| question about readf [message #49570] |
Thu, 03 August 2006 00:44  |
rlayberry
Messages: 33
Registered: November 2004
|
Member |
|
|
I am trying to get data out of MS-Excel and into IDL as a float array.
I laboriously save each Excel worksheet as a csv.
I then readf the CSV into a float array.
The problem is that there are some missing data values. for example
,10.0,,,0.0,10
so a missing data value is given by two commas next to each other.
readf assumes no data is there and all my formatting goes to pot. is
there anyway of getting either excel to write csvs out with dummy
values (-99.9 for example) or of getting IDL to read in a missing data
point as a dummy value?
thanks in advance
russ
|
|
|
|
| Re: question about readf [message #49623 is a reply to message #49570] |
Fri, 04 August 2006 07:23  |
savoie
Messages: 68
Registered: September 1996
|
Member |
|
|
rlayberry@hotmail.com writes:
> I am trying to get data out of MS-Excel and into IDL as a float array.
>
> I laboriously save each Excel worksheet as a csv.
>
> I then readf the CSV into a float array.
>
> The problem is that there are some missing data values. for example
> ,10.0,,,0.0,10
> so a missing data value is given by two commas next to each other.
> readf assumes no data is there and all my formatting goes to pot. is
> there anyway of getting either excel to write csvs out with dummy
> values (-99.9 for example) or of getting IDL to read in a missing data
> point as a dummy value?
I do this by reading in the entire line first, and then using stregex.
lformat = '(%"%100s")' ; a really long string to read in each line
readf, lun, FORMAT = lformat, line
delimiter = ','
parts = strsplit( line, delimiter, $
count = count, /EXTRACT, /PRESERVE_NULL )
parts = strcompress( parts, /REMOVE_ALL )
nullidx = where( parts eq '', count )
if count gt 0 then begin
parts[ nullidx ] = !values.F_NAN
endif
Just remember the keyword to strsplit: /PRESERVE_NULL.
Hope this helps. If not, I hope someone has a better answer.
Matt
--
Matthew Savoie - Scientific Programmer
National Snow and Ice Data Center
(303) 735-0785 http://nsidc.org
|
|
|
|
| Re: question about readf [message #49701 is a reply to message #49623] |
Fri, 11 August 2006 11:08 |
Paul[2]
Messages: 13
Registered: January 2006
|
Junior Member |
|
|
savoie@nsidc.org wrote:
> rlayberry@hotmail.com writes:
>
>> I am trying to get data out of MS-Excel and into IDL as a float array.
>>
>> I laboriously save each Excel worksheet as a csv.
>>
>> I then readf the CSV into a float array.
>>
>> The problem is that there are some missing data values. for example
>> ,10.0,,,0.0,10
>> so a missing data value is given by two commas next to each other.
>> readf assumes no data is there and all my formatting goes to pot. is
>> there anyway of getting either excel to write csvs out with dummy
>> values (-99.9 for example) or of getting IDL to read in a missing data
>> point as a dummy value?
>
> I do this by reading in the entire line first, and then using stregex.
>
>
> lformat = '(%"%100s")' ; a really long string to read in each line
> readf, lun, FORMAT = lformat, line
> delimiter = ','
> parts = strsplit( line, delimiter, $
> count = count, /EXTRACT, /PRESERVE_NULL )
> parts = strcompress( parts, /REMOVE_ALL )
> nullidx = where( parts eq '', count )
> if count gt 0 then begin
> parts[ nullidx ] = !values.F_NAN
> endif
>
>
> Just remember the keyword to strsplit: /PRESERVE_NULL.
>
> Hope this helps. If not, I hope someone has a better answer.
>
> Matt
>
> --
> Matthew Savoie - Scientific Programmer
> National Snow and Ice Data Center
> (303) 735-0785 http://nsidc.org
I'm now reading .xls files natively in IDL using IDL's COM import
bridge and the Excel Application Object. I *may* be able to preview it
if anyone wants to "alpha" test it. I can't put the code out yet
because we had to enhance one of IDL's internal dlls to make this
possible. Anyone interested in testing it? Is there a project anyone
is working on that could benefit from such functionality?? Contact me
directly for more info: psommer@ittvis.com
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
