[comp.unix.questions] Re: Reading IBM Tape

Paul O'Neill pvo at sapphire.OCE.ORST.EDU
Thu May 31 13:54:26 AEST 1990


Archive-name: readEBCDICansi/31-May-90
Original-posting-by: pvo at sapphire.OCE.ORST.EDU (Paul O'Neill)
Original-subject: Re: Reading IBM Tape
Reposted-by: emv at math.lsa.umich.edu (Edward Vielmetti)

[Reposted from comp.unix.questions.
Comments on this service to emv at math.lsa.umich.edu (Edward Vielmetti).]

In article <23453 at adm.BRL.MIL> zellich at stl-07sima.army.mil ( Rich Zellich) writes:
>If the problem is reading a standard-label EBCDIC tape, then .......
>
You might try this:

(tcopy the tape first and adjust the ibs block size accordingly)

#!/usr/local/bin/perl
#
#       readEBCDICansi
#
#       6 apr 90
#       Paul V. O'Neill
#	Coastal Imaging Lab
#	Oregon State University
#
#       read an ansi-labeled tape coded in EBCDIC!


while(1) {
    open(HEADER, 'dd if=/dev/nrmt0 ibs=80 conv=ascii cbs=80|');
    $name = '';                                 # undef for die test
    while(<HEADER>) {                           # read the ansi header w/ dd
        print $_;                               # verbosely
        $name = $1 if /^HDR1(\w*)/;     # found file name (ignores ".ext")
    }
    close(HEADER);
    print "$name\n\n";                          # more verbosity
    die "$0: Looks like EOT $! $?\n" if $?;     # dd return error
    $name || die "$0: No name found $! $?\n";   # dd ok, but no name!

    open(OUT, ">$name");                        # open the output disk file
    open(DATA, 'dd if=/dev/nrmt0 ibs=3600 conv=ascii|');
    while(<DATA>) {                             # read tape file w/ dd
        s/\015//;                               # strip out infidel CR's
        print OUT $_;                           # and write to disk file
    }
    close OUT;
    close DATA;
    system 'mt -f /dev/nrmt0 fsf 1';            # skip the trailer tape file
}


Paul O'Neill                 pvo at oce.orst.edu		DoD 000006
Coastal Imaging Lab
OSU--Oceanography
Corvallis, OR  97331         503-737-3251



More information about the Alt.sources mailing list