|Archivers
|
|Mark H. Solsman
|Documentation Training and Publications, Center for Academic Computing
|mhs108@psu.edu   9/14/93
I
 
 
Introduction
 
Program and data file archives exist in many forms today.
Public-access bulletin-board systems (BBS), BITNET-based
LISTSERV systems and Internet-based anonymous FTP sites are
just three of the more popular places where files are stored
for remote access and retrieval by computer users.
 
Archiving and compressing are separate functions that may be
performed by individual programs or by a single program,
depending on how the program was designed.  A third
function, encoding, is also sometimes used.  Following are
definitions of these functions:
 
Archiving means putting a number of files into a single file
for convenience.  Since many programs are actually comprised
of multiple files it is much easier to retrieve programs if
their component files have been archived into a single file.
Getting the individual files out of an archive is a process
known as "extracting" and is performed by the archiving
program.
 
Compressing means shifting the bits of a file in a way that
makes the file smaller.  This is done to reduce storage and
transfer expense.  Various compression algorithms may be
used to achieve compression.  A compressed file is typically
unreadable, or in the case of a program, unexecutable.  The
opposite function, decompressing, is used to recover the
original files.
 
Encoding means shifting the bits of a file to a form less
likely to suffer errors when transmitted through a network.
For example, binary files are often encoded to hexadecimal
characters and transmitted as text files through BITNET.
They must be decoded when received.  The encoding guarantees
that no transmission errors occur at any point along the
transmission path from sender to recipient.
 
The remainder of this document will describe the most
popular archiving, compressing and encoding tools utilized
on DOS, Macintosh, and Unix platforms.
 
 
DOS Systems
 
Many archiving programs exist for DOS-based personal
computers.  These archiving programs typically also
incorporate compression.  The most popular of these are ARC,
ZIP, PAK and ZOO.  They are all distributed as SHAREWARE.
 
If you receive an archive file created by one of these
programs you will know because the file extension will
identify the kind of archiving performed on the file.  For
example, a file archived with ZIP archiving will have a
program extension of .ZIP.
 
In some cases (such as PCLIB archives) an archive will be
made "self-extracting".  In this case the archive will have
an .EXE extension and will include the extraction code.  To
extract the files from the archive you would simply
execute the archive file.
 
Each of these DOS archiving programs is available from the
PSUVM file server PCLIB under the CAC-PC library and the ARC
directory.
 
 
Macintosh Systems
 
The most popular archiving programs for the Macintosh also
incorporate compression.  These programs are Stuffit, Packit
and Compact Pro.  Archive files created with these programs
typically have .sit, .pit, or .cpt appended to the file
name.
 
The most popular encoder for the Macintosh is Binhex.
Programs encoded with Binhex typically have the string .hqx
appended to the file name.
 
SHAREWARE and commercial versions of each of these programs
are available from the PSUVM file server PCLIB under the
CAC-MAC library and the UTIL directory.
 
 
Unix Systems
 
The predominant archiver program on unix systems is tar.
Tar archives typically have the characters .tar appended to
the end of the file name.
 
DOS versions of tar are also available.  A tar extractor for
DOS systems is available from the PSUVM file server PCLIB
under the CAC-PC library and the ARC directory.  A PSUVM
version of tar is also available.  Issue the command HELP
TAR for more information on this version.
 
Compress is the predominate compression program for Unix
systems.  Its corollary, uncompress, exists as a separate
program on Unix systems.  Files compressed with compress
typically have the string .Z appended to the end of their
file name.
 
U-COMP and U-DCOMP are PSUVM versions of compress and
uncompress located on the PSUVM STAGE disk.  To access these
programs issue the commands:
 
 product stage  product share
 
You may then look at the help files for these programs with
HELP U-COMP and HELP U-DCOMP.
 
Uuencode and uudecode are the primary encoders used on Unix
systems.  Files encoded with uuencode typically have the
string .uue appended to the end of their file name.
 
The PSUVM program ARCUTIL contains options for uuencode and
uudecode.  ARCUTIL also supports xxencode, ARC and ZIP
formats.  Issue the command HELP ARCUTIL for more
information.
 
 
Important information for PSUVM users
 
Many Penn State faculty, staff and students use PSUVM to
retrieve programs and data files from BITNET LISTSERV and
Internet archive sites.  Occasionally programs are also
obtained from NETNEWS postings.
 
Programs obtained from BITNET LISTSERV archives or NETNEWS
postings are typically encoded.  If the program is intended
for a Macintosh it will typically be encoded with Binhex.
In this case you should strip any mailer headers and
trailers from the file, download it as text to your
Macintosh, and run Binhex to unencode it.
 
If the program is encoded with uuencode or xxencode your
best bet is to use ARCUTIL to decode the file before
transferring it to its destination system for extracting or
decompressing.
 
Programs and data files obtained from Internet archives are
often archived with tar and compressed with compress.  If
the program is destined for a Unix system your best strategy
is to transfer the file to that system (using a binary
transfer method) for decompression and extraction.
 
Textual data files obtained from Internet archives pose a
special problem for PSUVM users.  While U-DCOMP may be used
to decompress these files and TAR may be used to extract
them they still may be unreadable.  If this is the case it
probably means that the original file came from a computer
using the ASCII character set, and thus it is unreadable on
PSUVM (which uses the EBCDIC character set).
 
When this is the case, the following pipe command may be
used to convert the file from ASCII to EBCDIC and to align
the records correctly:
 
pipe < infile | deblock linend 0a | xlate a2e | > outfile
 
infile is the name of the file to be converted and outfile
is the intended name for the converted file.
 
 
Getting help
 
Most of the programs described in this handout come with
help files or have help summaries built in.  If you are
having trouble getting or using an archiver, compressor or
encoder you may contact the CAC Help Desk.
 
CAC Help Desk,12 Willard Building,(814) 863-1035
email: HELPDESK