topic Re: Floating point numbers in batches possible? in Operating System - OpenVMS

Floating point numbers in batches possible?

matt burns — Thu, 21 Aug 2003 14:26:23 GMT

How can I store and manipulate floating point numbers from within a batch?

I'm well versed in batch logic and lexicals etcetera but mainly for file manipulation.

I know symbols can only hold integers and strings but I really need floating point numbers.

Can anyone help?

Re: Floating point numbers in batches possible?

Martin P.J. Zinser — Thu, 21 Aug 2003 16:00:38 GMT

Hello,

how about this

CFD002> a = "1.7"
CFD002> b = "2.1"
CFD002> icalc 'a' + 'b'
3.8
CFD002> sh sym icalc*
ICALC == "$PUBBIN:ICALC"
ICALC_OUT = " 3.8"

Icalc can be found at
http://zinser.no-ip.info:8080/www/vms/sw/icalc.htmlx

or
ftp://vms.process.com/ftp/vms-freeware/fileserv/

(the new home of the WKU archive, yes I know I need to update that page)

Re: Floating point numbers in batches possible?

matt burns — Tue, 26 Aug 2003 09:28:56 GMT

Cheers for the tip with ICALC, it was just the sort of thing I needed. Extra bonus for the trig functions cos I needed them too!

Re: Floating point numbers in batches possible?

matt burns — Fri, 29 Aug 2003 08:01:07 GMT

This is really another question but I thought I'd continue on this one...

The ICALC utility is great but, when not run interactively it still generates output.
I can't compile C code on the node I use and so cant change ICALC. I know I can divert SYS$OUTPUT into a text file, but is it possible to divert it into nothing like on unix?

If you're wandering what it's for, it to code a rotating wireframe cube made from asterixes on a VT100 terminal through telnet, I've attatched a basic working version if you're interested!

Re: Floating point numbers in batches possible?

Antoniov. — Fri, 29 Aug 2003 09:41:55 GMT

Hi Matt,
I downloaded your file and ICALC; I've compiled ICALC anad run CUBE. I can't understand exactly your problem, perhaps it's follow:
a) You can assingn output to null assign to device NLA0: (I suppose to change assign to *.TMP I've found so run quicly)
b) I've look for source and I've found ICALC can solve cos, sin and other trigonometric funcion.

HTH
Antoniov

Re: Floating point numbers in batches possible?

Martin P.J. Zinser — Fri, 29 Aug 2003 14:37:13 GMT

$ define/user sys$output nla0:
$ icalc 4.4 * 3.3
$ sh sym ica*
ICALC == "$PUBBIN:ICALC"
ICALC_OUT = " 14.52"

/user causes the define to be active for the next command invocation, so it is restored automatically after the icalc calculation

Greetings, Martin

Re: Floating point numbers in batches possible?

Hein van den Heuvel — Mon, 01 Sep 2003 02:49:28 GMT

matt> I've attatched a basic working version if you're interested!

That attachment was a zipped vms variable length record file with extention .com.
Which is perfectly fine by me and my vms systesm, but it upsets my billybox.
As I to read it on my pc, Notepad goes all funny.
The freeware 'textpad' that I use instead of notepad at least shows contents in its hex binary view mode.

To really read it I cobbled up a little perl script.
For your entertainment (nospan not supported :-)....

binmode STDIN;
while (read STDIN,$length_word,2) {
# avoid using "S" or "V". just do the math.
($length,$null) = unpack ("CC",$length_word);
$length += $null*256;
last if ($length > 32767);
$read = read STDIN,$line,$length;
# print "$length\t$line\n";
print "$line\n";
read STDIN,$null,1 if ($length & 1);
}

Re: Floating point numbers in batches possible?

matt burns — Tue, 02 Sep 2003 09:22:27 GMT

Cheers for the help guys, the nla0: tip was handy, and I've sped up quite a few of my batches with it :o)

To the chap with the perl script, you only had to ask and I would have sent it in windows format! nice script though!

Re: Floating point numbers in batches possible?

Jeroen Hartgers_2 — Mon, 29 Sep 2003 04:25:10 GMT

sometimes it is easier to do this
$b = 5
$c = 2
$a = (100*b)/c
$show symbol a
a = "250"
If you want to display the value of a the right way you simpley use f$extract to place the comma

$output = "''f$extract(0,1,a)',-
''f$extract(1,2,a)'"

show symbol output
output = "2,50"

Re: Floating point numbers in batches possible?

Willem Grooters — Mon, 29 Sep 2003 05:08:54 GMT

Just got this from the VMS technical Update: This feature will be - standard - be available in 7.3-2. Amongst others...