On 9/27/10 7:48 AM, Dave Poreh wrote:
> On Sep 27, 4:16 am, Ben Tupper<ben.bigh...@gmail.com> wrote:
>> On 9/27/10 5:05 AM, Dave Poreh wrote:
>>
>>
>>
>>
>>
>>> Folks
>>> Hi;
>>> I read some data like this:
>>> -1.8750000000 78.8750000000 1.317
>>> -1.8750000000 78.6250000000 1.284
>>> -1.8750000000 78.3750000000 1.216
>>> -1.8750000000 78.1250000000 1.148
>>> -1.8750000000 77.8750000000 1.080
>>> ���������� ��
>>> And when I want to write it,
>>> openw,1,'C.dat'
>>> z=transpose(reform([data[0,*],data[1,*],data[2,*]],n_element s(data[2,*]),
>>> 3))
>>> printf,1,z
>>> close,1
>>> it gives me something like this:
>>> -1.87500 0.897000 71.6250
>>> 78.8750 9.87500 0.645000
>>> 1.31700 70.3750 21.8750
>>> -1.87500 1.01300 71.3750
>>> 78.6250 9.87500 0.538000
>>> 1.28400 70.1250 21.8750
>>> ���.
>>> Will you please help me out whit this?
>>> Cheers
>>> Dave
>>
>> Hi Dave,
>>
>> You haven't said what you expected to see, but clearly the transpose is
>> making this come out 'funny'. Here's what I see when I simply readf
>> from a file ("data-int.txt") and then printf the data to a file
>> ("data-out.txt").
>>
>> Cheers,
>> Ben
>> { x86_64 darwin unix Mac OS X 7.1 Apr 21 2009 64 64}
>>
>> ;data-in.txt looks like the following
>>
>> ;-1.8750000000 78.8750000000 1.317
>> ;-1.8750000000 78.6250000000 1.284
>> ;-1.8750000000 78.3750000000 1.216
>> ;-1.8750000000 78.1250000000 1.148
>> ;-1.8750000000 77.8750000000 1.080
>>
>> infile = "data-in.txt"
>> n = FILE_LINES(infile)
>> s = FLTARR(3, n)
>> OPENR, U, infile, /GET_LUN
>> READF, U, s
>> FREE_LUN, u
>>
>> outfile = "data-out.txt"
>> OPENW,U, outfile,/GET_LUN
>> PRINTF, U, s
>> FREE_LUN, U
>>
>> ; data-out.txt looks like the following
>>
>> ; -1.87500 78.8750 1.31700
>> ; -1.87500 78.6250 1.28400
>> ; -1.87500 78.3750 1.21600
>> ; -1.87500 78.1250 1.14800
>> ; -1.87500 77.8750 1.08000
>
> Thanks Ben. You r way is perfect. I don�t know why my way does not
> working?
> Cheers
> Dave
Hi again,
I know that when I get bogged down with this kind of stuff (happens
often*) using an tiny simple example helps - I like to use BINDGEN to
created an ordered easy-to-read array. Below, you can see that I
separate the reformed array (rs) from the transposed reformed array
(trs). See how reform inserts the data into the new array in the
original order? Note that the top row in rs is ordered 0,1,2,3,4. Does
the rs value printed look like what you expected? Then also note that
down the fist column the order is 0, 5, 10. When you transpose it back
to being a 3x5 array the top row becomes 0,5,10. Thus your woes begin.
IDL> s = bindgen(3,5)
IDL> rs = reform([s[0,*],s[1,*],s[2,*]],n_elements(s[2,*]),3)
IDL> trs = transpose(rs)
IDL> print, s
0 1 2
3 4 5
6 7 8
9 10 11
12 13 14
IDL> print, rs
0 1 2 3 4
5 6 7 8 9
10 11 12 13 14
IDL> print, trs
0 5 10
1 6 11
2 7 12
3 8 13
4 9 14
Cheers,
Ben
* Especially now that I use R a lot. It's either row- or column- major.
I can't remember, but I know it is whatever IDL isn't.
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
