| Re: Unknown #INFO [message #13431] |
Fri, 13 November 1998 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
From one of my Aussie friends:
Using v5.2 Beta on OpenVMS I get :-
** Structure !VALUES, 4 tags, length=24:
F_INFINITY FLOAT Inf
F_NAN FLOAT NaN
D_INFINITY DOUBLE Infinity
D_NAN DOUBLE NaN
Looks like a nice pattern shaping up here. :-)
Cheers,
David
----------------------------------------------------------
David Fanning, Ph.D.
Fanning Software Consulting
E-Mail: davidf@dfanning.com
Phone: 970-221-0438, Toll-Free Book Orders: 1-888-461-0155
Coyote's Guide to IDL Programming: http://www.dfanning.com/
|
|
|
|
| Re: Unknown #INFO [message #13433 is a reply to message #13431] |
Thu, 12 November 1998 00:00  |
mgs
Messages: 144
Registered: March 1995
|
Senior Member |
|
|
In article <72fd75$isf$1@hammer.msfc.nasa.gov>, mallors@msfc.nasa.gov
(Robert S. Mallozzi) wrote:
> In article <MPG.10b49db91db8b3ae98971b@news.frii.com>,
> davidf@dfanning.com (David Fanning) writes:
>>
>> Hang on there. Here is what I get in IDL 5.2 beta 2:
>>
>> IDL> Print, !Version
>> { x86 Win32 Windows 5.2 Beta2 Sep 14 1998}
>>
>> IDL> Help, !Values, /Structure
>> ** Structure !VALUES, 4 tags, length=24:
>> F_INFINITY FLOAT Inf
>> F_NAN FLOAT NaN
>> D_INFINITY DOUBLE Infinity
>> D_NAN DOUBLE NaN
>>
>> IDL> Print, !Values
>> { Inf NaN Infinity NaN}
>>
>> IDL> Print, Float( [ 'Inf', 'NaN', 'Infinity', 'NaN' ] )
>> Inf NaN Inf NaN
>>
>
> FYI,
>
> Identical results on
> { x86 linux unix 5.2 Beta2 Sep 14 1998}
And Identical results on
IDL> print,!version
{ PowerMac MacOS MacOS 5.2 Beta2 Sep 22 1998}
--
Mike Schienle Interactive Visuals, LLC
mgs@ivsoftware.com http://www.ivsoftware.com/
|
|
|
|
| Re: Unknown #INFO [message #13435 is a reply to message #13433] |
Thu, 12 November 1998 00:00 |
mallors
Messages: 76
Registered: November 1997
|
Member |
|
|
In article <MPG.10b49db91db8b3ae98971b@news.frii.com>,
davidf@dfanning.com (David Fanning) writes:
>
> Hang on there. Here is what I get in IDL 5.2 beta 2:
>
> IDL> Print, !Version
> { x86 Win32 Windows 5.2 Beta2 Sep 14 1998}
>
> IDL> Help, !Values, /Structure
> ** Structure !VALUES, 4 tags, length=24:
> F_INFINITY FLOAT Inf
> F_NAN FLOAT NaN
> D_INFINITY DOUBLE Infinity
> D_NAN DOUBLE NaN
>
> IDL> Print, !Values
> { Inf NaN Infinity NaN}
>
> IDL> Print, Float( [ 'Inf', 'NaN', 'Infinity', 'NaN' ] )
> Inf NaN Inf NaN
>
FYI,
Identical results on
{ x86 linux unix 5.2 Beta2 Sep 14 1998}
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~
Robert S. Mallozzi 256-544-0887
Mail Code ES 84
Work: http://www.batse.msfc.nasa.gov/ Marshall Space Flight Center
Play: http://cspar.uah.edu/~mallozzir/ Huntsville, AL 35812
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~
|
|
|
|
| Re: Unknown #INFO [message #13439 is a reply to message #13433] |
Thu, 12 November 1998 00:00 |
Martin Schultz
Messages: 515
Registered: August 1997
|
Senior Member |
|
|
Stein Vidar Hagfors Haugan wrote:
> In article <72d8sr$9at$1@agate.berkeley.edu>
> korpela@albert.ssl.berkeley.edu (Eric J. Korpela) writes:
>
> [...]
Oh yeah, I love it!
IDL> print,!version
{ mipseb IRIX unix 5.1.1 Jul 20 1998}
IDL> print,!values
{ inf nan0x2000000 inf nan0x70000000}
> IDL> print,float(['INF','Inf','NaN','nanq'])
IDL> print,float(['INF','Inf','NaN','nanq'])
% Type conversion error: Unable to convert given STRING to Float.
% Detected at: $MAIN$
% Type conversion error: Unable to convert given STRING to Float.
% Detected at: $MAIN$
% Type conversion error: Unable to convert given STRING to Float.
% Detected at: $MAIN$
% Type conversion error: Unable to convert given STRING to Float.
% Detected at: $MAIN$
0.00000 0.00000 0.00000 0.00000
"All [systems] are equal, but some are more equal than others."
(G.Orwell)
BTW: I would say that it is alright for IDL to pick up the sign in
-float('Inf'). That's what people write below integrals all the time ;-)
So at least in theory you could test
for -Inf and replace it by the smallest (largest negative) number that
your system represents. But, I realize that
symbolic algebra seems rather mathematica's realm, and an average IDLer
would rather only have to test for
value eq !values.Inf instead of abs(value) ...
So long,
Martin.
PS: Hope they fixed this inconsistency across all platforms in 5.2
------------------------------------------------------------ -------
Dr. Martin Schultz
Department for Engineering&Applied Sciences, Harvard University
109 Pierce Hall, 29 Oxford St., Cambridge, MA-02138, USA
phone: (617)-496-8318
fax : (617)-495-4551
e-mail: mgs@io.harvard.edu
Internet-homepage: http://www-as.harvard.edu/people/staff/mgs/
------------------------------------------------------------ -------
|
|
|
|
| Re: Unknown #INFO [message #13441 is a reply to message #13433] |
Thu, 12 November 1998 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Stein Vidar Hagfors Haugan (steinhh@ulrik.uio.no) writes:
>> David Fanning <davidf@dfanning.com> wrote:
>>>
>>> IDL> Help, !Values, /Structure
>>> ** Structure !VALUES, 4 tags, length=24:
>>> F_INFINITY FLOAT 1.#INF0
>>> F_NAN FLOAT 1.#QNAN
>>> D_INFINITY DOUBLE 1.#INF000
>>> D_NAN DOUBLE 1.#QNAN00
>>>
>> My sun gives this:
>> IDL> print,!version
>> { sparc sunos unix 5.0 Apr 28 1997}
>> IDL> print,!values
>> { Inf NaN Infinity NaN}
>> IDL> print,float('Inf')
>> Inf
>> IDL> print,float('1.#INF0')
>> 1.00000
>
> I guess you're right on the spot with the assumption that the
> compiler (or rather, the c library) makes the difference:
>
> IDL> print,!version
> { alpha OSF unix 5.1.1 Jul 20 1998}
> IDL> print,!values
> { INF NaNQ INF NaNQ}
> IDL> print,float(['INF','Inf','NaN','nanq'])
> % Type conversion error: Unable to convert given STRING to Float.
> % Detected at: $MAIN$
> INF INF 0.00000 NaNQ
Hang on there. Here is what I get in IDL 5.2 beta 2:
IDL> Print, !Version
{ x86 Win32 Windows 5.2 Beta2 Sep 14 1998}
IDL> Help, !Values, /Structure
** Structure !VALUES, 4 tags, length=24:
F_INFINITY FLOAT Inf
F_NAN FLOAT NaN
D_INFINITY DOUBLE Infinity
D_NAN DOUBLE NaN
IDL> Print, !Values
{ Inf NaN Infinity NaN}
IDL> Print, Float( [ 'Inf', 'NaN', 'Infinity', 'NaN' ] )
Inf NaN Inf NaN
Cheers,
David
----------------------------------------------------------
David Fanning, Ph.D.
Fanning Software Consulting
E-Mail: davidf@dfanning.com
Phone: 970-221-0438, Toll-Free Book Orders: 1-888-461-0155
Coyote's Guide to IDL Programming: http://www.dfanning.com/
|
|
|
|
| Re: Unknown #INFO [message #13443 is a reply to message #13433] |
Thu, 12 November 1998 00:00 |
steinhh
Messages: 260
Registered: June 1994
|
Senior Member |
|
|
In article <72d8sr$9at$1@agate.berkeley.edu>
korpela@albert.ssl.berkeley.edu (Eric J. Korpela) writes:
> In article <MPG.10b3c2392717b0898971a@news.frii.com>,
> David Fanning <davidf@dfanning.com> wrote:
>>
>> IDL> Help, !Values, /Structure
>> ** Structure !VALUES, 4 tags, length=24:
>> F_INFINITY FLOAT 1.#INF0
>> F_NAN FLOAT 1.#QNAN
>> D_INFINITY DOUBLE 1.#INF000
>> D_NAN DOUBLE 1.#QNAN00
>>
>> I don't know. This may be a Windows thing. Looks a little
>> Bill Gatesh, doesn't it. :-)
>
> Yes it does look windowish. It probably has something to do
> with the compiler they used to make their windows version. What
> concerns me is how this affects reading and writing of text files.
> In the past I have relied on IDL correctly parsing 'Inf' and 'NaN'
> in input files. You can see the obvious portability problem.
>
> I'm not sure if IEEE specifies the text representation of 'Inf'
> and 'NaN' or if a program is required to be able to correctly parse
> them on input.
>
> Out of curiousity, could you try the following?
>
> print,float('Inf')
> print,float('1.#INF0')
>
> My sun gives this:
> IDL> print,!version
> { sparc sunos unix 5.0 Apr 28 1997}
> IDL> print,!values
> { Inf NaN Infinity NaN}
> IDL> print,float('Inf')
> Inf
> IDL> print,float('1.#INF0')
> 1.00000
I guess you're right on the spot with the assumption that the
compiler (or rather, the c library) makes the difference:
IDL> print,!version
{ alpha OSF unix 5.1.1 Jul 20 1998}
IDL> print,!values
{ INF NaNQ INF NaNQ}
IDL> print,float(['INF','Inf','NaN','nanq'])
% Type conversion error: Unable to convert given STRING to Float.
% Detected at: $MAIN$
INF INF 0.00000 NaNQ
So apparently, case doesn't matter, but the particular *spelling*
convention does make a difference, and that's even between two
unix flavors! What a bummer.. So much for the days when ascii
files were portable.. :-)
One other caveat:
IDL> print,float(['-Inf','Inf']),-float('Inf')
INF INF
-INF
So the sign isn't picked up. But this again may be platform
specific.... then again, I never expected it to be possible
to do float('NaNQ') in the first place :-)
Regards,
Stein Vidar
|
|
|
|
| Re: Unknown #INFO [message #13446 is a reply to message #13433] |
Thu, 12 November 1998 00:00 |
korpela
Messages: 59
Registered: September 1993
|
Member |
|
|
In article <MPG.10b3c2392717b0898971a@news.frii.com>,
David Fanning <davidf@dfanning.com> wrote:
>
> IDL> Help, !Values, /Structure
> ** Structure !VALUES, 4 tags, length=24:
> F_INFINITY FLOAT 1.#INF0
> F_NAN FLOAT 1.#QNAN
> D_INFINITY DOUBLE 1.#INF000
> D_NAN DOUBLE 1.#QNAN00
>
> I don't know. This may be a Windows thing. Looks a little
> Bill Gatesh, doesn't it. :-)
Yes it does look windowish. It probably has something to do
with the compiler they used to make their windows version. What
concerns me is how this affects reading and writing of text files.
In the past I have relied on IDL correctly parsing 'Inf' and 'NaN'
in input files. You can see the obvious portability problem.
I'm not sure if IEEE specifies the text representation of 'Inf'
and 'NaN' or if a program is required to be able to correctly parse
them on input.
Out of curiousity, could you try the following?
print,float('Inf')
print,float('1.#INF0')
(I'm guessing that this will give the same result as read on the same strings.)
My sun gives this:
IDL> print,!version
{ sparc sunos unix 5.0 Apr 28 1997}
IDL> print,!values
{ Inf NaN Infinity NaN}
IDL> print,float('Inf')
Inf
IDL> print,float('1.#INF0')
1.00000
Thanks,
Eric
--
Eric Korpela | An object at rest can never be
korpela@ssl.berkeley.edu | stopped.
<a href="http://sag-www.ssl.berkeley.edu/~korpela">Click for home page.</a>
|
|
|
|
| Re: Unknown #INFO [message #13453 is a reply to message #13433] |
Wed, 11 November 1998 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Eric J. Korpela (korpela@islay.ssl.berkeley.edu) writes:
> David Fanning <davidf@dfanning.com> wrote:
>>
>> That is the IDL notation for "infinite value".
>>
>> IDL> Print, !Values.F_Infinity
>> 1.#INF0
>>
>
> Whatever happened to 'Inf' and '-Inf'? Is this new notation
> with the latest IDL version? If so, what's the new notation for
> 'NaN'?
IDL> Help, !Values, /Structure
** Structure !VALUES, 4 tags, length=24:
F_INFINITY FLOAT 1.#INF0
F_NAN FLOAT 1.#QNAN
D_INFINITY DOUBLE 1.#INF000
D_NAN DOUBLE 1.#QNAN00
IDL> Print, !Version
{ x86 Win32 Windows 5.1.1 Jul 20 1998}
I don't know. This may be a Windows thing. Looks a little
Bill Gatesh, doesn't it. :-)
Cheers,
David
----------------------------------------------------------
David Fanning, Ph.D.
Fanning Software Consulting
E-Mail: davidf@dfanning.com
Phone: 970-221-0438, Toll-Free Book Orders: 1-888-461-0155
Coyote's Guide to IDL Programming: http://www.dfanning.com/
|
|
|
|
|
|
| Re: Unknown #INFO [message #13456 is a reply to message #13453] |
Wed, 11 November 1998 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Luis Oliveira (eu_luis@hotmail.com) writes:
> I have received this output from PRINT, when I passed it an array:
>
> prf->tprfdata::get(w='velocitycomponents')
> 1.#INF0 1.#INF0 1.#INF0 1.#INF0 1.#INF0
>
> Please, if you have steped already in this "1.#INFO" or "-1#INFO" thing let
> me know,
That is the IDL notation for "infinite value".
IDL> Print, !Values.F_Infinity
1.#INF0
Cheers,
David
----------------------------------------------------------
David Fanning, Ph.D.
Fanning Software Consulting
E-Mail: davidf@dfanning.com
Phone: 970-221-0438, Toll-Free Book Orders: 1-888-461-0155
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Note: A copy of this article was e-mailed to the original poster.
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
