Re: [ferret_users] read sequence of data files

[Ansley Manke <ansley.b.manke@xxxxxxxx>]

Hi all,
To build on what Ryo suggests, one can to do this kind of thing with the 
REPEAT/RANGE= /NAME=  syntax

 my_list = {"mydata001.nc","mydata002.nc","mydata003.nc","mydata004.nc"}
 let r1 = 1
 let r2 = 3
 repeat/range=`r1`:`r2`/name=m (let fname = my_list[i=`m`]; go process 
`fname`)

This would allow you to have a master list of files, and then the script 
could choose among the files, doing some subset of them if desired.

Ansley

Ryo Furue wrote:

> Hello Ilana,
>
> | A student is  writing a script to work with a big sequence of data
> | files. [. . .] I was wondering if there's any function that allows
> | ferret to read and select the specific files (batch mode?),
>
> The following solution isn't quite automatic, but it's manageable
> unless you have hundreds of files:
>
>   !--- main script file ---
>   go process  mydata001.nc
>   go process  mydata002.nc
>   go process  mydata003.nc
>   ! . . . and so on . . . .
>
>   !--- another script file named "process.jnl" ---
>   define symbol datafile "($1)"      !! get the argument
>   set data ($datafile)               !! open the file
>   ! . . . do something on the data . . .
>   cancel data/all                    !! close the file
>
> The idea is that you create a "subroutine" to process your data
> and you pass the filename to it as an argument.
>
> If I needed process hundreds of data files systematically,
> I would generate a ferret script corresponding to the
> main script above, by using a scripting language like Ruby,
> Perl, and shell.
>
> Ryo
>