contains either script/main.pl or main.pl, then it is used instead:
% perl -MPAR foo.par test.pl
# runs main.pl; @ARGV is 'test.pl'
Use in a program:
You can also use wildcard characters:
Since version 0.950, you can also use a different syntax for loading .par archives:
Why? Because you can also do this:
Foo::Bar will be searched in the system libs first and loaded from foo.par if it wasn't found!
This will load foo.par as usual and then execute the script/myscript file from the archive. Note that your program will not regain control. When script/myscript exits, so does your main program. To make this more useful, you can defer this to runtime: (otherwise equivalent)
require
PAR;
PAR->
import
( {
file
=>
'foo.par'
,
run
=>
'myscript'
} );
If you have PAR::Repository::Client installed, you can do this:
And PAR will fetch any modules you don't have from the specified PAR repository. For details on how this works, have a look at the SEE ALSO section below. Instead of an URL or local path, you can construct an PAR::Repository::Client object manually and pass that to PAR. If you specify the install => 1
option in the use PAR
line above, the distribution containing Module
will be permanently installed on your system. (use PAR { repository => 'http://foo/bar', install => 1 };
)
Furthermore, there is an upgrade => 1
option that checks for upgrades in the repository in addition to installing. Please note that an upgraded version of a module is only loaded on the next run of your application.
Adding the dependencies => 1
option will enable PAR::Repository::Client's static dependency resolution (PAR::Repository::Client 0.23 and up).
Finally, you can combine the run
and repository
options to run an application directly from a repository! (And you can add the install
option, too.)
# Will not reach this point as we executed my_app,
DESCRIPTION
This module lets you use special zip files, called Perl Archives, as libraries from which Perl modules can be loaded.
It supports loading XS modules by overriding DynaLoader bootstrapping methods; it writes shared object file to a temporary file at the time it is needed.
A .par file is mostly a zip of the blib/ directory after the build process of a CPAN distribution. To generate a .par file yourself, all you have to do is compress the modules under arch/ and lib/, e.g.:
% perl Makefile.PL
% make
% cd blib
% zip -r mymodule.par arch/ lib/
Afterward, you can just use mymodule.par anywhere in your @INC
, use PAR, and it will Just Work. Support for generating .par files is going to be in the next (beyond 0.2805) release of Module::Build.
For convenience, you can set the PERL5OPT
environment variable to -MPAR
to enable PAR
processing globally (the overhead is small if not used); setting it to -MPAR=/path/to/mylib.par
will load a specific PAR file. Alternatively, consider using the par.pl utility bundled with the PAR::Packer distribution, or using the self-contained parl utility which is also distributed with PAR::Packer on machines without PAR.pm installed.
Note that self-containing scripts and executables created with par.pl and pp may also be used as .par archives:
% pp -o packed.exe source.pl
# generate packed.exe (see PAR::Packer)
% perl -MPAR=packed.exe other.pl
# this also works
% perl -MPAR -Ipacked.exe other.pl
# ditto
Please see "SYNOPSIS" for most typical use cases.
NOTES
Settings in META.yml packed inside the PAR file may affect PAR's operation. For example, pp provides the -C
(--clean
) option to control the default behavior of temporary file creation.
Currently, pp-generated PAR files may attach four PAR-specific attributes in META.yml:
par:
clean: 0
# default value of PAR_CLEAN
signature:
''
# key ID of the SIGNATURE file
verbatim: 0
# was packed prerequisite's PODs preserved?
version: x.xx
# PAR.pm version that generated this PAR
User-defined environment variables, like PAR_GLOBAL_CLEAN, always overrides the ones set in META.yml. The algorithm for generating caching/temporary directory is as follows:
If PAR_GLOBAL_TEMP is specified, use it as the cache directory for extracted libraries, and do not clean it up after execution.
If PAR_GLOBAL_TEMP is not set, but PAR_GLOBAL_CLEAN is specified, set PAR_GLOBAL_TEMP to
TEMP/par-USERHEX/temp-PID/
, cleaning it after execution.If both are not set, use
TEMP/par-USERHEX/cache-HASH/
as the PAR_GLOBAL_TEMP, reusing any existing files inside.
Here is a description of the variables the previous paths.
TEMP is a temporary directory, which can be set via
$ENV{PAR_GLOBAL_TMPDIR}
,$ENV{TMPDIR}
,$ENV{TEMPDIR}
,$ENV{TEMP}
or$ENV{TMP}
, in that order of priority. If none of those are set, C:\TEMP, /tmp are checked. If neither of them exists, . is used.USERHEX is derived from the user name, or SYSTEM if none can be found. On Win32, this is
$Win32::LoginName
. On Unix, this is$ENV{USERNAME}
or$ENV{USER}
. Encode the raw bytes of the user name each as two hex digits to obtain USERHEX.PID is the process ID. Forked children use the parent's PID.
HASH is a crypto-hash of the entire par file or executable, calculated at creation time. This value can be overloaded with
pp
's --tempdir parameter.
By default, PAR strips POD sections from bundled modules. In case that causes trouble, you can turn this off by setting the environment variable PAR_VERBATIM
to 1
.
import options
When you "use PAR {...}" or call PAR->import({...}), the following options are available.
- file
-
The par filename.
You must pass one option of either 'file' or 'repository'.
- repository
-
A par repository (exclusive of file)
- fallback
-
Search the system
@INC
before the par.Off by default for loading .par files via
file =
...>. On by default for PAR repositories.To prefer loading modules from a repository over the locally installed modules, you can load the repository as follows:
- run
-
The name of a script to run in the par. Exits when done.
- no_shlib_unpack
-
Skip unpacking bundled dynamic libraries from shlib/$archname. The client may have them installed, or you may wish to cache them yourself. In either case, they must end up in the standard install location (such as /usr/local/lib/) or in $ENV{PAR_TEMP} before you require the module which needs them. If they are not accessible before you require the dependent module, perl will die with a message such as "cannot open shared object file..."
SEE ALSO
The PAR::Packer distribution which contains the packaging utilities: par.pl, parl, pp.
PAR::Dist for details on PAR distributions.
PAR::Repository::Client for details on accessing PAR repositories. PAR::Repository for details on how to set up such a repository.
Archive::Zip, "require" in perlfunc
ex::lib::zip, Acme::use::strict::with::pride
Steffen Mueller has detailed slides on using PAR for application deployment at http://steffen-mueller.net/talks/appdeployment/.
PAR supports the prefork module. It declares various run-time dependencies so you can use the prefork module to get streamlined processes in a forking environment.
ACKNOWLEDGMENTS
Nicholas Clark for pointing out the mad source filter hook within the (also mad) coderef @INC
hook, as well as (even madder) tricks one can play with PerlIO to avoid source filtering.
Ton Hospel for convincing me to ditch the Filter::Simple
implementation.
Uri Guttman for suggesting read_file
and par_handle
interfaces.
Antti Lankila for making me implement the self-contained executable options via par.pl -O
.
See the AUTHORS file in the distribution for a list of people who have sent helpful patches, ideas or comments.
AUTHORS
Audrey Tang <cpan@audreyt.org>
Steffen Mueller <smueller@cpan.org>
You can write to the mailing list at <par@perl.org>, or send an empty mail to <par-subscribe@perl.org> to participate in the discussion. Archives of the mailing list are available at <https://www.mail-archive.com/par@perl.org/> or <https://groups.google.com/g/perl.par>.
Please submit bug reports to <https://github.com/rschupp/PAR/issues>. If you need support, however, joining the <par@perl.org> mailing list is preferred.
COPYRIGHT
Copyright 2002-2010 by Audrey Tang <cpan@audreyt.org>. Copyright 2005-2010 by Steffen Mueller <smueller@cpan.org>
This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.
See LICENSE.