matplotlib-users Mailing List for matplotlib

Hi all.

I have 2 issues with the contour routine.  I'm just giving it a 2D  
array.  The first issue which is not a show stopper, is that, it uses  
j for the x position and i for the y-position (where my data is given  
by: M[i][j]).  This seems counter intuitive.  Can I switch the  
convention without transposing all the data?

The other issue seems more difficult.  It seems to be ignoring any  
"extent=" options I give it.  Is this because I just supplied a  
single 2D array rather than also supplying spatial x,y values too?

Thanks,
Craig

I worked out the slicing

Pxx, freqs, bins = matplotlib.mlab.specgram(x, NFFT, Fs, detrend,
             window, noverlap) #line 3377

x1,y1=Pxx.shape
q=zeros([x1,y1],'d')
q[:][x1/2:x1]=Pxx[:][0:x1/2]
q[:][0:x1/2]=Pxx[:][x1/2:x1]
Pxx=q

This would be the small change needed. To me it looks like the change
could be done in axes.py

One could add an argument wrapparound=False
and then 
if wrapparound 
	new code

I assume that one should do something about the y-axis as well.
extent = xmin, xmax, -amax(b)/2, amax(b)/2, is probably not 100% OK but
will do it for me... 


I guess I will have a look at psd now. I guess that will be easier

/Johan
	

On Wed, 2006-02-08 at 22:26 +0100, Johan Nilsson wrote:
> Dear group
> 
> I am rather new to matplotlib. I have a reasonably large file of complex
> valued data. Today I tried out specgram and psd, and they both work
> nicely. However I would like the negative frequencies to be on the
> negative side of the spectrum. I tried to look around, but could not
> find any parameter to set this. My question is, if this is easily
> possible or do I need to do something like
> 
> a,b,c,d=pylab.specgram(y,1024)
> pylab.show()
> 
> x1,y1=a.shape
> q=zeros([x1,y1],'d')
> q[:,x1/2:x1-1]=a[:,0:x1/2-1]
> q[:,0:x1/2-1]=a[:,x1-1:x1/2:-1]
> 
> pylab.imshow(q)
> 
> Apart from that I don't get the slicing to work properly yet, it would
> be nicer if there was just a parameter to set for specgram
> 
> 
> 
> /Johan
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

*doh*

There's not supposed to be a "dot" in the rc filename...

Works again.

--Craig

On Feb 9, 2006, at 11:44 AM, Craig Maloney wrote:

> Hi Jeff, et.al.
>
> Funny thing... somehow during the process of installation and  
> moving stuff around, matplotlib has gone back to using numpy  
> instead of Numeric (and giving me the error messages when trying to  
> run contour_demo.py).
>
> I still have
> "numerix      : Numeric   # Numeric or numarray"
> in my .rc file.
>
> How can I check to make sure about which .rc file matplotlib is  
> trying to use?
>
> Thanks, Craig
>
> -------------------------------
> Python 2.3.3 (#2, Jan  9 2004, 13:37:22)
> [GCC 3.0.1] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
> >>> from pylab import *
> >>>
> >>> rcParams['tick.direction'] = 'out'
> >>>
> >>> delta = 0.025
> >>> x = arange(-3.0, 3.0, delta)
> >>> y = arange(-2.0, 2.0, delta)
> >>> X, Y = meshgrid(x, y)
> >>> Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0)
> >>> Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1, 1)
> >>> # difference of Gaussians
> ... Z = 10.0 * (Z2 - Z1)
> >>>
> ...
> >>> type(X)
> <class 'numarray.numarraycore.NumArray'>
> ------------------------------------------------------
>
> PS  apparently, numpy *is* installed somewhere, I guess....
>
> On Feb 9, 2006, at 10:12 AM, Jeff Whitaker wrote:
>
>
>> Craig Maloney wrote:
>>
>>
>>> Thanks, Jeff.
>>> That did the trick.  I didn't even have a .matplotlibrc file.    
>>> Default behavior must be numpy, huh?
>>> --Craig
>>>
>>>
>>
>> Craig:
>>
>> If you have numpy installed, it will choose that as the default.   
>> If numpy is not installed, Numeric is the default.
>>
>> -Jeff
>>
>> -- 
>> Jeffrey S. Whitaker         Phone  : (303)497-6313
>> Meteorologist               FAX    : (303)497-6449
>> NOAA/OAR/PSD  R/PSD1        Email  : Jef...@no...
>> 325 Broadway                Office : Skaggs Research Cntr 1D-124
>> Boulder, CO, USA 80303-3328 Web    : http://tinyurl.com/5telg
>>
>>
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through  
> log files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD  
> SPLUNK!
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=103432&bid=230486&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

>>>>> "Craig" == Craig Maloney <cma...@ki...> writes:

    Craig> Hi Jeff, et.al.  Funny thing... somehow during the process
    Craig> of installation and moving stuff around, matplotlib has
    Craig> gone back to using numpy instead of Numeric (and giving me
    Craig> the error messages when trying to run contour_demo.py).

    Craig> I still have "numerix : Numeric # Numeric or numarray" in
    Craig> my .rc file.

    Craig> How can I check to make sure about which .rc file
    Craig> matplotlib is trying to use?
run your script with 

> python myscript.py --verbose-helpful

JDH

Hi Jeff, et.al.

Funny thing... somehow during the process of installation and moving  
stuff around, matplotlib has gone back to using numpy instead of  
Numeric (and giving me the error messages when trying to run  
contour_demo.py).

I still have
"numerix      : Numeric   # Numeric or numarray"
in my .rc file.

How can I check to make sure about which .rc file matplotlib is  
trying to use?

Thanks, Craig

-------------------------------
Python 2.3.3 (#2, Jan  9 2004, 13:37:22)
[GCC 3.0.1] on linux2
Type "help", "copyright", "credits" or "license" for more information.
 >>> from pylab import *
 >>>
 >>> rcParams['tick.direction'] = 'out'
 >>>
 >>> delta = 0.025
 >>> x = arange(-3.0, 3.0, delta)
 >>> y = arange(-2.0, 2.0, delta)
 >>> X, Y = meshgrid(x, y)
 >>> Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0)
 >>> Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1, 1)
 >>> # difference of Gaussians
... Z = 10.0 * (Z2 - Z1)
 >>>
...
 >>> type(X)
<class 'numarray.numarraycore.NumArray'>
------------------------------------------------------

PS  apparently, numpy *is* installed somewhere, I guess....

On Feb 9, 2006, at 10:12 AM, Jeff Whitaker wrote:

> Craig Maloney wrote:
>
>> Thanks, Jeff.
>> That did the trick.  I didn't even have a .matplotlibrc file.    
>> Default behavior must be numpy, huh?
>> --Craig
>>
>
> Craig:
>
> If you have numpy installed, it will choose that as the default.   
> If numpy is not installed, Numeric is the default.
>
> -Jeff
>
> -- 
> Jeffrey S. Whitaker         Phone  : (303)497-6313
> Meteorologist               FAX    : (303)497-6449
> NOAA/OAR/PSD  R/PSD1        Email  : Jef...@no...
> 325 Broadway                Office : Skaggs Research Cntr 1D-124
> Boulder, CO, USA 80303-3328 Web    : http://tinyurl.com/5telg
>

On Feb 9, 2006, at 12:08 PM, Darren Dale wrote:

> I don't know how to proceed on this. Do I check the platform and  
> apply the
> hack for darwin? Would that break usetex for any other mac users  
> out there?

This is a good question..

My suspicion is that this is a problem specific to the dvipng in the  
I-Installer teTeX distribution.  We could simply suggest to mac users  
to upgrade to 1.7, but dvipng is actually a pain to compile.  I tried  
to update, but the configure script was never able to find my  
freetype (I only have freetype2 installed -- does dvipng require  
freetype 1?), and I gave up.  It also wants other graphic libraries,  
which makes me think that this is something that people will hate  
more than the Ghostscript requirement.

My only suggestion is to poll Mac users about dvipng versions and  
problems (or lack of problems).  Perhaps a short term solution would  
be to create an rc option that users could toggle if they were having  
problems.  This would at least allow us to investigate the problem  
further, without too much hacking on the part of the user.

-Rob.

-----
Rob Hetland, Assistant Professor
Dept of Oceanography, Texas A&M University
p: 979-458-0096, f: 979-845-6331
e: he...@ta..., w: http://pong.tamu.edu

Bizarre.  I don't remember ever having installed numpy, it doesn't  
appear to be anywhere on my system, and if I do an "import numpy" it  
fails.  Maybe matplotlib is getting confused somehow?  How does it  
try to detect if numpy is installed?

On Feb 9, 2006, at 10:12 AM, Jeff Whitaker wrote:

> Craig Maloney wrote:
>
>> Thanks, Jeff.
>> That did the trick.  I didn't even have a .matplotlibrc file.    
>> Default behavior must be numpy, huh?
>> --Craig
>>
>
> Craig:
>
> If you have numpy installed, it will choose that as the default.   
> If numpy is not installed, Numeric is the default.
>
> -Jeff
>
> -- 
> Jeffrey S. Whitaker         Phone  : (303)497-6313
> Meteorologist               FAX    : (303)497-6449
> NOAA/OAR/PSD  R/PSD1        Email  : Jef...@no...
> 325 Broadway                Office : Skaggs Research Cntr 1D-124
> Boulder, CO, USA 80303-3328 Web    : http://tinyurl.com/5telg
>

On Thursday 09 February 2006 12:21, Robert Hetland wrote:
> On Feb 9, 2006, at 10:35 AM, Samuel M. Smith wrote:
> > So if I comment out fontenc, will that break .eps but fix .ps?
>
> Well, no.  In my experience, this fixes EPS (which is otherwise
> broken).  I also got the xpdf distiller to work, which places the
> fonts in the file -- good for editing in Illustrator later.

Getting this result was extremely important to me. I believe that some 
journals will insist on being able to open the image in, say, adobe 
illustrator, so they can change the fonts.

Craig Maloney wrote:
> Thanks, Jeff.
> 
> That did the trick.  I didn't even have a .matplotlibrc file.   Default 
> behavior must be numpy, huh?
> 
> --Craig
> 

Craig:

If you have numpy installed, it will choose that as the default.  If 
numpy is not installed, Numeric is the default.

-Jeff

-- 
Jeffrey S. Whitaker         Phone  : (303)497-6313
Meteorologist               FAX    : (303)497-6449
NOAA/OAR/PSD  R/PSD1        Email  : Jef...@no...
325 Broadway                Office : Skaggs Research Cntr 1D-124
Boulder, CO, USA 80303-3328 Web    : http://tinyurl.com/5telg

Thanks, Jeff.

That did the trick.  I didn't even have a .matplotlibrc file.   
Default behavior must be numpy, huh?

--Craig

On Feb 9, 2006, at 10:03 AM, Jeff Whitaker wrote:

> Craig Maloney wrote:
>
>> I'd like to use Numeric-23.1 since that's what I have installed.
>> Will that not work with matplotlib?
>> Thanks,
>> Craig
>>
>
> Craig:  That should work.  Do you have numerix='Numeric' in your  
> matplotlibrc?
>
> You might want to upgrade to Numeric 24.2, but I don't think this  
> is the source of your problem.
>
> -Jeff
>
>
> -- 
> Jeffrey S. Whitaker         Phone  : (303)497-6313
> Meteorologist               FAX    : (303)497-6449
> NOAA/OAR/PSD  R/PSD1        Email  : Jef...@no...
> 325 Broadway                Office : Skaggs Research Cntr 1D-124
> Boulder, CO, USA 80303-3328 Web    : http://tinyurl.com/5telg
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through  
> log files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD  
> SPLUNK!
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=103432&bid=230486&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

On Thursday 09 February 2006 10:42, John Hunter wrote:
> >>>>> "Darren" == Darren Dale <dd...@co...> writes:
>
>     Darren> There is some issue related to the dvipng hack that I dont
>     Darren> understand. Robert and Samuel want to apply the hack to
>     Darren> dvipng version 1.6. I also have version 1.6, but if I
>     Darren> apply the hack, then my output doesnt look right. Whats
>     Darren> going on here? Are there two OS dependent versions of
>     Darren> dvipng-1.6?
>
> According to the release notes for 1.6
>
>   The 1.6 release adds proper alpha channel support and xcolor
>   support.  It also reestablishes identification of preview-latex'
>   tightpage mode option.
>
> Not sure why we are seeing differences between 1.6 on different
> platforms, unless there is an endian issue in the png files.

I don't know how to proceed on this. Do I check the platform and apply the 
hack for darwin? Would that break usetex for any other mac users out there? 

Craig Maloney wrote:
> I'd like to use Numeric-23.1 since that's what I have installed.
> 
> Will that not work with matplotlib?
> 
> Thanks,
> Craig

Craig:  That should work.  Do you have numerix='Numeric' in your 
matplotlibrc?

You might want to upgrade to Numeric 24.2, but I don't think this is the 
source of your problem.

-Jeff


-- 
Jeffrey S. Whitaker         Phone  : (303)497-6313
Meteorologist               FAX    : (303)497-6449
NOAA/OAR/PSD  R/PSD1        Email  : Jef...@no...
325 Broadway                Office : Skaggs Research Cntr 1D-124
Boulder, CO, USA 80303-3328 Web    : http://tinyurl.com/5telg

I'd like to use Numeric-23.1 since that's what I have installed.

Will that not work with matplotlib?

Thanks,
Craig

On Feb 9, 2006, at 4:53 AM, Jeff Whitaker wrote:

>
> On Wed, 8 Feb 2006 21:17:51 -0800, "Craig Maloney"
> <cma...@ki...> said:
>
>> Hi all.
>>
>> The contour plot demos are failing on my installation with a
>> "ValueError: Arguments x, y, z, mask (if present) must be 2D arrays."
>>
>> Anyone know what's up?
>>
>> Thanks.
>> --Craig
>>
>
> Craig:  I assume you are using numpy?  If so, you need the latest
> matplotlib (0.86.2) - there may even be a few fixes committed to  
> CVS to
> fix numpy support that didn't make it in to 0.86.2.  The latest CVS
> version works well with numpy.
>
> -Jeff
> ------
> Jeffrey S. Whitaker
> Phone: (303)497-6313
> FAX: (303)497-6449
> NOAA/OAR/CDC  R/CDC1
> 325 Broadway, Boulder, CO, USA 80305-3328
> http://www.cdc.noaa.gov/people/jeffrey.s.whitaker
>
>

On Feb 9, 2006, at 10:35 AM, Samuel M. Smith wrote:

> So if I comment out fontenc, will that break .eps but fix .ps?

Well, no.  In my experience, this fixes EPS (which is otherwise  
broken).  I also got the xpdf distiller to work, which places the  
fonts in the file -- good for editing in Illustrator later.

-Rob

-----
Rob Hetland, Assistant Professor
Dept of Oceanography, Texas A&M University
p: 979-458-0096, f: 979-845-6331
e: he...@ta..., w: http://pong.tamu.edu

look at two_scales.py, it says it is possible with 2 X axes
Panos Kassianidis wrote:

> Hi,
>
> I would like to create a date plot containing historic data overlayed 
> upon the actual data.
> For example plot the temperature this December vs the temperature from 
> November last year.
> Can I use two x-axes, one at the top and one at the bottom to plot 
> these 2 different temperature lines?
> Thank you very much.
>
>             Panos
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log 
> files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

On 09 Feb, 2006, at 08:58, Robert Hetland wrote:

>
> This was a font problem I was having in LaTeX compilation.  Many of  
> the issues that you and I are having are slightly beyond mpl, and  
> have to do with different ways different versions of LaTeX deal  
> with fonts, etc.  The change to texmanager.py (below) was to not  
> load the fontenc package.   When this package was loaded, LaTeX was  
> producing illegible postscript (for reasons I don't entirely  
> understand).   This was *separate* from the dvipng hack, which was  
> required for me to get decent bitmapped images.  The excluding the  
> fontenc package is required for EPS (vector) output, so these two  
> issues are completely separate.

I have never been able to get .ps files that will convert to .pdf  
using adobe distiller. Always get a font related error and distiller  
bails.
But the .eps files work fine for converting to .pdf.
So if I comment out fontenc, will that break .eps but fix .ps?

>
> The texmanager.py simply performs a number of LaTeX opperations.   
> The best advice I got was to try to do as much of this by hand as  
> possible, to see where the potential problems lie.  Turn on the  
> debug-annoying to see everything that mpl is doing while it creates  
> your plots.
>
> -Rob
>
>
> On Feb 9, 2006, at 8:59 AM, Samuel M. Smith wrote:
>
>> what is this doing differently than the 1.6 hack?
>> what is wrong with the eps files?
>>
>>>   Also, commenting out line 141 in texmanager:
>>>
>>>         self._font_preamble = '\n'.join([r'\usepackage{type1cm}',
>>>                              cmd,
>>> #                             r'\usepackage[T1]{fontenc}',     #  
>>> The guilty line
>>>                              r'\usepackage{textcomp}'])k
>>>
>>> allowed me to save EPS files that work very well.
>>>
>>> Finally, for those of you interested in editing EPS files in  
>>> Illustrator, you need to worry about the absence of the Computer  
>>> Modern fonts in Adobe products.  This link (really work checking  
>>> out if you do scientific graphics on a Mac) explains how to get  
>>> everything working for Illustrator CS and CS2:
>>
>
> -----
> Rob Hetland, Assistant Professor
> Dept of Oceanography, Texas A&M University
> p: 979-458-0096, f: 979-845-6331
> e: he...@ta..., w: http://pong.tamu.edu
>

**********************************************************************
Samuel M. Smith Ph.D.
2966 Fort Hill Road
Eagle Mountain, Utah 84043
801-768-2768 voice
801-768-2769 fax
**********************************************************************
"The greatest source of failure and unhappiness in the world is
giving up what we want most for what we want at the moment"
**********************************************************************

On 06 Feb, 2006, at 12:34, Christopher Barker wrote:

> Samuel M. Smith wrote:
>> I would like to create an animation that shows the path of a  
>> vehicle moving
>> along a trajectory. The vehicle could be represented by a small  
>> circle with and arrow
>> pointing along the vehicle's current heading. It would be nice to  
>> leave a bread crumb trail
>> of the vehicles path as a line.
>
> Sam,
>
> This isn't a MPL solution, but as I notice you're using wxAgg, I  
> thought you might be interested in a wxPython solution. It would be  
> very easy to do this with wx.lib.floatcanvas. There is a demo for  
> it in the wxPython demo, and, if you want, I can send you the  
> latest version with a few more demos of it's capabilities.

Thanks for the suggestion. I haven't really used WxPython so there is  
a little bit of a learning curve and I was trying to get something
to work I had a presentation and not a lot of time.
I have installed the wxpython 2.6.2.1 demo distribution. Do you have  
more demos than what is in the distribution. If so I would appreciate  
getting them
as I would like to explore using wx python

I figured out a way to do it in Matplotlib. What I did is just draw  
lines and then use the line.set_data() method to update the position  
for each frame
My code is a little ugly and not the most speed efficient but I was  
able to generate the frames for movies. I used the "export slideshow  
to movie" command in GraphicConverter to convert the frames to a  
quicktime movie.
I couldn't really grok how transforms work so I couldn't move patches  
around.
The biggest problem is keeping the axis scaled which I never solved.  
I would have to play with the figure size.

I couldn't get real time animation to work, the screen never updates,  
but I could make movies.
Have you ever done an animation in WXAgg? When I run the example code  
from the cookbook, the screen doesn't update.
I think its broken on OS X.



Here is my code

>>>>> "Darren" == Darren Dale <dd...@co...> writes:
    Darren> There is some issue related to the dvipng hack that I dont
    Darren> understand. Robert and Samuel want to apply the hack to
    Darren> dvipng version 1.6. I also have version 1.6, but if I
    Darren> apply the hack, then my output doesnt look right. Whats
    Darren> going on here? Are there two OS dependent versions of
    Darren> dvipng-1.6?

According to the release notes for 1.6

  The 1.6 release adds proper alpha channel support and xcolor
  support.  It also reestablishes identification of preview-latex'
  tightpage mode option.

Not sure why we are seeing differences between 1.6 on different
platforms, unless there is an endian issue in the png files.

JDH

On Thursday 09 February 2006 09:59, Samuel M. Smith wrote:
> what is this doing differently than the 1.6 hack?
> what is wrong with the eps files?

The following discussion addresses a problem found in the latex code, and has 
nothing to do with the dvipng hack. I am working on fixing these problems in 
cvs, but I need to do some testing first.

> >   Also, commenting out line 141 in texmanager:
> >
> >         self._font_preamble = '\n'.join([r'\usepackage{type1cm}',
> >                              cmd,
> > #                             r'\usepackage[T1]{fontenc}',     #
> > The guilty line
> >                              r'\usepackage{textcomp}'])k
> >
> > allowed me to save EPS files that work very well.

There is some issue related to the dvipng hack that I dont understand. Robert 
and Samuel want to apply the hack to dvipng version 1.6. I also have version 
1.6, but if I apply the hack, then my output doesnt look right. Whats going 
on here? Are there two OS dependent versions of dvipng-1.6?

Darren

Hi,

I would like to create a date plot containing historic data overlayed 
upon the actual data.
For example plot the temperature this December vs the temperature from 
November last year.
Can I use two x-axes, one at the top and one at the bottom to plot 
these 2 different temperature lines?
Thank you very much.

			Panos

what is this doing differently than the 1.6 hack?
what is wrong with the eps files?

>   Also, commenting out line 141 in texmanager:
>
>         self._font_preamble = '\n'.join([r'\usepackage{type1cm}',
>                              cmd,
> #                             r'\usepackage[T1]{fontenc}',     #  
> The guilty line
>                              r'\usepackage{textcomp}'])k
>
> allowed me to save EPS files that work very well.
>
> Finally, for those of you interested in editing EPS files in  
> Illustrator, you need to worry about the absence of the Computer  
> Modern fonts in Adobe products.  This link (really work checking  
> out if you do scientific graphics on a Mac) explains how to get  
> everything working for Illustrator CS and CS2:

I made the change and that fixed the problem. The output of dvipng -- 
version is below.
Thank you very much.
I assume that the fix will be in the cvs version?

One more question. Even though I have my rc file set to use type1cm font
font.latex.package  : type1cm
It uses an non latex font for on screen display. The saved version of  
the plot uses
a latex font but the latex font is always serif regardless of the  
font.family setting.

How do I get the onscreen display and the file saved to use the serif  
version of the typelcm font?


dvipng --version
This is dvipng 1.6 Copyright 2002-2005 Jan-Ake Larsson
dvipng 1.6
kpathsea version 3.5.5
Copyright (C) 2002-2005 Jan-Ake Larsson.
There is NO warranty.  You may redistribute this software
under the terms of the GNU General Public License.
For more information about these matters, see the files
named COPYING and dvipng.c.


On 09 Feb, 2006, at 06:47, John Hunter wrote:

>>>>>> "Samuel" == Samuel M Smith <sm...@sa...> writes:
>
>     Samuel> I am having a similar problem with online display Latex in
>     Samuel> matplotlib WXAgg where the display is low resolution .  It
>     Samuel> sounds like you fixed it. I tried to follow the email
>     Samuel> exchange but I am not sure everything I need to know is
>     Samuel> there.  Could you summarize exactly what the fix is?
>
>     Samuel> I have matplotlib 0.86.2 and OSX 10.4.4
>
> Try editing matplotlib/texmanager.py and replacing the < with <= in
> the following code; it should read
>
>             if vers<='1.6':
>                 # hack the alpha channel as described in comment above
>                 alpha = sqrt(1-X[:,:,0])
>             else:
>                 alpha = X[:,:,-1]
>
> also, print the version number and let us know what you have.
>
> Make sure you rm -rf the ~/.matplotlib/tex.cache after reinstalling.
>
> JDH

**********************************************************************
Samuel M. Smith Ph.D.
2966 Fort Hill Road
Eagle Mountain, Utah 84043
801-768-2768 voice
801-768-2769 fax
**********************************************************************
"The greatest source of failure and unhappiness in the world is
giving up what we want most for what we want at the moment"
**********************************************************************

I am using the I-Installer version of teTeX on Mac OS X.  I imagine  
that many other Mac users might be experiencing similar problems.   
Here is what worked for me, with many thanks to Darren for helping me  
work through this:

First, using the dvipng 'hack' for dvipng versions less than 1.6  
worked to get me PNG output using LaTeX.  Also, commenting out line  
141 in texmanager:

         self._font_preamble = '\n'.join([r'\usepackage{type1cm}',
                              cmd,
#                             r'\usepackage[T1]{fontenc}',     # The  
guilty line
                              r'\usepackage{textcomp}'])k

allowed me to save EPS files that work very well.

Finally, for those of you interested in editing EPS files in  
Illustrator, you need to worry about the absence of the Computer  
Modern fonts in Adobe products.  This link (really work checking out  
if you do scientific graphics on a Mac) explains how to get  
everything working for Illustrator CS and CS2:

http://darkwing.uoregon.edu/~noeckel/Illustrator.html


Finally, an answer to Darren's qustion:

On Feb 9, 2006, at 8:06 AM, Darren Dale wrote:

> WOuld you send me the output of dvipng -v?

~$ dvipng --version
This is dvipng 1.6 Copyright 2002-2005 Jan-Ake Larsson
dvipng 1.6
kpathsea version 3.5.5
Copyright (C) 2002-2005 Jan-Ake Larsson.
There is NO warranty.  You may redistribute this software
under the terms of the GNU General Public License.
For more information about these matters, see the files
named COPYING and dvipng.c.


-Rob


-----
Rob Hetland, Assistant Professor
Dept of Oceanography, Texas A&M University
p: 979-458-0096, f: 979-845-6331
e: he...@ta..., w: http://pong.tamu.edu

Hi Robert,

I am working on clearing up the code for cvs, but wanted to double check with 
you: what did you have to do to fix the dvipng problem? I ask because I also 
have dvipng version 1.6, and dont have any problems with the cvs codebase. 
WOuld you send me the output of dvipng -v?

On Tuesday 07 February 2006 17:35, Robert Hetland wrote:
> On Feb 7, 2006, at 12:38 PM, John Hunter wrote:
> >             if vers<'1.6':
>
> This seemed to be the problem (perhaps vers<='1.6' is meant).  I use
> version dvipng version 1.6, but the alpha hack still seems to be
> required.  When I force the hack, the PNG output looks good.  Three
> cheers for the MPL-users list!  (Four for JDH!)
>
> Now, if you'll bear with me, I need to get EPS (then, ultimately,
> encapsulated PDF) output.  I am getting an error like (from
>
> verbose:helpful) :
>  >>> savefig('foo.eps')
>
> [.....]
>
> dvips: Font ecrm1000 at 72 not found; scaling 600 instead.
> dvips: Such scaling will generate extremely poor output.
>
>
> This is dvips(k) 5.95b Copyright 2005 Radical Eye Software
> (www.radicaleye.com)
> ' TeX output 2006.02.07:1621' -> /tmp/
> bc5f66f5eb5ed881504e9f305289dc95.ps
> <tex.pro><psfrag.pro><special.pro><color.pro>. [1
> <bc5f66f5eb5ed881504e9f305289dc95.eps>]
>
> GNU Ghostscript 8.16: Unrecoverable error, exit code 1
>
>
>
> Any clues?
>
> -Rob
>
>
> -----
> Rob Hetland, Assistant Professor
> Dept of Oceanography, Texas A&M University
> p: 979-458-0096, f: 979-845-6331
> e: he...@ta..., w: http://pong.tamu.edu

-- 
Darren S. Dale, Ph.D.
Cornell High Energy Synchrotron Source
Cornell University
200L Wilson Lab
Rt. 366 & Pine Tree Road
Ithaca, NY 14853

dd...@co...
office: (607) 255-9894
fax: (607) 255-9001

>>>>> "Samuel" == Samuel M Smith <sm...@sa...> writes:

    Samuel> I am having a similar problem with online display Latex in
    Samuel> matplotlib WXAgg where the display is low resolution .  It
    Samuel> sounds like you fixed it. I tried to follow the email
    Samuel> exchange but I am not sure everything I need to know is
    Samuel> there.  Could you summarize exactly what the fix is?

    Samuel> I have matplotlib 0.86.2 and OSX 10.4.4

Try editing matplotlib/texmanager.py and replacing the < with <= in
the following code; it should read

            if vers<='1.6':
                # hack the alpha channel as described in comment above
                alpha = sqrt(1-X[:,:,0])
            else:
                alpha = X[:,:,-1]

also, print the version number and let us know what you have.

Make sure you rm -rf the ~/.matplotlib/tex.cache after reinstalling.

JDH