NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
Hi Jeff this is Alejandro from CATHALAC Panamá. I try to do what you recommend below but since am creating netcdf using grads all my netcdf file doesn't have the time dimension so when I write <?xml version="1.0" encoding="ISO-8859-1"?> <isl debug="true" offscreen="true"> <echo message="Loop iteration: ${loopindex} at: ${yyyy}-${MM}-${dd} ${HH}:${mm}:${ss}"/> <echo>PRUEBA</echo> <bundle clear="true" file="${islpath}/testfire.xidv" wait="true"> <setfiles datasource="/home/idv/DATANC/RepDom/acc24hr/"> <fileset dir="/home/idv/DATANC/RepDom/acc24hr/" pattern=".nc" last="5"/> </setfiles> </bundle> <image file="${saveDirect}/t.png"/> </isl> I get an aggregation error by time. (am going to fix this in the next couple of days). I start to try other solutions and I found this one: <?xml version="1.0" encoding="ISO-8859-1"?> <isl debug="true" offscreen="true"> <echo message="Loop iteration: ${loopindex} at: ${yyyy}-${MM}-${dd} ${HH}:${mm}:${ss}"/> <echo>PRUEBA</echo> <mkdir file="/home/idv/script/tmp"/> <exec command="find /home/idv/DATANC/RepDom/acc24hr/ -type f -mtime -31 -exec cp -f {} tmp/ \;"/> </isl> When I ran this like: [idv@cathalac1 isl]$ runIDV sourcefiles.isl Loop iteration: 0 at: 2011-04-13 15:27:40 PRUEBA Wed Apr 13 10:27:40 EST 2011: Calling exec:find /home/idv/DATANC/RepDom/acc24hr/ -type f -mtime -31 -exec cp -f {} tmp/ \; Exec: find /home/idv/DATANC/RepDom/acc24hr/ -type f -mtime -31 -exec cp -f {} tmp/ \; returned:1 the script create the tmp dir but the find command is not executed as I could do it manual look at the attachment. Thanks From: Jeff McWhirter [mailto:jeffmc@xxxxxxxxxxxxxxxx] Sent: martes, 12 de abril de 2011 11:05 p.m. To: Luis Del Castillo Cc: idvusers@xxxxxxxxxxxxxxxx Subject: Re: [idvusers] isl scripting Sorry, I hit tab/return in gmail and the mail got sent. There are undocumented fileset attributes sort, sort direction, first and last. So you can do something like: <fileset <http://www.unidata.ucar.edu/software/idv/docs/userguide/isl/FileTags.html#fileset> dir="the directory to look at" pattern="file pattern to match" sort="time" sortdir="ascending" /> Forgot to include the attributes: first="number of files in the beginning of the list to use" last="# files at the end of the list" Maybe Yuan and/or Julien can check this out to see if it works? (My main machine is baroque). And also update the docs? -Jeff
[idv@cathalac1 script]$ mkdir tmp [idv@cathalac1 script]$ ls datos isl jython tmp xidv [idv@cathalac1 script]$ find /home/idv/DATANC/RepDom/acc24hr/ -type f -mtime -31 -exec cp -f {} tmp/ \; [idv@cathalac1 script]$ cd tmp/ [idv@cathalac1 tmp]$ ls 2011.03.14.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.26.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.15.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.27.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.16.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.28.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.17.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.29.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.18.05.mm5.rdom.9km_accum24_pcp.nc 2011.04.07.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.19.05.mm5.rdom.9km_accum24_pcp.nc 2011.04.08.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.20.05.mm5.rdom.9km_accum24_pcp.nc 2011.04.10.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.21.05.mm5.rdom.9km_accum24_pcp.nc 2011.04.11.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.22.05.mm5.rdom.9km_accum24_pcp.nc 2011.04.12.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.23.05.mm5.rdom.9km_accum24_pcp.nc 2011.04.13.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.24.05.mm5.rdom.9km_accum24_pcp.nc 2011.04.14.05.mm5.rdom.9km_accum24_pcp.nc 2011.03.25.05.mm5.rdom.9km_accum24_pcp.nc [idv@cathalac1 tmp]$ ls
idvusers
archives: