comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » Re: open,/record - classification
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Switch to threaded view of this topic Create a new topic Submit Reply
Re: open,/record - classification [message #711] Wed, 25 November 1992 03:01
knipp is currently offline  knipp
Messages: 68
Registered: January 1993
Member
William Thompson (thompson@stars.gsfc.nasa.gov (William Thompson, code 682.1, x2040)
writes ...

> My experience is with IDL, not PV-WAVE, but this should still hold true. You
> can open a tape device in Unix and do I/O using READU and WRITEU. At least I
> was able to do this with an Exabyte tape drive on a Sun. You can also spawn
> out to the OS to send commands such as
>
> mt -f /dev/nrst1 fsf 1
>
> ...
>
> At one time I tried to write IDL procedures for Unix that imitated as closely
> as possible the VMS tape utilities: TAPRD, TAPWRT, etc. I stopped working on
> it, though, when I realized that there was no way to obtain the actual number
> of bytes read in, as one could with the VMS routines. One place that was
> crucial for us was in reading FITS tapes--we needed to be able to tell if the
> tape was written using 2880 byte or 28800 byte records.
>
> Bill Thompson

Dear Bill,
thank you for your answer.

I've tried OPENU together with READU/WRITEU. It worked perfectely well
for records up to ~ 1000 bytes long.
Unfortunately it did NOT with larger record-lengths (say 10000, 16000 or
even 19000 bytes long - which I have to deal with).

If anybody could help me to overcome this problem (without spawning
a command like 'dd' to the OS) I'll be very grateful.


Regarding your problem (2880 or 28800 byte records) :
Have you tried the UNIX-command 'tcopy' ?

K.Knipp
(Does nobody know something about classification in WAVE ?)


Karlheinz Knipp Tel. 0511 - 762 4922
Universitaet Hannover Fax. 0511 - 762 2483
Institut fuer Photogrammetrie Mail. knipp@ipi.uni-hannover.de
Nienburger Str.1
3000 Hannover 1
FRG
Re: open,/record - classification [message #717 is a reply to message #711] Mon, 16 November 1992 15:22 Go to previous message
thompson is currently offline  thompson
Messages: 584
Registered: August 1991
Senior Member
In article <1992Nov16.152127.17585@newsserver.rrzn.uni-hannover.de>, knipp@ipi.uni-hannover.de writes...
> Hello there,
>
> We're runnig Pv~Wave Cl. 4.0 under SunOs 4.1.2
>
> There are 2 questions :
>
> a) Has anybody ever tried to open a device (p.e. a streamer-tape)
> for record-wise reading/writing ?. It should work under VMS, but we must do it
> under UNIX.
>
> b) Is there anybody who has programmed a land-use-classification in
> WAVE (or knows about an existing one) ?
>
> Thanks, KK

My experience is with IDL, not PV-WAVE, but this should still hold true. You
can open a tape device in Unix and do I/O using READU and WRITEU. At least I
was able to do this with an Exabyte tape drive on a Sun. You can also spawn
out to the OS to send commands such as

mt -f /dev/nrst1 fsf 1

or

mt -f /dev/nrst1 rewind

As I recalled, I just opened the device with the command

OPENU, UNIT, '/dev/nrst1'

I didn't use any keywords.

At one time I tried to write IDL procedures for Unix that imitated as closely
as possible the VMS tape utilities: TAPRD, TAPWRT, etc. I stopped working on
it, though, when I realized that there was no way to obtain the actual number
of bytes read in, as one could with the VMS routines. One place that was
crucial for us was in reading FITS tapes--we needed to be able to tell if the
tape was written using 2880 byte or 28800 byte records.

Bill Thompson
  Switch to threaded view of this topic Create a new topic Submit Reply
Previous Topic: Re: Floating point system variables on RS6000
Next Topic: Re: Scatter Plots

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Wed Oct 08 13:31:21 PDT 2025

Total time taken to generate the page: 0.00833 seconds