CMSTAPE

Section: User Commands (1)
Updated: Columbia University
Index Return to Main Contents
 

NAME

cmstape - manipulate an IBM VM/SP CMS dump tape  

SYNOPSIS

cmstape [ -tvcrxfiud ] tape [ -Fnn ] file ...  

DESCRIPTION

Cmstape manipulates an IBM VM/SP CMS dump tape. Flags are:
-t
(default) type tape directory of tape file.
-v
verbose
-c
create a CMS tape dump.
-r
append to a CMS tape dump ('r' is non-mnemonic but matches tar).
-x
extract files from tape. If no names given all are extracted. otherwise only those named are extracted. Syntax for names is: fn.ft.
-f
tape is in readtape(1) format: each file block is preceded by a two byte block length with vax byte ordering. If ommitted, /dev/nrmt8 is used.
-i
image (no EBCDIC/ASCII translation or tacking on of newlines). Handled differently under UNIX and CMS: Under UNIX, V-format files are extracted with a halfword length header preceding each record. This makes image extracts and creates a reversible process for V as well as F format files under UNIX. Under CMS, V-format files are written to the CMS filesystem which maintains the record length information, so no length headers are needed in the data.
-u
don't convert filenames to lowercase and vice-versa. By default, lowercase UNIX names become uppercase CMS names and vice-versa.
-d
debug (multiple occurences increase volume of output).

Cmstape reads and writes either a tape device or a file that has been read off a tape by readtape(l). (That is, a file that contains a 2 byte vax order tape block size before each data block). This allows you to get the files off a tape with readtape(l) and transfer them over the network to the system where they will actually be extracted. You can also do the opposite: create a tape dump file, transfer it over the network, and write it to a tape device with writetape(l).

When writing a tape dump under UNIX, file names preceded by -Fnn cause those files to be written in fixed instead of the default variable format. If nn is specified, it will be used for the record length, otherwise 80 is assumed. Under CMS, this switch is ignored; the actual record length and format maintained by the filesystem are used.  

FILES

cms??????
temporary UNIX work file.
CMSTAPE.CMSUT1
temporary CMS work file.
 

EXAMPLES

Get directory of a CMS tape mounted on your favorite UNIX system:
    % cmstape -tv
    F/80         46 Apr  2 17:51 EXPCPU   ASSEMBLE A1
    V/70         43 Apr  2 18:07 AUTHORIZ EXEC     A1
    V/34         40 Jun 11 12:40 SENDCTLS EXEC     A1
    V/50         56 Apr 17 14:43 MKTOOLS  EXEC     A1
    V/76         95 Jun 11 12:39 CW       EXEC     A1

Extract all files from a CMS tape:

    % cmstape -x

Create a CMS tape dump on your UNIX system to give to your poor friend forced to use CMS:

    % cmstape -c /dev/nrmt8 cmstape.c -F foo.assemble

Read a CMS tape dump on your UNIX system, send it over the network to your CMS system and extract the files there:

    % readtape
    Reading from /dev/nrmt0
    Writing to file1    123456 bytes.
    Writing to file2         0 bytes.
    % telnet ibmhost
    ftp vaxhost
    ...
    ftp> type i
    ftp> get file1 file1.file
    ftp> quit
    cmstape -xfi file1    
 

SEE ALSO

readtape(1), writetape(1) by Jim Guyton, Rand Corp.  

AUTHOR

E. Alan Crosswell
Columbia University
 

BUGS

Due to the odd format of CMS tape files, you will need twice the working space as that taken up by the final files. Also, even files you don't want to extract get copied to a temporary file and are then discarded since the name of the file is at the end of the data!

Only manipulates files under CMS; use the CMS TAPE command for tape devices.

Does not know about CMS VMFPLC2 format.


 

Index

NAME
SYNOPSIS
DESCRIPTION
FILES
EXAMPLES
SEE ALSO
AUTHOR
BUGS

This document was created by man2html, using the manual pages.
Time: 06:21:54 GMT, December 12, 2024