[gradsusr] Memory Allocation Error When Using PRINTIM

Thu Dec 2 11:55:31 EST 2010

Stephen

After NOAA finally let me download the shapefiles, I used sdfopen, loaded the shapefiles, and attached is the result from printim. There was no memory error.

NOW, having said that, the Cygwin build is like the pure Linux builds, and not like opengrads; so, the opengrads extensions do not work. Here is what I used.

set mproj nps
set mpdset mres
set map 15
set lat 10 85
set lon -200 -20
set mpdraw off
set grads off
set rgb 16 0 0 125
set gxout shaded
d tmp2m
set line 7
draw shp admin98
set line 16
draw recf 6.6 3.7 8.1 4.8
set rgb 17 175 175 255
set line 17
draw shp ln_ca
set line 11
draw shp ln_us
set line 11
*shp_polyf admin98 2342
draw shp ln_ak
set line 1
draw shp ln_ca
set line 1
draw shp ln_us
set line 1
*shp_lines admin98 2342
draw shp ln_ak
printim uscan.map.png x960 y720 white

John Huddleston, PhD
Cooperative Institute for Research in the Atmosphere

From: gradsusr-bounces at gradsusr.org [mailto:gradsusr-bounces at gradsusr.org] On Behalf Of Stephen McMillan
Sent: Thursday, December 02, 2010 9:10 AM
To: GrADS Users Forum
Subject: Re: [gradsusr] Memory Allocation Error When Using PRINTIM

John,

Thanks for offering to do this.  Since I don't have an FTP or HTTP site for placing the files, I am attaching only a somewhat modified version of my original script.  No control file used or needed.

Before running, I opened the latest GFS operational forecast, for example, http://nomads.ncep.noaa.gov:9090/dods/gfs_hd/gfs_hd20101202/gfs_hd_06z.  Script uses the "tmp2m" variable to set environment.  Near top of script are sources for the three "boundaries" shapefiles (tar.gz format) used.  I put extracted files in the SupportData folder where the "admin98" shapefile is.  In script, edit the "outfile" variable as needed.

Stephen Mc
On Thu, Dec 2, 2010 at 10:29 AM, Huddleston, John <Huddleston at cira.colostate.edu<mailto:Huddleston at cira.colostate.edu>> wrote:
Stephen

I build the Cygwin ftp://iges.org/grads/2.0/grads-2.0.a9-i686-pc-cygwin.tar version of GrADS and would be willing to test your data.

Can you zip it up include CTL ans GS files and put it on a FTP or HTTP site to download?

John

John Huddleston, PhD
Cooperative Institute for Research in the Atmosphere

From: gradsusr-bounces at gradsusr.org<mailto:gradsusr-bounces at gradsusr.org> [mailto:gradsusr-bounces at gradsusr.org<mailto:gradsusr-bounces at gradsusr.org>] On Behalf Of Stephen McMillan
Sent: Thursday, December 02, 2010 8:22 AM
To: GrADS Users Forum
Subject: Re: [gradsusr] Memory Allocation Error When Using PRINTIM

Arllindo, Jennifer,
Obviously the choice of shapefiles and output image dimensions play a critical part, not to mention the GrADS version you mentioned.  Still using 2.0.a7.oga.3, by replacing one shapefile for just the Alaska portion, I was able to "printim" an image without the memory allocation error.  In this case, I replaced the shapefile "ln_ak" with "admin98 2342" for AK.  However, if I tried to output a higher-resolution image (say, x1440 y1080), GrADS simply crashed (I briefly saw an error about something being corrupted).  I can do 960x720 or 1200x900 without a problem.

I have win32 superpack 2.0.a9.oga.1 on my other machine at home, so I'll try the original script on that this evening.  That could be interesting, since it's about a 10-years-old Windows XP laptop with much less RAM (500MB?).  For now, I'd rather not attempt the COLA version 2.0.a9, since I ran into problems trying to install it several weeks ago.  Perhaps someone else with a9 would like to give it a shot.

Thanks for your comments and suggestions!
Stephen
On Wed, Dec 1, 2010 at 8:07 PM, Arlindo da Silva <dasilva at alum.mit.edu<mailto:dasilva at alum.mit.edu>> wrote:

On Wed, Dec 1, 2010 at 7:59 PM, Jennifer Adams <jma at cola.iges.org<mailto:jma at cola.iges.org>> wrote:
Hi, Stephen --
The error message you're seeing occurs when the printim code tries to allocate an array that contains the vertices of a filled polygon. It may be that your shapefiles have single polygons with so many vertices (e.g. the one that fills in most of Canada) that your 3.2Gb of RAM is not enough, or maybe an integer isn't big enough to represent the number of elements in the array, so it passes malloc some junk and you get an error. To complicate things, you are also using a deprecated shapefile interface that I didn't write.

See if you can reliably reproduce this error with the latest COLA version of GrADS. If so, then send me the simplest possible script and the required shapefiles, and I will try to reproduce the error on my own system that has lots and lots of available memory.

I agree, the new shapefile interface is the way to go. The latest win32 superpack 2.0.a9.oga.1 has exactly the same shapefile code that Jennifer wrote, you could try that as well.

   Arlindo

--
Arlindo da Silva
dasilva at alum.mit.edu<mailto:dasilva at alum.mit.edu>

_______________________________________________
gradsusr mailing list
gradsusr at gradsusr.org<mailto:gradsusr at gradsusr.org>
http://gradsusr.org/mailman/listinfo/gradsusr

***************************************************

The information contained in this e-mail message

is intended only for the use of the recipient(s)

named above and may contain information that is

privileged, confidential, and/or proprietary.

If you are not the intended recipient, you may not

review, copy or distribute this message. If you have

received this communication in error, please notify

the sender immediately by e-mail, and delete the original message.

***************************************************

_______________________________________________
gradsusr mailing list
gradsusr at gradsusr.org<mailto:gradsusr at gradsusr.org>
http://gradsusr.org/mailman/listinfo/gradsusr

***************************************************

The information contained in this e-mail message

is intended only for the use of the recipient(s)

named above and may contain information that is

privileged, confidential, and/or proprietary.

If you are not the intended recipient, you may not

review, copy or distribute this message. If you have

received this communication in error, please notify

the sender immediately by e-mail, and delete the original message.

***************************************************

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gradsusr.org/pipermail/gradsusr/attachments/20101202/7f393d85/attachment-0003.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: uscan.map.png
Type: image/png
Size: 32375 bytes
Desc: uscan.map.png
Url : http://gradsusr.org/pipermail/gradsusr/attachments/20101202/7f393d85/attachment-0003.png 