| Re: the precision of PRINT and HELP statements [message #26928] |
Thu, 04 October 2001 10:24 |
Paul van Delst
Messages: 364
Registered: March 1997
|
Senior Member |
|
|
George McCabe wrote:
>
> hello,
>
> this one made me laugh when i realized what was going on -
>
> IDL> help, chkN
> CHKN LONG = 140486
> IDL> help, float(Nhb)/Nreq0
> <Expression> FLOAT = 140486.
> IDL> print, chkN - float(Nhb)/Nreq0
> -0.500000
>
> perhaps one should expect IDL's PRINT statement to act this way,
>
> IDL> print, float(Nhb)/Nreq0
> 140486.
> IDL> print, format='(g)', float(Nhb)/Nreq0
> 140486.5
>
> but, the HELP command i feel should show the actual value to the
> precision of the variable type.
Hmm. My personal opinion is that if you need to know the actual value of a variable to its
precision (e.g. X number of significant digits, or 6 decimal places, or something else) then it
is encumbent upon the programmer to tell exactly what he/she wants printed/displayed. Relying
on default behaviour is means you don't care what the output format is. Why use HELP to print
out number values when you know a PRINT, FORMAT=..etc.. will give you what you want? And the
online help states for the expression arguments to HELP:
HELP
Expression(s)
The arguments are interpreted differently depending on the keyword selected. If no keyword is
selected, HELP displays basic information for its parameters. For example, to see the type and
structure of the variable A, enter:
HELP, A
The important phrase being "basic information."
I actually find the truncation that occurs in cases like the one that you mention helpful as it
has taught me to be careful when printing out numbers for inspection. The same applies in my
Fortran code too.
paulv
--
Paul van Delst Religious and cultural
CIMSS @ NOAA/NCEP purity is a fundamentalist
Ph: (301)763-8000 x7274 fantasy
Fax:(301)763-8545 V.S.Naipaul
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
