Re: reading files with date- and time - columns

[Craig Markwardt <craigmnet(at)cow.physics.wisc.edu>]

"Zeschke" <zeschke@bessy.de> writes:

> Hallo,
> 
> I have files for example in such a format:
> 
> File : s.log
> ;DATE          TIME       n x VALUES
> 23.01.2001 08:39:41 21 21.4 0
> 23.01.2001 08:40:16 21 21.3 0
> 23.01.2001 08:40:51 21 21.3 0
> 23.01.2001 08:41:26 21 21.4 0
> 
> or such a format:
> 
> ;DATE         TIME                n x VALUES
> 2/23/2001   08:39:41.643  34.660
> 2/23/2001   08:49:41.456  34.960

There are tons of ways to do this in IDL.  In principle the best way
to do this is read the data with a structure.  This is what my
TRANSREAD does internally.  Here is how I would do it, first
initializing the data, and then reading it.

mm = 0 & dd = 0 & yyyy = 0 & hh = 0 & mi = 0 & ss = 0 & nn = 0 & x = 0. & i = 0
openr, unit, 'myfile.dat'
transread, data, mm, dd, yyyy, hh, mi, ss, nn, x, i, comment=';', /debug, $
  format='(I2,1X,I2,1X,I4,1X,I2,1X,I2,1X,I2,1X,I0,1X,F0,1X,I0)'
;          mm  . dd  . yy    hh  : mi  : ss    nn    val   zero

I've never had much success using explicit quoted style format codes
for *reading* data, so I tend to use the "X" code instead for skipping
dead characters.  The use of the /DEBUG keyword is crucial for the
initial phases to be sure your format code is working.

After you get this, then there is the (trivial) matter of assembling
the component data into a single date, presumably with CALDATE, etc.

Craig

TRANSREAD found at http://cow.physics.wisc.edu/~craigm/idl/idl.html

-- 
--------------------------------------------------------------------------
Craig B. Markwardt, Ph.D.         EMAIL:    craigmnet@cow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
--------------------------------------------------------------------------