Due to the current gap in continued funding from the U.S. National Science Foundation (NSF), the NSF Unidata Program Center has temporarily paused most operations. See NSF Unidata Pause in Most Operations for details.

Re: How to save a ZIP file within a NetCDF file

  • To: Ed Hill <eh3@xxxxxxx>
  • Subject: Re: How to save a ZIP file within a NetCDF file
  • From: "Richard P Signell" <rsignell@xxxxxxxx>
  • Date: Tue, 23 Nov 2004 18:58:45 -0500
<FONT face="Default Sans Serif, Verdana, Arial, Helvetica, sans-serif" 
size=2><DIV>Ed,</DIV><DIV>&nbsp;</DIV><DIV>The code hash idea sounds useful for 
huge&nbsp;codes.&nbsp;&nbsp;&nbsp;</DIV><DIV>&nbsp;</DIV><DIV>But I still think 
the idea of stuffing the whole code into the</DIV><DIV>netcdf file will work 
for many folks and be the most desirable</DIV><DIV>route.&nbsp; I 
guess&nbsp;by&nbsp;your&nbsp;definition of "medium",&nbsp;the code I'm 
using</DIV><DIV>(ROMS) is "small", for though it is about 100,000 lines of 
FORTRAN code, </DIV><DIV>it compresses to less than 1 Mb!&nbsp;&nbsp;Since my 
netcdf output </DIV><DIV>files are&nbsp;typically&nbsp;300 -3000 Mb, I won't 
even notice the difference.</DIV><DIV>How big is your 
code?!</DIV><DIV>&nbsp;</DIV><DIV>For all the guys who are faithful to&nbsp;CVS 
and the like, this </DIV><DIV>code embedding stuff&nbsp;is certainly not 
necessary, but most of </DIV><DIV>the ocean modelers I know have 15 different 
unofficial versions </DIV><DIV>of any particular official version number, and 
they are&nbsp;are not keen </DIV><DIV>on giving any of them a finer grained 
version number for some</DIV><DIV>pretty good reasons (can't put a fine grained 
version number on</DIV><DIV>a 
band-aid)!</DIV><DIV>&nbsp;</DIV><DIV>-Rich</DIV><DIV><BR>Dr.&nbsp;Richard&nbsp;P.&nbsp;Signell&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;rsignell@xxxxxxxx<BR>U.S.&nbsp;Geological&nbsp;Survey&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;(508)&nbsp;457-2229&nbsp;<BR>384&nbsp;Woods&nbsp;Hole&nbsp;Road&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Fax:&nbsp;(508)&nbsp;457-2310<BR>Woods&nbsp;Hole,&nbsp;MA&nbsp;&nbsp;02543-1598&nbsp;&nbsp;&nbsp;<BR></DIV><DIV><DIV><BR></DIV><FONT
 color=#990099>-----owner-netcdfgroup@xxxxxxxxxxxxxxxx wrote: 
-----<BR><BR></FONT>To: Rich Signell &lt;rsignell@xxxxxxxx&gt;, 
netcdfgroup@xxxxxxxxxxxxxxxx<BR>From: Ed Hill &lt;eh3@xxxxxxx&gt;<BR>Sent by: 
owner-netcdfgroup@xxxxxxxxxxxxxxxx<BR>Date: 11/23/2004 05:41PM<BR>Subject: Re: 
How to save a ZIP file within a NetCDF file<BR><BR><font face="monospace" 
size=2>On&nbsp;Tue,&nbsp;2004-11-23&nbsp;at&nbsp;14:08&nbsp;-0500,&nbsp;Rich&nbsp;Signell&nbsp;wrote:<BR>&gt;&nbsp;Folks,<BR>&gt;&nbsp;<BR>&gt;&nbsp;Here's&nbsp;a&nbsp;problem&nbsp;we&nbsp;encounter&nbsp;all&nbsp;the&nbsp;time&nbsp;in&nbsp;the<BR>&gt;&nbsp;modeling&nbsp;business:&nbsp;&nbsp;we&nbsp;have&nbsp;a&nbsp;model&nbsp;run&nbsp;(netcdf&nbsp;output,<BR>&gt;&nbsp;of&nbsp;course),&nbsp;and&nbsp;we&nbsp;can't&nbsp;quite&nbsp;figure&nbsp;out&nbsp;what&nbsp;version&nbsp;of<BR>&gt;&nbsp;the&nbsp;code&nbsp;actually&nbsp;was&nbsp;used&nbsp;to&nbsp;produce&nbsp;it.&nbsp;&nbsp;Oh&nbsp;sure,<BR>&gt;&nbsp;it&nbsp;says&nbsp;"version&nbsp;2.1"&nbsp;but&nbsp;that&nbsp;was&nbsp;before&nbsp;we&nbsp;hacked<BR>&gt;&nbsp;the&nbsp;code&nbsp;to&nbsp;fix&nbsp;X,&nbsp;or&nbsp;put&nbsp;in&nbsp;that&nbsp;special&nbsp;little&nbsp;patch<BR>&gt;&nbsp;just&nbsp;for&nbsp;this&nbsp;run&nbsp;(or&nbsp;was&nbsp;it&nbsp;taken&nbsp;out)?&nbsp;&nbsp;Etc.<BR><BR><BR>Hi&nbsp;Rich,<BR><BR>It&nbsp;sounds&nbsp;like&nbsp;you&nbsp;could&nbsp;use&nbsp;a&nbsp;finer-grained&nbsp;version&nbsp;number.&nbsp;&nbsp;And/or&nbsp;a<BR>few&nbsp;more&nbsp;descriptive&nbsp;attributes&nbsp;that&nbsp;mention&nbsp;who&nbsp;is&nbsp;running&nbsp;the&nbsp;code&nbsp;and<BR>what&nbsp;they&nbsp;have&nbsp;or&nbsp;have&nbsp;not&nbsp;have&nbsp;done&nbsp;with&nbsp;it.<BR><BR>But&nbsp;getting&nbsp;back&nbsp;to&nbsp;your&nbsp;idea:&nbsp;for&nbsp;"medium"&nbsp;[1]&nbsp;and&nbsp;larger&nbsp;code&nbsp;bases,<BR>embedding&nbsp;the&nbsp;entire&nbsp;source&nbsp;code&nbsp;into&nbsp;the&nbsp;model&nbsp;output&nbsp;isn't&nbsp;necessarily<BR>a&nbsp;great&nbsp;idea.&nbsp;&nbsp;It&nbsp;doesn't&nbsp;scale.<BR><BR>Instead,&nbsp;have&nbsp;you&nbsp;considered&nbsp;embedding&nbsp;a&nbsp;hash&nbsp;of&nbsp;your&nbsp;model?&nbsp;&nbsp;For<BR>instance,&nbsp;during&nbsp;the&nbsp;model&nbsp;build&nbsp;stage&nbsp;you&nbsp;could&nbsp;easily&nbsp;create&nbsp;a&nbsp;list&nbsp;of<BR>the&nbsp;source&nbsp;files&nbsp;and&nbsp;their&nbsp;MD5SUMs&nbsp;and&nbsp;then&nbsp;embed&nbsp;that&nbsp;list&nbsp;into&nbsp;one&nbsp;or<BR>more&nbsp;output&nbsp;files&nbsp;(which&nbsp;may&nbsp;or&nbsp;may&nbsp;not&nbsp;be&nbsp;in&nbsp;a&nbsp;NetCDF&nbsp;format).<BR>Admittedly,&nbsp;it&nbsp;wouldn't&nbsp;tell&nbsp;you&nbsp;exactly&nbsp;what&nbsp;modifications&nbsp;were&nbsp;made&nbsp;to<BR>the&nbsp;code.&nbsp;&nbsp;But&nbsp;it&nbsp;would&nbsp;tell&nbsp;you&nbsp;which&nbsp;files&nbsp;had&nbsp;been&nbsp;modified&nbsp;and&nbsp;it<BR>would&nbsp;scale&nbsp;to&nbsp;work&nbsp;with&nbsp;much&nbsp;larger&nbsp;projects.&nbsp;&nbsp;And&nbsp;any&nbsp;hash(es)&nbsp;can&nbsp;be<BR>though&nbsp;of&nbsp;as&nbsp;a&nbsp;sort&nbsp;of&nbsp;"extended&nbsp;version&nbsp;number".<BR><BR>If&nbsp;you'd&nbsp;like,&nbsp;I'd&nbsp;be&nbsp;happy&nbsp;to&nbsp;work&nbsp;with&nbsp;you&nbsp;on&nbsp;writing&nbsp;up&nbsp;a&nbsp;quick&nbsp;pair<BR>of&nbsp;hash-list-embed&nbsp;and&nbsp;hash-list-extract&nbsp;utilities.<BR><BR>Ed<BR><BR>[1]&nbsp;lets&nbsp;define&nbsp;"medium"&nbsp;as&nbsp;"big&nbsp;enough&nbsp;to&nbsp;make&nbsp;the&nbsp;output<BR>&nbsp;&nbsp;&nbsp;&nbsp;file(s)&nbsp;unwieldy&nbsp;even&nbsp;when&nbsp;the&nbsp;source&nbsp;code&nbsp;has&nbsp;been<BR>&nbsp;&nbsp;&nbsp;&nbsp;compressed"<BR><BR>--&nbsp;<BR>Edward&nbsp;H.&nbsp;Hill&nbsp;III,&nbsp;PhD<BR>office:&nbsp;&nbsp;MIT&nbsp;Dept.&nbsp;of&nbsp;EAPS;&nbsp;&nbsp;Rm&nbsp;54-1424;&nbsp;&nbsp;77&nbsp;Massachusetts&nbsp;Ave.<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Cambridge,&nbsp;MA&nbsp;02139-4307<BR>emails:&nbsp;&nbsp;eh3@xxxxxxx&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;ed@xxxxxxx<BR>URLs:&nbsp;&nbsp;&nbsp;&nbsp;<A&NBSP;HREF="HTTP:
 designtimesp="7808" ? eh3 
web.mit.edu>http://web.mit.edu/eh3/</A>&nbsp;&nbsp;&nbsp;&nbsp;<A&NBSP;HREF="HTTP:
 designtimesp="7809" ? 
eh3.com>http://eh3.com/</A><BR>phone:&nbsp;&nbsp;&nbsp;617-253-0098<BR>fax:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;617-253-4464<BR><BR><BR><BR>***PRIVILEGED&nbsp;AND&nbsp;CONFIDENTIAL***&nbsp;<BR>The&nbsp;information&nbsp;contained&nbsp;in&nbsp;this&nbsp;e-mail&nbsp;message&nbsp;(including&nbsp;any&nbsp;attached&nbsp;files)&nbsp;is&nbsp;intended&nbsp;for&nbsp;the&nbsp;use&nbsp;of&nbsp;the&nbsp;addressee(s)&nbsp;only&nbsp;and&nbsp;is&nbsp;privileged&nbsp;information.&nbsp;The&nbsp;information&nbsp;should&nbsp;neither&nbsp;be&nbsp;posted&nbsp;to&nbsp;the&nbsp;Internet,&nbsp;nor&nbsp;published&nbsp;in&nbsp;any&nbsp;other&nbsp;public&nbsp;domain,&nbsp;without&nbsp;the&nbsp;express&nbsp;permission&nbsp;of&nbsp;the&nbsp;sender.&nbsp;If&nbsp;you&nbsp;are&nbsp;not&nbsp;the&nbsp;intended&nbsp;recipient(s)&nbsp;or&nbsp;the&nbsp;recipient's&nbsp;representative,&nbsp;you&nbsp;are&nbsp;hereby&nbsp;notified&nbsp;that&nbsp;any&nbsp;use,&nbsp;disclosure,&nbsp;copying&nbsp;or&nbsp;distribution&nbsp;of&nbsp;this&nbsp;communication&nbsp;is&nbsp;prohibited.&nbsp;If&nbsp;you&nbsp;have&nbsp;received&nbsp;this&nbsp;communication&nbsp;in&nbsp;error&nbsp;please&nbsp;notify&nbsp;us&nbsp;immediately&nbsp;at&nbsp;postmaster@xxxxxxxxxxxxxxxxx,&nbsp;and&nbsp;remove&nbsp;this&nbsp;message&nbsp;from&nbsp;your&nbsp;system.<BR><BR><BR></font></DIV></FONT>
  • 2004 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the netcdfgroup archives: