[lug] Getting frustrated with gnuplot

Steve Sullivan sullivan at mathcom.com
Fri Oct 13 05:32:58 MDT 2006


Hi Daniel,
Here are some other plotting packagess you might try.
This only a partial list ... have you tried searching sourceforge?
Also have you tried searching on "scientific visualization"?

gmt (generic mapping tools): http://gmt.soest.hawaii.edu/
plplot: http://plplot.sourceforge.net/
openSceneGraph: http://66.220.18.234/
open visualization data explorer: http://www.research.ibm.com/dx/
gri: http://gri.sourceforge.net/
VisAD: http://www.ssec.wisc.edu/~billh/visad.html
Vis5D: http://www.ssec.wisc.edu/~billh/vis5d.html
gnuplot: http://www.gnuplot.info/  (well, you know this one!)
gnu plotutils: http://www.gnu.org/software/plotutils/plotutils.html
kmatplot: http://kmatplot.sourceforge.net/
grace: http://plasma-gate.weizmann.ac.il/Grace/

*** Can you post the result of your exploration back on BLUG
*** so the rest of us can benefit?

Thanks,

Steve

On Thu, Oct 12, 2006 at 12:00:07PM -0600, lug-request at lug.boulder.co.us wrote:
> Send LUG mailing list submissions to
> 	lug at lug.boulder.co.us
> 
> To subscribe or unsubscribe via the World Wide Web, visit
> 	http://lists.lug.boulder.co.us/mailman/listinfo/lug
> or, via email, send a message with subject or body 'help' to
> 	lug-request at lug.boulder.co.us
> 
> You can reach the person managing the list at
> 	lug-owner at lug.boulder.co.us
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of LUG digest..."
> 
> 
> Today's Topics:
> 
>    1. excluding directories with rsync (Paul Nowosielski)
>    2. Re: excluding directories with rsync (Dan Ferris)
>    3. Re: excluding directories with rsync (Zan Lynx)
>    4. Getting frustrated with gnuplot (Daniel Webb)
>    5. Re: Getting frustrated with gnuplot (Evelyn Mitchell)
>    6. Re: Getting frustrated with gnuplot (Daniel Webb)
>    7. BLUG Meeting Announcement - 2006-10-12 (Chris Riddoch)
>    8. Re: Getting frustrated with gnuplot (Todd Ruskell)
>    9. Re: Getting frustrated with gnuplot (Daniel Webb)
>   10. Re: excluding directories with rsync (Sean Reifschneider)
>   11. Re: Getting frustrated with gnuplot (Andrew GIlmore)
>   12. Re: Getting frustrated with gnuplot (Matt Thompson)
> 
> 
> ----------------------------------------------------------------------
> 
> Message: 1
> Date: Wed, 11 Oct 2006 15:43:57 -0600
> From: Paul Nowosielski <paul at celebrityaccess.com>
> Subject: [lug] excluding directories with rsync
> To: "Boulder (Colorado) Linux Users Group -- General Mailing List"
> 	<lug at lug.boulder.co.us>
> Message-ID: <200610111543.57245.paul at celebrityaccess.com>
> Content-Type: text/plain;  charset="us-ascii"
> 
> Dear All,
> 
> I'm trying to setup an rsync script to push data from a dev box to a live box.
> 
> The files live in different directories under /var/www
> 
> I need to exclude 5 of the directories and their content from being 
> propagated.
> 
> I've found a command to exclude files but not directories.
> 
> Does anyone know of a command or switch to do this?
> 
> Thank you,
> 
> -- 
> Paul Nowosielski
> Webmaster
> 
> 
> 
> ------------------------------
> 
> Message: 2
> Date: Wed, 11 Oct 2006 16:14:27 -0600
> From: Dan Ferris <dan at usrsbin.com>
> Subject: Re: [lug] excluding directories with rsync
> To: paul at celebrityaccess.com,	"Boulder (Colorado) Linux Users Group --
> 	General Mailing List"	<lug at lug.boulder.co.us>
> Message-ID: <452D6CC3.6020901 at usrsbin.com>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
> 
> I use --exclude-file= and then a file with the list of stuff to exclude.
> 
> It takes files or directories.
> 
> Dan
> 
> Paul Nowosielski wrote:
> > Dear All,
> >
> > I'm trying to setup an rsync script to push data from a dev box to a live box.
> >
> > The files live in different directories under /var/www
> >
> > I need to exclude 5 of the directories and their content from being 
> > propagated.
> >
> > I've found a command to exclude files but not directories.
> >
> > Does anyone know of a command or switch to do this?
> >
> > Thank you,
> >
> >   
> 
> -- 
> Build a man a fire and warm him for a night.  Set a man on fire and warm him for the rest of his life.
> 
> I'm not lazy.  I just hang out a lot.
> 
> 
> 
> ------------------------------
> 
> Message: 3
> Date: Wed, 11 Oct 2006 16:14:36 -0600
> From: Zan Lynx <zlynx at acm.org>
> Subject: Re: [lug] excluding directories with rsync
> To: paul at celebrityaccess.com,	"Boulder (Colorado) Linux Users Group --
> 	General Mailing List"	<lug at lug.boulder.co.us>
> Message-ID: <1160604876.16635.2.camel at localhost>
> Content-Type: text/plain; charset="us-ascii"
> 
> On Wed, 2006-10-11 at 15:43 -0600, Paul Nowosielski wrote:
> > Dear All,
> > 
> > I'm trying to setup an rsync script to push data from a dev box to a live box.
> > 
> > The files live in different directories under /var/www
> > 
> > I need to exclude 5 of the directories and their content from being 
> > propagated.
> > 
> > I've found a command to exclude files but not directories.
> > 
> > Does anyone know of a command or switch to do this?
> > 
> > Thank you,
> > 
> 
> --exclude works fine for both files and directories.
> For example, here is the command from a backup script I used to use:
> 
> ssh oberon rsync -az \
>         --password-file=/root/rsync-password \
>         --verbose \
>         --verbose \
>         --numeric-ids \
>         --delete --delete-excluded \
>         --exclude=/proc \
>         --exclude=/sys \
>         --exclude=/tmp \
>         --exclude=/var/tmp \
>         --exclude=/usr/tmp \
>         / snapshot at green::snapshots/${newdir#/home/snapshots} \
>         > /tmp/oberon.log 2>/tmp/oberon.errors </dev/null
> 
> -- 
> Zan Lynx <zlynx at acm.org>
> -------------- next part --------------
> A non-text attachment was scrubbed...
> Name: not available
> Type: application/pgp-signature
> Size: 189 bytes
> Desc: This is a digitally signed message part
> Url : http://lists.community.tummy.com/pipermail/lug/attachments/20061011/b1eed55c/attachment-0001.pgp
> 
> ------------------------------
> 
> Message: 4
> Date: Wed, 11 Oct 2006 20:58:57 -0600
> From: Daniel Webb <lists at danielwebb.us>
> Subject: [lug] Getting frustrated with gnuplot
> To: "Boulder (Colorado) Linux Users Group -- General Mailing List"
> 	<lug at lug.boulder.co.us>
> Message-ID: <20061012025857.GU3963 at drivebymail.com>
> Content-Type: text/plain; charset="us-ascii"
> 
> It stinks for publishing 3D data in black and white (like journals).  No 3D
> bar plots.  No ability to print the contour level on the contours (like a topo
> map).  How stupid is that to have to look over at a legend to figure out which
> of the 8 slightly different kinds of dotted lines correspond to which level?
> 
> Know of anything better for 3D that runs on Linux?
> 
> -------------- next part --------------
> A non-text attachment was scrubbed...
> Name: not available
> Type: application/pgp-signature
> Size: 189 bytes
> Desc: Digital signature
> Url : http://lists.community.tummy.com/pipermail/lug/attachments/20061011/878f6a4d/attachment-0001.pgp
> 
> ------------------------------
> 
> Message: 5
> Date: Wed, 11 Oct 2006 22:06:28 -0600
> From: Evelyn Mitchell <efm at tummy.com>
> Subject: Re: [lug] Getting frustrated with gnuplot
> To: "Boulder (Colorado) Linux Users Group -- General Mailing List"
> 	<lug at lug.boulder.co.us>
> Message-ID: <20061012040628.GA5216 at tummy.com>
> Content-Type: text/plain; charset=us-ascii
> 
> * On 2006-10-11 20:59 Daniel Webb <lists at danielwebb.us> wrote:
> > It stinks for publishing 3D data in black and white (like journals).  No 3D
> > bar plots.  No ability to print the contour level on the contours (like a topo
> > map).  How stupid is that to have to look over at a legend to figure out which
> > of the 8 slightly different kinds of dotted lines correspond to which level?
> > 
> > Know of anything better for 3D that runs on Linux?
> > 
> R does lovely plots:
>    http://www.r-project.org/
>    http://www.r-project.org/screenshots/screenshots.html
> 
> 3d bar charts, unless your data actually has 3 dimensions, are terribly
> misleading.
> 
> -- 
> Regards,                    tummy.com, ltd 
> Evelyn Mitchell             Linux Consulting since 1995
> efm at tummy.com               Senior System and Network Administrators
>                             http://www.tummy.com/
> 
> 
> ------------------------------
> 
> Message: 6
> Date: Wed, 11 Oct 2006 23:00:59 -0600
> From: Daniel Webb <lists at danielwebb.us>
> Subject: Re: [lug] Getting frustrated with gnuplot
> To: "Boulder (Colorado) Linux Users Group -- General Mailing List"
> 	<lug at lug.boulder.co.us>
> Message-ID: <20061012050059.GV3963 at drivebymail.com>
> Content-Type: text/plain; charset=us-ascii
> 
> On Wed, Oct 11, 2006 at 10:06:28PM -0600, Evelyn Mitchell wrote:
> 
> > R does lovely plots:
> >    http://www.r-project.org/
> >    http://www.r-project.org/screenshots/screenshots.html
> > 
> > 3d bar charts, unless your data actually has 3 dimensions, are terribly
> > misleading.
> 
> I looked briefly at that page even before I wrote the email, but didn't see
> that it can create contour plots like a topo map.
> 
> My data is from a parametric study with 5 values each of 2 independent
> parameters, for a total of 25 data.  Each combination of parameters results in
> a real-valued output and also a true/false output.  If you can think of a
> better way to represent that than a 3D bar plot with the height of the bars
> determined by the real-value output and the thatching of the bar determined by
> the true/false output, I would love to know because I'd love to use gnuplot
> and not have to learn something new.  Any non-discrete representation of this
> data will be somewhat misleading because it is a small enough amount of data
> that it shouldn't be considered continuous.
> 
> 
> 
> ------------------------------
> 
> Message: 7
> Date: Wed, 11 Oct 2006 22:35:28 -0600
> From: "Chris Riddoch" <riddochc at gmail.com>
> Subject: [lug] BLUG Meeting Announcement - 2006-10-12
> To: lug at lug.boulder.co.us
> Cc: announce at lug.boulder.co.us
> Message-ID:
> 	<6efbd9b70610112135pa74697ej7319a3d9c312399e at mail.gmail.com>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
> 
> Where:    Wolf Law, Room 307, on the University of Colorado campus
>           Approximately at Latitude 40.000852, Longitude -105.262402
>           A map is at: http://www.colorado.edu/directories/webmap/
>           Join us informally for a bite to eat at Noodles and Company
>           before the meeting, around 5:30 P.M.  Noodles is on Baseline,
>           just east of Highway 36/28th street.
> 
> When:     7:00 pm on Thursday October 12, 2006
> 
> Speaker:  Various
> 
> Talk:     Demo Day
> 
> Abstract: This is an opportunity for everyone to show off interesting tools and
>           hacks on Linux.
> 
> Driving Directions:
> 
>       * Go west on Baseline and turn right on Broadway.
>       * Turn right on Regent Drive.
>       * Take the first right that is not a parking lot. This is
>         before you get to the pedestrian crosswalk and the
>         Planetarium.
>       * Take a right into the first parking lot. Parking should be
>         free after 5:00.
>       * The law school building will be to your south, there is an
>         entrance on the north side.  Go down the hall, turn right,
>         and go to the main lobby.
>       * The room we're in is on the south side of the lobby.
> 
> -- 
> epistemological humility
>   Chris Riddoch
> 
> 
> ------------------------------
> 
> Message: 8
> Date: Wed, 11 Oct 2006 22:39:21 -0600
> From: Todd Ruskell <truskell at mines.edu>
> Subject: Re: [lug] Getting frustrated with gnuplot
> To: "Boulder (Colorado) Linux Users Group -- General Mailing List"
> 	<lug at lug.boulder.co.us>
> Message-ID: <200610112239.22071.truskell at mines.edu>
> Content-Type: text/plain;  charset="iso-8859-1"
> 
> Can't say I've played a lot with it, but you could try plotting all the 
> contours with the same line type, and then use the label command to label a 
> few of the contours so the reader can figure out the slope and size of the 
> contour steps.
> 
> Certainly not a perfect solution, but depending on your application, it might 
> not be too bad.  On the other hand, I agree that the sample contour plots 
> from R do at least look pretty.
> 
> You didn't necessarily specify "free" software, so I suppose I should mention 
> Matlab and Mathematica, if you happen to have access to them.  
> 
> Todd
> 
> On Wednesday 11 October 2006 08:58 pm, Daniel Webb wrote:
> > It stinks for publishing 3D data in black and white (like journals).  No 3D
> > bar plots.  No ability to print the contour level on the contours (like a
> > topo map).  How stupid is that to have to look over at a legend to figure
> > out which of the 8 slightly different kinds of dotted lines correspond to
> > which level?
> >
> > Know of anything better for 3D that runs on Linux?
> 
> -- 
> Dr. Todd G. Ruskell                         Phone: 303-384-2080
> Senior Lecturer, Department of Physics      FAX:   303-273-3919
> Colorado School of Mines
> Golden, CO 80401
> 
> 
> ------------------------------
> 
> Message: 9
> Date: Thu, 12 Oct 2006 00:02:29 -0600
> From: Daniel Webb <lists at danielwebb.us>
> Subject: Re: [lug] Getting frustrated with gnuplot
> To: "Boulder (Colorado) Linux Users Group -- General Mailing List"
> 	<lug at lug.boulder.co.us>
> Message-ID: <20061012060229.GW3963 at drivebymail.com>
> Content-Type: text/plain; charset=us-ascii
> 
> On Wed, Oct 11, 2006 at 10:39:21PM -0600, Todd Ruskell wrote:
> 
> > Can't say I've played a lot with it, but you could try plotting all the 
> > contours with the same line type, and then use the label command to label a 
> > few of the contours so the reader can figure out the slope and size of the 
> > contour steps.
> 
> I am considering doing that.  It's a pain though, because there are 8 of these
> plots and I have already changed the data several times.  I may label them as
> you suggest when I am sure they're not going to change, but it will be a pain
> to manually set all those labels if the data shifts.
>  
> > Certainly not a perfect solution, but depending on your application, it might 
> > not be too bad.  On the other hand, I agree that the sample contour plots 
> > from R do at least look pretty.
> 
> I can't evaluate if R can do either of the things I asked about (contours with
> levels or 3D bar).  For example here is one that can do the contours:
> 
> http://www.mps.mpg.de/dislin/exa_cont.html
> 
> It's not free as in speech, but then, neither is gnuplot and I've been using
> that.
>  
> > You didn't necessarily specify "free" software, so I suppose I should mention 
> > Matlab and Mathematica, if you happen to have access to them.  
> 
> I don't have either of those on my computers or I would probably just use
> Matlab.  Plus, once I leave school they're #@#$! expensive considering what
> little I would use them for personally.
> 
> 
> 
> ------------------------------
> 
> Message: 10
> Date: Thu, 12 Oct 2006 04:10:20 -0600
> From: Sean Reifschneider <jafo at tummy.com>
> Subject: Re: [lug] excluding directories with rsync
> To: paul at celebrityaccess.com,	"Boulder (Colorado) Linux Users Group --
> 	General Mailing List"	<lug at lug.boulder.co.us>
> Message-ID: <20061012101020.GA5158 at tummy.com>
> Content-Type: text/plain; charset=us-ascii
> 
> On Wed, Oct 11, 2006 at 03:43:57PM -0600, Paul Nowosielski wrote:
> >I need to exclude 5 of the directories and their content from being 
> >propagated.
> 
> As others have mentioned, rsync can do directories as well.  Read the man
> page, it has *EXTENSIVE* options for file selection that only reading the
> man page can really do justice to it.  The short form is that you can
> specify a directory by giving a trailing /.  So "sys/" matches any
> directory named "sys", where "sys" matches a file or directory.  And
> "/sys/" matches only the directory named "sys" at the root of the rsync
> file-system.
> 
> Thanks,
> Sean
> -- 
>  Brooks's Law of Prototypes: Plan to throw one away, you will anyhow.
> Sean Reifschneider, Member of Technical Staff <jafo at tummy.com>
> tummy.com, ltd. - Linux Consulting since 1995: Ask me about High Availability
>       Back off man. I'm a scientist.   http://HackingSociety.org/
> 
> 
> 
> ------------------------------
> 
> Message: 11
> Date: Thu, 12 Oct 2006 13:50:27 +0000
> From: Andrew GIlmore <agilmore at wirelessbeehive.com>
> Subject: Re: [lug] Getting frustrated with gnuplot
> To: "Boulder (Colorado) Linux Users Group -- General Mailing List"
> 	<lug at lug.boulder.co.us>
> Message-ID: <1160661027.20866.112.camel at lorax.uc.usbr.gov>
> Content-Type: text/plain; charset="us-ascii"
> 
> On Wed, 2006-10-11 at 23:00 -0600, Daniel Webb wrote:
> > On Wed, Oct 11, 2006 at 10:06:28PM -0600, Evelyn Mitchell wrote:
> > 
> > > R does lovely plots:
> > >    http://www.r-project.org/
> > >    http://www.r-project.org/screenshots/screenshots.html
> > > 
> > > 3d bar charts, unless your data actually has 3 dimensions, are terribly
> > > misleading.
> > 
> > I looked briefly at that page even before I wrote the email, but didn't see
> > that it can create contour plots like a topo map.
> 
> I've done this with R. I can't compare it with gnuplot, because I've had
> the same problem getting gnuplot to do contours. I know I've seen it
> suggested that gnuplot can, though.
> 
> I will not say that R is great for plots though. Extreme control over
> axes and text labels is a real pain.
> 
> The attached plot was created with the following R command:
> 
> contour(c(2008:2060),breaks[2:19],d,col=terrain.colors(20),
> zlim=c(0,40))
> 
> d is a sample of modeling results for Lake Powell elevation. :)
> 
> I would also say that the colors, etc. on this plot are not great, but
> it is an example.
> 
> Andrew
> -------------- next part --------------
> A non-text attachment was scrubbed...
> Name: R-contour.png
> Type: image/png
> Size: 10059 bytes
> Desc: not available
> Url : http://lists.community.tummy.com/pipermail/lug/attachments/20061012/f8d42095/R-contour-0001.png
> 
> ------------------------------
> 
> Message: 12
> Date: Thu, 12 Oct 2006 09:12:24 -0600
> From: Matt Thompson <thompsma at jilau1.colorado.edu>
> Subject: Re: [lug] Getting frustrated with gnuplot
> To: "Boulder (Colorado) Linux Users Group -- General Mailing List"
> 	<lug at lug.boulder.co.us>
> Message-ID: <452E5B58.5010908 at jilau1.colorado.edu>
> Content-Type: text/plain; charset=UTF-8; format=flowed
> 
> The heavens stretched forth and Daniel Webb decreed the following on or 
> about 10/11/2006 08:58 PM:
> > It stinks for publishing 3D data in black and white (like journals).  No 3D
> > bar plots.  No ability to print the contour level on the contours (like a topo
> > map).  How stupid is that to have to look over at a legend to figure out which
> > of the 8 slightly different kinds of dotted lines correspond to which level?
> 
> A dumb question, but did you try the label_contours.awk script that 
> Gnuplot has on it's script page:
> 
> http://gnuplot.sourceforge.net/scripts/index.html#tricks-here
> 
> It's not the greatest, but it does sort of label the contours.  Their 
> example is a wee bit busy though (the textcolor=-1 can be...psychedelic).
> 
> Matt
> -- 
> The mayfly lives only one day, and sometimes it rains. - Geo. Carlin
>     Matt Thompson -- http://ucsub.colorado.edu/~thompsma/
>     440 UCB, Boulder, CO  80309-0440
>     JILA A510, 303-492-4662
> 
> 
> ------------------------------
> 
> _______________________________________________
> LUG mailing list
> LUG at lug.boulder.co.us
> http://lists.lug.boulder.co.us/mailman/listinfo/lug
> 
> 
> End of LUG Digest, Vol 36, Issue 12
> ***********************************

-- 

========================================
Steve Sullivan    sullivan at mathcom.com

http://www.mathcom.com    303-494-7115
========================================



More information about the LUG mailing list