######################################################################
   Archive::Tar::Wrapper 0.23
######################################################################

NAME
   Archive::Tar::Wrapper - API wrapper around the 'tar' utility

SYNOPSIS
       use Archive::Tar::Wrapper;

       my $arch = Archive::Tar::Wrapper->new();

           # Open a tarball, expand it into a temporary directory
       $arch->read("archive.tgz");

           # Iterate over all entries in the archive
       $arch->list_reset(); # Reset Iterator
                            # Iterate through archive
       while(my $entry = $arch->list_next()) {
           my($tar_path, $phys_path) = @$entry;
           print "$tar_path\n";
       }

           # Get a huge list with all entries
       for my $entry (@{$arch->list_all()}) {
           my($tar_path, $real_path) = @$entry;
           print "Tarpath: $tar_path Tempfile: $real_path\n";
       }

           # Add a new entry
       $arch->add($logic_path, $file_or_stringref);

           # Remove an entry
       $arch->remove($logic_path);

           # Find the physical location of a temporary file
       my($tmp_path) = $arch->locate($tar_path);

           # Create a tarball
       $arch->write($tarfile, $compress);

DESCRIPTION
   Archive::Tar::Wrapper is an API wrapper around the 'tar' command line
   utility. It never stores anything in memory, but works on temporary
   directory structures on disk instead. It provides a mapping between the
   logical paths in the tarball and the 'real' files in the temporary
   directory on disk.

   It differs from Archive::Tar in two ways:

   *   Archive::Tar::Wrapper doesn't hold anything in memory. Everything is
       stored on disk.

   *   Archive::Tar::Wrapper is 100% compliant with the platform's "tar"
       utility, because it uses it internally.

METHODS
   my $arch = Archive::Tar::Wrapper->new()
       Constructor for the tar wrapper class. Finds the "tar" executable by
       searching "PATH" and returning the first hit. In case you want to
       use a different tar executable, you can specify it as a parameter:

           my $arch = Archive::Tar::Wrapper->new(tar => '/path/to/tar');

       Since "Archive::Tar::Wrapper" creates temporary directories to store
       tar data, the location of the temporary directory can be specified:

           my $arch = Archive::Tar::Wrapper->new(tmpdir => '/path/to/tmpdir');

       Tremendous performance increases can be achieved if the temporary
       directory is located on a ram disk. Check the "Using RAM Disks"
       section below for details.

       Additional options can be passed to the "tar" command by using the
       "tar_read_options" and "tar_write_options" parameters. Example:

            my $arch = Archive::Tar::Wrapper->new(
                          tar_read_options => "p"
                       );

       will use "tar xfp archive.tgz" to extract the tarball instead of
       just "tar xf archive.tgz". Gnu tar supports even more options, these
       can be passed in via

            my $arch = Archive::Tar::Wrapper->new(
                           tar_gnu_read_options => ["--numeric-owner"],
                       );

       Similarily, "tar_gnu_write_options" can be used to provide
       additional options for Gnu tar implementations. For example, the tar
       object

           my $tar = Archive::Tar::Wrapper->new(
                         tar_gnu_write_options => ["--exclude=foo"],
                     );

       will call the "tar" utility internally like

           tar cf tarfile --exclude=foo ...

       when the "write" method gets called.

       By default, the "list_*()" functions will return only file entries.
       Directories will be suppressed. To have "list_*()" return
       directories as well, use

            my $arch = Archive::Tar::Wrapper->new(
                          dirs  => 1
                       );

       If more files are added to a tarball than the command line can
       handle, "Archive::Tar::Wrapper" will switch from using the command

           tar cfv tarfile file1 file2 file3 ...

       to

           tar cfv tarfile -T filelist

       where "filelist" is a file containing all file to be added. The
       default for this switch is 512, but it can be changed by setting the
       parameter "max_cmd_line_args":

            my $arch = Archive::Tar::Wrapper->new(
                max_cmd_line_args  => 1024
            );

   $arch->read("archive.tgz")
       "read()" opens the given tarball, expands it into a temporary
       directory and returns 1 on success und "undef" on failure. The
       temporary directory holding the tar data gets cleaned up when $arch
       goes out of scope.

       "read" handles both compressed and uncompressed files. To find out
       if a file is compressed or uncompressed, it tries to guess by
       extension, then by checking the first couple of bytes in the
       tarfile.

       If only a limited number of files is needed from a tarball, they can
       be specified after the tarball name:

           $arch->read("archive.tgz", "path/file.dat", "path/sub/another.txt");

       The file names are passed unmodified to the "tar" command, make sure
       that the file paths match exactly what's in the tarball, otherwise
       "read()" will fail.

   $arch->list_reset()
       Resets the list iterator. To be used before the first call to
       $arch-list_next()>.

   my($tar_path, $phys_path, $type) = $arch->list_next()
       Returns the next item in the tarfile. It returns a list of three
       scalars: the relative path of the item in the tarfile, the physical
       path to the unpacked file or directory on disk, and the type of the
       entry (f=file, d=directory, l=symlink). Note that by default,
       Archive::Tar::Wrapper won't display directories, unless the "dirs"
       parameter is set when running the constructor.

   my $items = $arch->list_all()
       Returns a reference to a (possibly huge) array of items in the
       tarfile. Each item is a reference to an array, containing two
       elements: the relative path of the item in the tarfile and the
       physical path to the unpacked file or directory on disk.

       To iterate over the list, the following construct can be used:

               # Get a huge list with all entries
           for my $entry (@{$arch->list_all()}) {
               my($tar_path, $real_path) = @$entry;
               print "Tarpath: $tar_path Tempfile: $real_path\n";
           }

       If the list of items in the tarfile is big, use "list_reset()" and
       "list_next()" instead of "list_all".

   $arch->add($logic_path, $file_or_stringref, [$options])
       Add a new file to the tarball. $logic_path is the virtual path of
       the file within the tarball. $file_or_stringref is either a scalar,
       in which case it holds the physical path of a file on disk to be
       transferred (i.e. copied) to the tarball. Or it is a reference to a
       scalar, in which case its content is interpreted to be the data of
       the file.

       If no additional parameters are given, permissions and user/group id
       settings of a file to be added are copied. If you want different
       settings, specify them in the options hash:

           $arch->add($logic_path, $stringref,
                      { perm => 0755, uid => 123, gid => 10 });

       If $file_or_stringref is a reference to a Unicode string, the
       "binmode" option has to be set to make sure the string gets written
       as proper UTF-8 into the tarfile:

           $arch->add($logic_path, $stringref, { binmode => ":utf8" });

   $arch->remove($logic_path)
       Removes a file from the tarball. $logic_path is the virtual path of
       the file within the tarball.

   $arch->locate($logic_path)
       Finds the physical location of a file, specified by $logic_path,
       which is the virtual path of the file within the tarball. Returns a
       path to the temporary file "Archive::Tar::Wrapper" created to
       manipulate the tarball on disk.

   $arch->write($tarfile, $compress)
       Write out the tarball by tarring up all temporary files and
       directories and store it in $tarfile on disk. If $compress holds a
       true value, compression is used.

   $arch->tardir()
       Return the directory the tarball was unpacked in. This is sometimes
       useful to play dirty tricks on "Archive::Tar::Wrapper" by
       mass-manipulating unpacked files before wrapping them back up into
       the tarball.

   $arch->is_gnu()
       Checks if the tar executable is a GNU tar by running 'tar --version'
       and parsing the output for "GNU".

Using RAM Disks
   On Linux, it's quite easy to create a RAM disk and achieve tremendous
   speedups while untarring or modifying a tarball. You can either create
   the RAM disk by hand by running

      # mkdir -p /mnt/myramdisk
      # mount -t tmpfs -o size=20m tmpfs /mnt/myramdisk

   and then feeding the ramdisk as a temporary directory to
   Archive::Tar::Wrapper, like

      my $tar = Archive::Tar::Wrapper->new( tmpdir => '/mnt/myramdisk' );

   or using Archive::Tar::Wrapper's built-in option 'ramdisk':

      my $tar = Archive::Tar::Wrapper->new(
          ramdisk => {
              type => 'tmpfs',
              size => '20m',   # 20 MB
          },
      );

   Only drawback with the latter option is that creating the RAM disk needs
   to be performed as root, which often isn't desirable for security
   reasons. For this reason, Archive::Tar::Wrapper offers a utility
   functions that mounts the ramdisk and returns the temporary directory
   it's located in:

         # Create new ramdisk (as root):
       my $tmpdir = Archive::Tar::Wrapper->ramdisk_mount(
           type => 'tmpfs',
           size => '20m',   # 20 MB
       );

         # Delete a ramdisk (as root):
       Archive::Tar::Wrapper->ramdisk_unmount();

   Optionally, the "ramdisk_mount()" command accepts a "tmpdir" parameter
   pointing to a temporary directory for the ramdisk if you wish to set it
   yourself instead of letting Archive::Tar::Wrapper create it
   automatically.

KNOWN LIMITATIONS
   *   Currently, only "tar" programs supporting the "z" option (for
       compressing/decompressing) are supported. Future version will use
       "gzip" alternatively.

   *   Currently, you can't add empty directories to a tarball directly.
       You could add a temporary file within a directory, and then
       "remove()" the file.

   *   If you delete a file, the empty directories it was located in stay
       in the tarball. You could try to "locate()" them and delete them.
       This will be fixed, though.

   *   Filenames containing newlines are causing problems with the list
       iterators. To be fixed.

   *   If you ask Archive::Tar::Wrapper to add a file to a tarball, it
       copies it into a temporary directory and then calls the system tar
       to wrap up that directory into a tarball.

       This approach has limitations when it comes to file permissions: If
       the file to be added belongs to a different user/group,
       Archive::Tar::Wrapper will adjust the uid/gid/permissions of the
       target file in the temporary directory to reflect the original
       file's settings, to make sure the system tar will add it like that
       to the tarball, just like a regular tar run on the original file
       would. But this will fail of course if the original file's uid is
       different from the current user's, unless the script is running with
       superuser rights. The tar program by itself (without
       Archive::Tar::Wrapper) works differently: It'll just make a note of
       a file's uid/gid/permissions in the tarball (which it can do without
       superuser rights) and upon extraction, it'll adjust the permissions
       of newly generated files if the -p option is given (default for
       superuser).

BUGS
   Archive::Tar::Wrapper doesn't currently handle filenames with embedded
   newlines.

LEGALESE
   Copyright 2005 by Mike Schilli, all rights reserved. This program is
   free software, you can redistribute it and/or modify it under the same
   terms as Perl itself.

AUTHOR
   2005, Mike Schilli <[email protected]>

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.