matplotlib-users Mailing List for matplotlib

Gerardo,
I think the approach used in this code should do what you want. Dorry I 
don't have time for more explanation...
-Chris
Gerardo Rivera wrote:
> Hi,
> 
> I'm trying to draw a line given an angle, magnitude, x0 and y0 
> location on a 2d line plot. I use the "cosine" and "sine" to find 
> the x1 and y1 locations so that I can draw a line segment from x0, y0 
> to x1, y1. This works fine on a 200 x 200 (Width x Height) pixel 
> graph. However, I have an input form where the user can change the 
> plot size, for example 200x800 or 800x200.
> 
> The "magnitude" is along the y-axis and "time" is along the x-axis. 
> At a particular point in time the magnitude and angle is given and I 
> have to calculate the x1 and y1 values. The max magnitude along the 
> y-axis (height) is always the same whether it is 200 pixels in height 
> or 800 pixels in height.
> 
> The scientist I'm working with wants to see the magnitude (x1,y1) of 
> each plot look correct if the graph is 800 pixels or 200 pixels in 
> height. He also wants the correct angle if you stretch the graph 
> from 200 pixels wide to 800 pixels wide and then down to 100 pixels 
> wide.
> 
> I've been trying to use a ratio of the graph size to maintain the 
> correct sizing but with no luck. Basically, the scientist want to 
> maintain the same angle and reduce or increase the magnitude in the 
> y-axis if the graph height is reduced or increased.
> 
> Does anyone have an answer to this solution? Maintaining the same 
> angle and varying the magnitude based on the y-axis height.
> 
> Since, I have a lot of points I started to use a LineCollection to 
> speed the rendering speed.
> 
> Gerardo
> 
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
-- 
Christopher Barker, Ph.D.
Oceanographer
 		
NOAA/OR&R/HAZMAT (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chr...@no...

Hi,
I have different questions to use tex with matplotlib.
The first one:
If I'm doing:
rc('text', usetex=True)
xlabel('$\textrm{toto}$') # bad
xlabel(r'$\textrm{toto}$') # ok
The results are not the same and I don't understand why.
and for:
xlabel('$\textit{toto}$') #bad
xlabel(r'$\textit{toto}$') #ok
same things, but it's working fine for:
xlabel('$\it{toto}$') #ok
xlabel(r'$\it{toto}$') #ok
but it's not working for \bf:
xlabel('$\textbf{toto}$') # bad
xlabel(r'$\textbf{toto}$') # ok
xlabel('$\bf{toto}$') # bad
xlabel(r'$\bf{toto}$') # ok
And this bring to the second question, I would like to create the xlabel with 
latex stuf inside somewhere in my script:
label='$\textrm{test}_2$'
but that can't work:
xlabel(label) # bad
How to tell that I want the "r" before the chains to have TeX working like 
expected?
Thanks,
N.

Hi,
I'm trying to draw a line given an angle, magnitude, x0 and y0 
location on a 2d line plot. I use the "cosine" and "sine" to find 
the x1 and y1 locations so that I can draw a line segment from x0, y0 
to x1, y1. This works fine on a 200 x 200 (Width x Height) pixel 
graph. However, I have an input form where the user can change the 
plot size, for example 200x800 or 800x200.
The "magnitude" is along the y-axis and "time" is along the x-axis. 
At a particular point in time the magnitude and angle is given and I 
have to calculate the x1 and y1 values. The max magnitude along the 
y-axis (height) is always the same whether it is 200 pixels in height 
or 800 pixels in height.
The scientist I'm working with wants to see the magnitude (x1,y1) of 
each plot look correct if the graph is 800 pixels or 200 pixels in 
height. He also wants the correct angle if you stretch the graph 
from 200 pixels wide to 800 pixels wide and then down to 100 pixels 
wide.
I've been trying to use a ratio of the graph size to maintain the 
correct sizing but with no luck. Basically, the scientist want to 
maintain the same angle and reduce or increase the magnitude in the 
y-axis if the graph height is reduced or increased.
Does anyone have an answer to this solution? Maintaining the same 
angle and varying the magnitude based on the y-axis height.
Since, I have a lot of points I started to use a LineCollection to 
speed the rendering speed.
Gerardo

John Hunter wrote:
>>>>>> "Karl" == Karl Guertin <gra...@gm...> writes:
>>>>>> 
>
> Karl> I have a very customized matplotlibrc for rendering charts
> Karl> for web pages. The biggest change is that I render
> Karl> everything white on black by default. This has worked fine
> Karl> up through 0.87.5 (the last release I grabbed) but in
> Karl> 0.87.7, the ticks on the axes are black (the axes color is
> Karl> white). Am I missing a setting somewhere or is this a bug?
>
>
> This looks like it may be a bug introduced by some recent changes
> Norbert made to the rc handling, but hard to say.
>
> Make sure matplotlib is picking up the rc file you think it is by
> running with --verbose-helpful and if so, send that file along and
> we'll take a look.
> 
Maybe, there is a connection, but I don't quite understand how it would
come.
Ticks are rendered as linemarkers. They are not filled, so, in the new
version, markeredgecolor should follow the line color automatically. If
you set the line color to white, the tick color should also become white.
Could you send along your matplotlibrc and specify in detail how to
reproduce the problem?

John Hunter wrote:
>>>>>> "David" == David Goldsmith <Dav...@no...> writes:
>>>>>> 
>
> David> OK, I was afraid of that; in that case, is there some way
> David> to get the height and width of the legend (so I can do what
> David> I want programatically)? Thanks again,
>
> Again, afraid not. At least nothing obvious. The legend placement is
> done dynamically at draw time, and so it will be difficult to get this
> information ahead of time. 
Actually, I don't care (tremendously) if I have to wait 'til after it's 
drawn to get the info (I've already found other instances where, to do 
what I've wanted, I've had to draw, then get/modify/set then redraw - as 
you say, inelegant, but serviceable; it's not like the old days where 
re-drawing meant waiting half a minute) so if that info is available 
after the figure's been drawn, that's good enough for me (as long as 
it's possible to set it as well as get it) for now. Is that what 
legendPatch.get_window_extent().get_bounds() does? (I'll probably find 
out using dir before you reply, but a reply would be appreciated anyway, 
so the info will be in the archive).
> There might be some cleverness that can be
> applied, but nothing easy.
>
> One option would be connect to the draw event, and then inspect the
> legend properties, and then place it where you want knowing the width
> and the height. Not too elegant, but serviceable. Here is an 
> untested sketch
>
>
> def ondraw(event):
> if ondraw.done: return 
> # in pixels
> left,bottom,width,height = leg.legendPatch.get_window_extent().get_bounds()
> # move your legend....
> ondraw.done = True
> ondraw.done = False 
>
> fig = figure()
> ax = fig.add_subplot(111)
> leg = ax.legend(blah)
> fig.canvas.mpl_connect('draw_event', ondraw)
>
>
> It might be better to patch legend directly to do what you want and
> send the patch our way. 
I'm getting closer to the point where I feel like I might be capable of 
doing this, but I'm not quite there yet. :-)
> Or subclass it.
> 
This is closer to my confidence level.
Thanks!
DG
> JDH
> 
-- 
HMRD/ORR/NOS/NOAA <http://response.restoration.noaa.gov/emergencyresponse/>

John Hunter wrote:
> The legend placement is done dynamically at draw time,
Ah, so it looks like it does make sense for the user to specify an 
alignment, and have it figured out at draw time.
> It might be better to patch legend directly to do what you want and
> send the patch our way. Or subclass it.
That does seem the way to go. David, I've done stuff like this for the 
wxPython FloatCavnas -- perhaps I can help, if you want to do it.
-Chris
-- 
Christopher Barker, Ph.D.
Oceanographer
 		
NOAA/OR&R/HAZMAT (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chr...@no...

On 2006年10月26日, Michele Vallisneri apparently wrote: 
> I've been building matplotlib 0.87.7 on a 10.4.7 OS X system, with 
> Universal Python 2.4.3 from pythonmac.org. I've been following 
> instructions at 
> http://sourceforge.net/mailarchive/message.php?msg_id=36901627 
Please keep posting your experiences.
At work I'm about to spend my first ever time on a Mac,
and I expected this to be a blessing,
but such reports worry me!
Thanks,
Alan Isaac

>>>>> "David" == David Goldsmith <Dav...@no...> writes:
 David> OK, I was afraid of that; in that case, is there some way
 David> to get the height and width of the legend (so I can do what
 David> I want programatically)? Thanks again,
Again, afraid not. At least nothing obvious. The legend placement is
done dynamically at draw time, and so it will be difficult to get this
information ahead of time. There might be some cleverness that can be
applied, but nothing easy.
One option would be connect to the draw event, and then inspect the
legend properties, and then place it where you want knowing the width
and the height. Not too elegant, but serviceable. Here is an 
untested sketch
def ondraw(event):
 if ondraw.done: return 
 # in pixels
 left,bottom,width,height = leg.legendPatch.get_window_extent().get_bounds()
 # move your legend....
 ondraw.done = True
ondraw.done = False 
fig = figure()
ax = fig.add_subplot(111)
leg = ax.legend(blah)
fig.canvas.mpl_connect('draw_event', ondraw)
It might be better to patch legend directly to do what you want and
send the patch our way. Or subclass it.
JDH

John Hunter wrote:
>>>>>> "David" == David Goldsmith <Dav...@no...> writes:
>>>>>> 
>
> David> Hi! OK, loc=(a,b) positions the legend, and appears to
> David> place the lower left corner at (a,b) (axes coords.), right?
>
> yes
>
> David> Is there some way to say that (a,b) should specify the
> David> location of, say, the center of the legend? Thanks!
>
> Afraid not.
> 
OK, I was afraid of that; in that case, is there some way to get the 
height and width of the legend (so I can do what I want 
programatically)? Thanks again,
DG
> JDH
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 

>>>>> "David" == David Goldsmith <Dav...@no...> writes:
 David> Hi! OK, loc=(a,b) positions the legend, and appears to
 David> place the lower left corner at (a,b) (axes coords.), right?
yes
 David> Is there some way to say that (a,b) should specify the
 David> location of, say, the center of the legend? Thanks!
Afraid not.
JDH

>>>>> "Eric" == Eric Firing <ef...@ha...> writes:
 >> I am trying to build Matplotlib 0.87.7 on an RHEL 4.4-derived
 >> system, with Python 2.3.4. I am running into a weird
 >> problem. When I do "python setup.py build", I get the following
 >> error:
It looks like this bug
http://matplotlib.sourceforge.net/faq.html#PYGTK24

Stephen,
On the dev list was an ANN post which stated that 0.87.7:
This release is compiled against numpy-1.0 final. The binaries are
fresh on sourceforge, so they may take some time to propagate to the
mirrors.
http://cheeseshop.python.org/pypi/matplotlib/
http://sourceforge.net/project/showfiles.php?group_id=80706&package_id=82474
Note: There is a compile error on python2.3 that will probably require
a slight modification to the released source. Please allow some time
for us to fix this and post the binaries.
Werner
Stephen Uhlhorn wrote:
> Sorry if this has been answered already, but I'm having trouble
> finding the answer--
> 
> What version of matplotlib is compatible with numpy 1.0 (final)? Or,
> should I continue to use numpy-1.0rc2 w/ mpl 0.87.7? I'm building from
> source on mac os using fink, and have had no trouble until numpy
> 1.0rc3 came out.
> 
> -stephen
> 
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

Sorry if this has been answered already, but I'm having trouble
finding the answer--
What version of matplotlib is compatible with numpy 1.0 (final)? Or,
should I continue to use numpy-1.0rc2 w/ mpl 0.87.7? I'm building from
source on mac os using fink, and have had no trouble until numpy
1.0rc3 came out.
-stephen

Hi! OK, loc=(a,b) positions the legend, and appears to place the lower 
left corner at (a,b) (axes coords.), right? Is there some way to say 
that (a,b) should specify the location of, say, the center of the 
legend? Thanks!
DG

I wrote a plotting module for matplotlib that can be used with Boa or easily
integrated into your own wxPython application.
If you are interested in the plugin, pls. send e-mail to kh...@in....
There are some examples here.
file: Frame1.py was greatet by Boa.
I added a few lines of codes there.
Two pictures are added for illustration.
Igor Khromushin
file: Frame1.py
#Boa:Frame:Frame1
import wx
from scipy import *
from pylab import meshgrid
import wx.lib.igorplot
def create(parent):
 return Frame1(parent)
[wxID_FRAME1, wxID_FRAME1BUTTON1, wxID_FRAME1IGORPLOTCANVAS1,
 wxID_FRAME1IGORPLOTCANVAS2, wxID_FRAME1IGORPLOTCANVAS3,
 wxID_FRAME1IGORPLOTCANVAS4, wxID_FRAME1IGORPLOTCANVAS5,
 wxID_FRAME1IGORPLOTCANVAS6, wxID_FRAME1IGORPLOTCANVAS7,
 wxID_FRAME1IGORPLOTCANVAS8, wxID_FRAME1IGORPLOTCANVAS9,
] = [wx.NewId() for _init_ctrls in range(11)]
class Frame1(wx.Frame):
 def _init_coll_flexGridSizer1_Items(self, parent):
 # generated method, don't edit
 parent.AddWindow(self.igorPlotCanvas1, 1, border=0, flag=wx.GROW)
 parent.AddWindow(self.igorPlotCanvas2, 1, border=0, flag=wx.GROW)
 parent.AddWindow(self.igorPlotCanvas3, 1, border=0, flag=wx.GROW)
 parent.AddWindow(self.igorPlotCanvas4, 1, border=0, flag=wx.GROW)
 parent.AddWindow(self.igorPlotCanvas5, 1, border=0, flag=wx.GROW)
 parent.AddWindow(self.igorPlotCanvas6, 1, border=0, flag=wx.GROW)
 parent.AddWindow(self.igorPlotCanvas7, 1, border=0, flag=wx.GROW)
 parent.AddWindow(self.igorPlotCanvas8, 1, border=0, flag=wx.GROW)
 parent.AddWindow(self.igorPlotCanvas9, 1, border=0, flag=wx.GROW)
 parent.AddSpacer(wx.Size(8, 8), border=0, flag=0)
 parent.AddWindow(self.button1, 0, border=0, flag=wx.ALIGN_CENTER)
 def _init_coll_flexGridSizer1_Growables(self, parent):
 # generated method, don't edit
 parent.AddGrowableRow(0)
 parent.AddGrowableRow(1)
 parent.AddGrowableRow(2)
 parent.AddGrowableCol(0)
 parent.AddGrowableCol(1)
 parent.AddGrowableCol(2)
 def _init_sizers(self):
 # generated method, don't edit
 self.flexGridSizer1 = wx.FlexGridSizer(cols=3, hgap=0, rows=4,
vgap=0)
 self._init_coll_flexGridSizer1_Items(self.flexGridSizer1)
 self._init_coll_flexGridSizer1_Growables(self.flexGridSizer1)
 self.SetSizer(self.flexGridSizer1)
 def _init_ctrls(self, prnt):
 # generated method, don't edit
 wx.Frame.__init__(self, id=wxID_FRAME1, name='', parent=prnt,
 pos=wx.Point(187, 73), size=wx.Size(1044, 847),
 style=wx.DEFAULT_FRAME_STYLE, title='Frame1')
 self.SetClientSize(wx.Size(1036, 813))
 self.igorPlotCanvas1 =
wx.lib.igorplot.IgorPlotCanvas(id=wxID_FRAME1IGORPLOTCANVAS1,
 name='igorPlotCanvas1', parent=self, pos=wx.Point(0, 0),
 size=wx.Size(337, 263), style=0)
 self.igorPlotCanvas2 =
wx.lib.igorplot.IgorPlotCanvas(id=wxID_FRAME1IGORPLOTCANVAS2,
 name='igorPlotCanvas2', parent=self, pos=wx.Point(337, 0),
 size=wx.Size(362, 263), style=0)
 self.igorPlotCanvas3 =
wx.lib.igorplot.IgorPlotCanvas(id=wxID_FRAME1IGORPLOTCANVAS3,
 name='igorPlotCanvas3', parent=self, pos=wx.Point(699, 0),
 size=wx.Size(337, 263), style=0)
 self.igorPlotCanvas4 =
wx.lib.igorplot.IgorPlotCanvas(id=wxID_FRAME1IGORPLOTCANVAS4,
 name='igorPlotCanvas4', parent=self, pos=wx.Point(0, 263),
 size=wx.Size(337, 263), style=0)
 self.igorPlotCanvas5 =
wx.lib.igorplot.IgorPlotCanvas(id=wxID_FRAME1IGORPLOTCANVAS5,
 name='igorPlotCanvas5', parent=self, pos=wx.Point(337, 263),
 size=wx.Size(362, 263), style=0)
 self.igorPlotCanvas6 =
wx.lib.igorplot.IgorPlotCanvas(id=wxID_FRAME1IGORPLOTCANVAS6,
 name='igorPlotCanvas6', parent=self, pos=wx.Point(699, 263),
 size=wx.Size(337, 263), style=0)
 self.igorPlotCanvas7 =
wx.lib.igorplot.IgorPlotCanvas(id=wxID_FRAME1IGORPLOTCANVAS7,
 name='igorPlotCanvas7', parent=self, pos=wx.Point(0, 526),
 size=wx.Size(337, 263), style=0)
 self.igorPlotCanvas8 =
wx.lib.igorplot.IgorPlotCanvas(id=wxID_FRAME1IGORPLOTCANVAS8,
 name='igorPlotCanvas8', parent=self, pos=wx.Point(337, 526),
 size=wx.Size(362, 263), style=0)
 self.igorPlotCanvas9 =
wx.lib.igorplot.IgorPlotCanvas(id=wxID_FRAME1IGORPLOTCANVAS9,
 name='igorPlotCanvas9', parent=self, pos=wx.Point(699, 526),
 size=wx.Size(337, 263), style=0)
 self.button1 = wx.Button(id=wxID_FRAME1BUTTON1, label='Plot',
 name='button1', parent=self, pos=wx.Point(480, 789),
 size=wx.Size(75, 23), style=0)
 self.button1.Bind(wx.EVT_BUTTON, self.OnButton1Button,
 id=wxID_FRAME1BUTTON1)
 self._init_sizers()
 def __init__(self, parent):
 self._init_ctrls(parent)
 def OnButton1Button(self, event):
 x=y=r_[-5:5:200j]
 X,Y=meshgrid(x,y)
 Z=sin(X)+cos(Y)
 self.igorPlotCanvas1.imshow(Z,extent=(-5,5,-5,5),origin='lower')
 self.igorPlotCanvas1.axes.set_title('Test imshow with colorbar')
 
self.igorPlotCanvas2.imshowwithcontour(Z,extent=(-5,5,-5,5),origin='lower')
 self.igorPlotCanvas3.contourf(Z,extent=(-5,5,-5,5),origin='lower')
 self.igorPlotCanvas4.contour(Z,extent=(-5,5,-5,5),origin='lower')
 
self.igorPlotCanvas5.contour(Z,extent=(-5,5,-5,5),origin='lower',colors=('k',))
 
self.igorPlotCanvas6.imshow(Z,extent=(-5,5,-5,5),origin='lower',colorbar=False)
 self.igorPlotCanvas7.plot(x,x**2,x,x**3)
 self.igorPlotCanvas7.axes.grid(True)
 self.igorPlotCanvas7.axes.legend(('plot 1','plot 2'),loc='best')
 self.igorPlotCanvas8.tbvisible=False # No toolbar
 self.igorPlotCanvas8.plot(x,x**2,x,x**3)
 self.igorPlotCanvas9.tbvisible=False # No toolbar
 
self.igorPlotCanvas9.imshowwithcontour(Z,extent=(-5,5,-5,5),origin='lower',colorbar=False)
http://www.nabble.com/file/3900/runtimesample2.jpg 
http://www.nabble.com/file/3901/boadesigntime2.jpg 
-- 
View this message in context: http://www.nabble.com/Matplotlib-Boa-constructor-plug-in-tf2519494.html#a7027147
Sent from the matplotlib - users mailing list archive at Nabble.com.

Just to archive this in case anyone else has this problem, it seems
like Times as the serif font and Helvetica as the sans-serif are the
magic combination for Miktex-2.5. I can make any other fonts work
with true type fonts and usetex: true.
On 10/28/06, Darren Dale <dd...@co...> wrote:
> On Saturday 28 October 2006 10:20 am, you wrote:
> > Thanks Darren. type1cm got installed on the fly by miktex the first
> > time I ran this, but that must not have taken effect fast enough or
> > something. The miktex package manager now shows it as installed and
> > deleting my matplotlib tex.cache and ttfont.cache and re-running my
> > test leads to the usual, beautiful results (without changing the
> > default fonts).
> >
> > Thank again,
>
> You're welcome.
>
> Darren
>
>
> > On 10/28/06, Darren Dale <dd...@co...> wrote:
> > > You could try changing your font settings in matplotlibrc, maybe times
> > > for serif and helvetica for sans-serif. dvipng is trying to locate a font
> > > that hasn't been (or cannot be) generated. This is usually an issue with
> > > bitmapped fonts, which are generated on demand, so you would get lots of
> > > these messages when working with a new Miktex install.
> > >
> > > I dont know why bitmapped fonts are being used though, since usetex
> > > specifically asks for type1 fonts. Maybe you need to go into your miktex
> > > package manager and make sure you have type1cm available on your system.
> > >
> > > Darren
> > >
> > > On Saturday 28 October 2006 12:25 am, Ryan Krauss wrote:
> > > > I am trying to use latex with matplolib on Windows XP with MikTex 2.5
> > > > and dvipng 1.8.
> > > >
> > > > I am getting this warning:
> > > > dvipng warning: font cmsy10 at 512 dpi not found, characters will be
> > > > left blank
> > > >
> > > > when I do this:
> > > >
> > > > t=arange(0,1,0.01)
> > > > y=sin(2*pi*t)
> > > > plot(t,y)
> > > >
> > > > and this
> > > > ylabel('$y(t)$')
> > > >
> > > > leads to
> > > > dvipng warning: font cmr17 at 356 dpi not found, characters will be
> > > > left blank
> > > >
> > > > with no ylabel.
> > > >
> > > > I am also getting a lot of other output (see below).
> > > >
> > > > How do I fix this?
> > > >
> > > > Thanks,
> > > >
> > > > Ryan
> > > >
> > > > ==============================================
> > > > Trying to make PK font cmmi12 at 512 DPI...
> > > > Creating cmmi12.pk...
> > > > Running mf...
> > > > This is METAFONT, Version 2.71828 (MiKTeX 2.5)
> > > >
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmmi12.mf
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmbase.mf)
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\mathit.mf
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanu.mf [65] [66]
> > > > [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80]
> > > > [81] [82] [83] [84] [85] [86] [87] [88] [89] [90])
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\itall.mf [97] [98]
> > > > [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110]
> > > > [111] [112] [113] [114] [115] [116] [117] [118] [119] [120] [121]
> > > > [122]) (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greeku.mf
> > > > [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10])
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greekl.mf [11] [12]
> > > > [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26]
> > > > [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39])
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\italms.mf [64] [96]
> > > > [123] [124] [125] [126] [127])
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\olddig.mf [48] [49]
> > > > [50] [51] [52] [53] [54] [55] [56] [57])
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romms.mf [40] [41]
> > > > [42] [43] [44] [45] [46] [47] [58] [59] [60] [61] [62] [63] [91] [92]
> > > > [93] [94] [95]) ) )
> > > > Font metrics written on cmmi12.tfm.
> > > > Output written on cmmi12.512gf (128 characters, 25960 bytes).
> > > > Transcript written on cmmi12.log.
> > > > METAFONT failed for some reason
> > > > ignoring 0 strange path(s)
> > > > Running gftopk...
> > > > This is GFtoPK, Version 2.3 (MiKTeX 2.5)
> > > > 'METAFONT output 2006年10月27日:2322'
> > > > 25960 bytes packed to 14164 bytes.
> > > > dvipng warning: font cmmi12 at 512 dpi not found, characters will be
> > > > left blan Trying to make PK font cmr17 at 356 DPI...
> > > > Creating cmr17.pk...
> > > > Running mf...
> > > > This is METAFONT, Version 2.71828 (MiKTeX 2.5)
> > > >
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmr17.mf
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmbase.mf)
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\roman.mf
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanu.mf [65] [66]
> > > > [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80]
> > > > [81] [82] [83] [84] [85] [86] [87] [88] [89] [90])
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanl.mf [97] [98]
> > > > [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110]
> > > > [111] [112] [113] [114] [115] [116] [117] [118] [119] [120] [121]
> > > > [122]) (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greeku.mf
> > > > [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10])
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romand.mf [48] [49]
> > > > [50] [51] [52] [53] [54] [55] [56] [57])
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanp.mf [36] [38]
> > > > [63] [62]) (C:\Program Files\MiKTeX
> > > > 2.5\fonts\source\public\cm\romspl.mf [16] [17] [25] [26] [27] [28])
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romspu.mf [29] [30]
> > > > [31]) (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\punct.mf [33]
> > > > [60] [35] [37] [39] [40] [41] [42] [43] [44] [46] [47] [58] [59] [61]
> > > > [64] [91] [93] [96]) (C:\Program Files\MiKTeX
> > > > 2.5\fonts\source\public\cm\accent.mf [18] [19] [20] [21] [22] [23] [24]
> > > > [32] [94] [95] [125] [126] [127]) (C:\Program Files\MiKTeX
> > > > 2.5\fonts\source\public\cm\romlig.mf [11] [12] [13] [14] [15])
> > > > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\comlig.mf [34] [45]
> > > > [92] [123] [124]) ) )
> > > > Font metrics written on cmr17.tfm.
> > > > Output written on cmr17.356gf (128 characters, 24612 bytes).
> > > > Transcript written on cmr17.log.
> > > > METAFONT failed for some reason
> > > > ignoring 0 strange path(s)
> > > > Running gftopk...
> > > > This is GFtoPK, Version 2.3 (MiKTeX 2.5)
> > > > 'METAFONT output 2006年10月27日:2322'
> > > > 24612 bytes packed to 11084 bytes.
> > > > dvipng warning: font cmr17 at 356 dpi not found, characters will be
> > > > left blank
> > > >
> > > > -----------------------------------------------------------------------
> > > >-- Using Tomcat but need to do more? Need to support web services,
> > > > security? Get stuff done quickly with pre-integrated technology to make
> > > > your job easier Download IBM WebSphere Application Server v.1.0.1 based
> > > > on Apache Geronimo
> > > > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=12164
> > > >2 _______________________________________________
> > > > Matplotlib-users mailing list
> > > > Mat...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> > >
> > > --
> > > Darren S. Dale, Ph.D.
> > > dd...@co...
>
> --
> Darren S. Dale, Ph.D.
> dd...@co...
>

Thanks Darren. type1cm got installed on the fly by miktex the first
time I ran this, but that must not have taken effect fast enough or
something. The miktex package manager now shows it as installed and
deleting my matplotlib tex.cache and ttfont.cache and re-running my
test leads to the usual, beautiful results (without changing the
default fonts).
Thank again,
Ryan
On 10/28/06, Darren Dale <dd...@co...> wrote:
> You could try changing your font settings in matplotlibrc, maybe times for
> serif and helvetica for sans-serif. dvipng is trying to locate a font that
> hasn't been (or cannot be) generated. This is usually an issue with bitmapped
> fonts, which are generated on demand, so you would get lots of these messages
> when working with a new Miktex install.
>
> I dont know why bitmapped fonts are being used though, since usetex
> specifically asks for type1 fonts. Maybe you need to go into your miktex
> package manager and make sure you have type1cm available on your system.
>
> Darren
>
>
> On Saturday 28 October 2006 12:25 am, Ryan Krauss wrote:
> > I am trying to use latex with matplolib on Windows XP with MikTex 2.5
> > and dvipng 1.8.
> >
> > I am getting this warning:
> > dvipng warning: font cmsy10 at 512 dpi not found, characters will be left
> > blank
> >
> > when I do this:
> >
> > t=arange(0,1,0.01)
> > y=sin(2*pi*t)
> > plot(t,y)
> >
> > and this
> > ylabel('$y(t)$')
> >
> > leads to
> > dvipng warning: font cmr17 at 356 dpi not found, characters will be left
> > blank
> >
> > with no ylabel.
> >
> > I am also getting a lot of other output (see below).
> >
> > How do I fix this?
> >
> > Thanks,
> >
> > Ryan
> >
> > ==============================================
> > Trying to make PK font cmmi12 at 512 DPI...
> > Creating cmmi12.pk...
> > Running mf...
> > This is METAFONT, Version 2.71828 (MiKTeX 2.5)
> >
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmmi12.mf
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmbase.mf)
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\mathit.mf
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanu.mf [65] [66]
> > [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81]
> > [82] [83] [84] [85] [86] [87] [88] [89] [90])
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\itall.mf [97] [98] [99]
> > [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111]
> > [112] [113] [114] [115] [116] [117] [118] [119] [120] [121] [122])
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greeku.mf [0] [1] [2]
> > [3] [4] [5] [6] [7] [8] [9] [10])
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greekl.mf [11] [12]
> > [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27]
> > [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39])
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\italms.mf [64] [96]
> > [123] [124] [125] [126] [127])
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\olddig.mf [48] [49]
> > [50] [51] [52] [53] [54] [55] [56] [57])
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romms.mf [40] [41] [42]
> > [43] [44] [45] [46] [47] [58] [59] [60] [61] [62] [63] [91] [92] [93] [94]
> > [95]) ) )
> > Font metrics written on cmmi12.tfm.
> > Output written on cmmi12.512gf (128 characters, 25960 bytes).
> > Transcript written on cmmi12.log.
> > METAFONT failed for some reason
> > ignoring 0 strange path(s)
> > Running gftopk...
> > This is GFtoPK, Version 2.3 (MiKTeX 2.5)
> > 'METAFONT output 2006年10月27日:2322'
> > 25960 bytes packed to 14164 bytes.
> > dvipng warning: font cmmi12 at 512 dpi not found, characters will be left
> > blan Trying to make PK font cmr17 at 356 DPI...
> > Creating cmr17.pk...
> > Running mf...
> > This is METAFONT, Version 2.71828 (MiKTeX 2.5)
> >
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmr17.mf
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmbase.mf)
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\roman.mf
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanu.mf [65] [66]
> > [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81]
> > [82] [83] [84] [85] [86] [87] [88] [89] [90])
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanl.mf [97] [98]
> > [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110]
> > [111] [112] [113] [114] [115] [116] [117] [118] [119] [120] [121] [122])
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greeku.mf [0] [1] [2]
> > [3] [4] [5] [6] [7] [8] [9] [10])
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romand.mf [48] [49]
> > [50] [51] [52] [53] [54] [55] [56] [57])
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanp.mf [36] [38]
> > [63] [62]) (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romspl.mf
> > [16] [17] [25] [26] [27] [28])
> > (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romspu.mf [29] [30]
> > [31]) (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\punct.mf [33]
> > [60] [35] [37] [39] [40] [41] [42] [43] [44] [46] [47] [58] [59] [61] [64]
> > [91] [93] [96]) (C:\Program Files\MiKTeX
> > 2.5\fonts\source\public\cm\accent.mf [18] [19] [20] [21] [22] [23] [24]
> > [32] [94] [95] [125] [126] [127]) (C:\Program Files\MiKTeX
> > 2.5\fonts\source\public\cm\romlig.mf [11] [12] [13] [14] [15]) (C:\Program
> > Files\MiKTeX 2.5\fonts\source\public\cm\comlig.mf [34] [45] [92] [123]
> > [124]) ) )
> > Font metrics written on cmr17.tfm.
> > Output written on cmr17.356gf (128 characters, 24612 bytes).
> > Transcript written on cmr17.log.
> > METAFONT failed for some reason
> > ignoring 0 strange path(s)
> > Running gftopk...
> > This is GFtoPK, Version 2.3 (MiKTeX 2.5)
> > 'METAFONT output 2006年10月27日:2322'
> > 24612 bytes packed to 11084 bytes.
> > dvipng warning: font cmr17 at 356 dpi not found, characters will be left
> > blank
> >
> > -------------------------------------------------------------------------
> > Using Tomcat but need to do more? Need to support web services, security?
> > Get stuff done quickly with pre-integrated technology to make your job
> > easier Download IBM WebSphere Application Server v.1.0.1 based on Apache
> > Geronimo
> > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> > _______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
> --
> Darren S. Dale, Ph.D.
> dd...@co...
>

You could try changing your font settings in matplotlibrc, maybe times for 
serif and helvetica for sans-serif. dvipng is trying to locate a font that 
hasn't been (or cannot be) generated. This is usually an issue with bitmapped 
fonts, which are generated on demand, so you would get lots of these messages 
when working with a new Miktex install. 
I dont know why bitmapped fonts are being used though, since usetex 
specifically asks for type1 fonts. Maybe you need to go into your miktex 
package manager and make sure you have type1cm available on your system.
Darren
On Saturday 28 October 2006 12:25 am, Ryan Krauss wrote:
> I am trying to use latex with matplolib on Windows XP with MikTex 2.5
> and dvipng 1.8.
>
> I am getting this warning:
> dvipng warning: font cmsy10 at 512 dpi not found, characters will be left
> blank
>
> when I do this:
>
> t=arange(0,1,0.01)
> y=sin(2*pi*t)
> plot(t,y)
>
> and this
> ylabel('$y(t)$')
>
> leads to
> dvipng warning: font cmr17 at 356 dpi not found, characters will be left
> blank
>
> with no ylabel.
>
> I am also getting a lot of other output (see below).
>
> How do I fix this?
>
> Thanks,
>
> Ryan
>
> ==============================================
> Trying to make PK font cmmi12 at 512 DPI...
> Creating cmmi12.pk...
> Running mf...
> This is METAFONT, Version 2.71828 (MiKTeX 2.5)
>
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmmi12.mf
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmbase.mf)
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\mathit.mf
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanu.mf [65] [66]
> [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81]
> [82] [83] [84] [85] [86] [87] [88] [89] [90])
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\itall.mf [97] [98] [99]
> [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111]
> [112] [113] [114] [115] [116] [117] [118] [119] [120] [121] [122])
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greeku.mf [0] [1] [2]
> [3] [4] [5] [6] [7] [8] [9] [10])
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greekl.mf [11] [12]
> [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27]
> [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39])
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\italms.mf [64] [96]
> [123] [124] [125] [126] [127])
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\olddig.mf [48] [49]
> [50] [51] [52] [53] [54] [55] [56] [57])
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romms.mf [40] [41] [42]
> [43] [44] [45] [46] [47] [58] [59] [60] [61] [62] [63] [91] [92] [93] [94]
> [95]) ) )
> Font metrics written on cmmi12.tfm.
> Output written on cmmi12.512gf (128 characters, 25960 bytes).
> Transcript written on cmmi12.log.
> METAFONT failed for some reason
> ignoring 0 strange path(s)
> Running gftopk...
> This is GFtoPK, Version 2.3 (MiKTeX 2.5)
> 'METAFONT output 2006年10月27日:2322'
> 25960 bytes packed to 14164 bytes.
> dvipng warning: font cmmi12 at 512 dpi not found, characters will be left
> blan Trying to make PK font cmr17 at 356 DPI...
> Creating cmr17.pk...
> Running mf...
> This is METAFONT, Version 2.71828 (MiKTeX 2.5)
>
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmr17.mf
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmbase.mf)
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\roman.mf
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanu.mf [65] [66]
> [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81]
> [82] [83] [84] [85] [86] [87] [88] [89] [90])
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanl.mf [97] [98]
> [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110]
> [111] [112] [113] [114] [115] [116] [117] [118] [119] [120] [121] [122])
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greeku.mf [0] [1] [2]
> [3] [4] [5] [6] [7] [8] [9] [10])
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romand.mf [48] [49]
> [50] [51] [52] [53] [54] [55] [56] [57])
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanp.mf [36] [38]
> [63] [62]) (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romspl.mf
> [16] [17] [25] [26] [27] [28])
> (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romspu.mf [29] [30]
> [31]) (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\punct.mf [33]
> [60] [35] [37] [39] [40] [41] [42] [43] [44] [46] [47] [58] [59] [61] [64]
> [91] [93] [96]) (C:\Program Files\MiKTeX
> 2.5\fonts\source\public\cm\accent.mf [18] [19] [20] [21] [22] [23] [24]
> [32] [94] [95] [125] [126] [127]) (C:\Program Files\MiKTeX
> 2.5\fonts\source\public\cm\romlig.mf [11] [12] [13] [14] [15]) (C:\Program
> Files\MiKTeX 2.5\fonts\source\public\cm\comlig.mf [34] [45] [92] [123]
> [124]) ) )
> Font metrics written on cmr17.tfm.
> Output written on cmr17.356gf (128 characters, 24612 bytes).
> Transcript written on cmr17.log.
> METAFONT failed for some reason
> ignoring 0 strange path(s)
> Running gftopk...
> This is GFtoPK, Version 2.3 (MiKTeX 2.5)
> 'METAFONT output 2006年10月27日:2322'
> 24612 bytes packed to 11084 bytes.
> dvipng warning: font cmr17 at 356 dpi not found, characters will be left
> blank
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job
> easier Download IBM WebSphere Application Server v.1.0.1 based on Apache
> Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
-- 
Darren S. Dale, Ph.D.
dd...@co...

On 28/10/06, Eric Firing <ef...@ha...> wrote:
>
> The error is not in matplotlib, it is in a pygtk header file, which is
> presumably coming from a Redhat package. I don't know why it is
 True.
> happening or what the best solution is, though. It looks like the
> pygobject.h that I have has "typename_", not "typename", so I am
> thinking maybe this is a pygtk bug, since fixed, in which "typename" is
> in conflict with something.
 C++, typename is a reserved word there.
 You might try editing your copy of
> pygobject.h, adding the trailing underscore to "typename" on both lines
> where it occurs. It can't hurt; these are just dummy variable names.
 That is the solution, and you say the same that was adopted by the
project to solve the problem.
> Eric
--=20
Jos=E9 Ab=EDlio

I am trying to use latex with matplolib on Windows XP with MikTex 2.5
and dvipng 1.8.
I am getting this warning:
dvipng warning: font cmsy10 at 512 dpi not found, characters will be left blank
when I do this:
t=arange(0,1,0.01)
y=sin(2*pi*t)
plot(t,y)
and this
ylabel('$y(t)$')
leads to
dvipng warning: font cmr17 at 356 dpi not found, characters will be left blank
with no ylabel.
I am also getting a lot of other output (see below).
How do I fix this?
Thanks,
Ryan
==============================================
Trying to make PK font cmmi12 at 512 DPI...
Creating cmmi12.pk...
Running mf...
This is METAFONT, Version 2.71828 (MiKTeX 2.5)
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmmi12.mf
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmbase.mf)
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\mathit.mf
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanu.mf [65] [66]
[67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81]
[82] [83] [84] [85] [86] [87] [88] [89] [90])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\itall.mf [97] [98] [99]
[100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111]
[112] [113] [114] [115] [116] [117] [118] [119] [120] [121] [122])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greeku.mf [0] [1] [2]
[3] [4] [5] [6] [7] [8] [9] [10])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greekl.mf [11] [12]
[13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27]
[28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\italms.mf [64] [96]
[123] [124] [125] [126] [127])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\olddig.mf [48] [49]
[50] [51] [52] [53] [54] [55] [56] [57])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romms.mf [40] [41] [42]
[43] [44] [45] [46] [47] [58] [59] [60] [61] [62] [63] [91] [92] [93] [94]
[95]) ) )
Font metrics written on cmmi12.tfm.
Output written on cmmi12.512gf (128 characters, 25960 bytes).
Transcript written on cmmi12.log.
METAFONT failed for some reason
ignoring 0 strange path(s)
Running gftopk...
This is GFtoPK, Version 2.3 (MiKTeX 2.5)
'METAFONT output 2006年10月27日:2322'
25960 bytes packed to 14164 bytes.
dvipng warning: font cmmi12 at 512 dpi not found, characters will be left blan
Trying to make PK font cmr17 at 356 DPI...
Creating cmr17.pk...
Running mf...
This is METAFONT, Version 2.71828 (MiKTeX 2.5)
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmr17.mf
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\cmbase.mf)
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\roman.mf
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanu.mf [65] [66]
[67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81]
[82] [83] [84] [85] [86] [87] [88] [89] [90])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanl.mf [97] [98]
[99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111]
[112] [113] [114] [115] [116] [117] [118] [119] [120] [121] [122])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\greeku.mf [0] [1] [2]
[3] [4] [5] [6] [7] [8] [9] [10])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romand.mf [48] [49]
[50] [51] [52] [53] [54] [55] [56] [57])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romanp.mf [36] [38]
[63] [62]) (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romspl.mf
[16] [17] [25] [26] [27] [28])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romspu.mf [29] [30]
[31]) (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\punct.mf [33]
[60] [35] [37] [39] [40] [41] [42] [43] [44] [46] [47] [58] [59] [61] [64]
[91] [93] [96]) (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\accent.mf
[18] [19] [20] [21] [22] [23] [24] [32] [94] [95] [125] [126] [127])
(C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\romlig.mf [11] [12]
[13] [14] [15]) (C:\Program Files\MiKTeX 2.5\fonts\source\public\cm\comlig.mf
[34] [45] [92] [123] [124]) ) )
Font metrics written on cmr17.tfm.
Output written on cmr17.356gf (128 characters, 24612 bytes).
Transcript written on cmr17.log.
METAFONT failed for some reason
ignoring 0 strange path(s)
Running gftopk...
This is GFtoPK, Version 2.3 (MiKTeX 2.5)
'METAFONT output 2006年10月27日:2322'
24612 bytes packed to 11084 bytes.
dvipng warning: font cmr17 at 356 dpi not found, characters will be left blank

Victor Danilchenko wrote:
> 	Hi all,
> 
> 	I am trying to build Matplotlib 0.87.7 on an RHEL 4.4-derived system, 
> with Python 2.3.4. I am running into a weird problem. When I do "python 
> setup.py build", I get the following error:
> 
> gcc: src/_gtkagg.cpp
> In file included from /usr/include/python2.3/Python.h:8,
> from /usr/include/pygtk-2.0/pygobject.h:5,
> from src/_gtkagg.cpp:11:
> /usr/include/python2.3/pyconfig.h:850:1: warning: "_POSIX_C_SOURCE" 
> redefined
> In file included from /usr/include/string.h:26,
> from 
> /usr/lib/gcc/i386-redhat-linux/3.4.6/../../../../include/c++/3.4.6/cstring:51,
> from src/_gtkagg.cpp:2:
> /usr/include/features.h:150:1: warning: this is the location of the 
> previous definition
> In file included from src/_gtkagg.cpp:11:
> /usr/include/pygtk-2.0/pygobject.h:140: error: expected `,' or `...' 
> before "typename"
> /usr/include/pygtk-2.0/pygobject.h:147: error: expected `,' or `...' 
> before "typename"
[...]
The error is not in matplotlib, it is in a pygtk header file, which is 
presumably coming from a Redhat package. I don't know why it is 
happening or what the best solution is, though. It looks like the 
pygobject.h that I have has "typename_", not "typename", so I am 
thinking maybe this is a pygtk bug, since fixed, in which "typename" is 
in conflict with something. You might try editing your copy of 
pygobject.h, adding the trailing underscore to "typename" on both lines 
where it occurs. It can't hurt; these are just dummy variable names.
Eric

Thanks for the support.
I uninstalled numarray, and then compiled/installed most recent numpy and 
matplotlib from the tarballs.
And I am happy to announce I can now perform the mundane task of plotting 
barcharts and histograms. :)
Thanks again everyone,
Suresh
On 2006年10月26日, Eric Firing wrote:
> Suresh,
>
> This is baffling, and particularly annoying for someone just starting out 
> with mpl. I don't think there is any problem with the barchart code--the 
> demo works for me with numarray--and I don't see anything obvious in the 
> traceback. Are your matplotlib and numarray both from Mandriva packages?
>
> I would be inclined to download numpy and matplotlib as nice, fresh tarballs, 
> uninstall any previous versions (including numarray), build and install 
> numpy, and then do the same for matplotlib. If you have not done this before 
> you will probably trip over some missing -devel packages (furnishing header 
> files for python, gtk, etc), but once all that is straightened out it should 
> go smoothly.
>
> Eric

	Hi all,
	I am trying to build Matplotlib 0.87.7 on an RHEL 4.4-derived system, 
with Python 2.3.4. I am running into a weird problem. When I do "python 
setup.py build", I get the following error:
gcc: src/_gtkagg.cpp
In file included from /usr/include/python2.3/Python.h:8,
 from /usr/include/pygtk-2.0/pygobject.h:5,
 from src/_gtkagg.cpp:11:
/usr/include/python2.3/pyconfig.h:850:1: warning: "_POSIX_C_SOURCE" 
redefined
In file included from /usr/include/string.h:26,
 from 
/usr/lib/gcc/i386-redhat-linux/3.4.6/../../../../include/c++/3.4.6/cstring:51,
 from src/_gtkagg.cpp:2:
/usr/include/features.h:150:1: warning: this is the location of the 
previous definition
In file included from src/_gtkagg.cpp:11:
/usr/include/pygtk-2.0/pygobject.h:140: error: expected `,' or `...' 
before "typename"
/usr/include/pygtk-2.0/pygobject.h:147: error: expected `,' or `...' 
before "typename"
In file included from /usr/include/python2.3/Python.h:8,
 from /usr/include/pygtk-2.0/pygobject.h:5,
 from src/_gtkagg.cpp:11:
/usr/include/python2.3/pyconfig.h:850:1: warning: "_POSIX_C_SOURCE" 
redefined
In file included from /usr/include/string.h:26,
 from 
/usr/lib/gcc/i386-redhat-linux/3.4.6/../../../../include/c++/3.4.6/cstring:51,
 from src/_gtkagg.cpp:2:
/usr/include/features.h:150:1: warning: this is the location of the 
previous definition
In file included from src/_gtkagg.cpp:11:
/usr/include/pygtk-2.0/pygobject.h:140: error: expected `,' or `...' 
before "typename"
/usr/include/pygtk-2.0/pygobject.h:147: error: expected `,' or `...' 
before "typename"
	So I did a google search, and found someone recommending that "#include 
<Python.h>" be prepended to the offending file. I have prepended it to 
src/_gtkagg.cpp, did a clean, and run the build again. The error 
changed, and the "_POSIX_C_SOURCE" problem went away, but the build is 
still failing, at that same file:
gcc: src/_gtkagg.cpp
In file included from src/_gtkagg.cpp:11:
/usr/include/pygtk-2.0/pygobject.h:140: error: expected `,' or `...' 
before "typename"
/usr/include/pygtk-2.0/pygobject.h:147: error: expected `,' or `...' 
before "typename"
In file included from src/_gtkagg.cpp:11:
/usr/include/pygtk-2.0/pygobject.h:140: error: expected `,' or `...' 
before "typename"
/usr/include/pygtk-2.0/pygobject.h:147: error: expected `,' or `...' 
before "typename"
	Does anyone have an idea of what is going wrong here? I could really 
use some help, as I am totally stumped.
	Many thanks in advance.

Yes, both are mandriva packages:
python-matplotlib-0.87.4-2mdv2007.0
python-numarray-1.5.2-1mdv2007.0
Given there isnt a python-numpy package for x64-64 right now (there is a 
devel one, so not sure why no main package), I will go ahead and try to 
build from scratch.
Thanks,
Suresh
On 2006年10月26日, Eric Firing wrote:
> Suresh,
>
> This is baffling, and particularly annoying for someone just starting out 
> with mpl. I don't think there is any problem with the barchart code--the 
> demo works for me with numarray--and I don't see anything obvious in the 
> traceback. Are your matplotlib and numarray both from Mandriva packages?
>
> I would be inclined to download numpy and matplotlib as nice, fresh tarballs, 
> uninstall any previous versions (including numarray), build and install 
> numpy, and then do the same for matplotlib. If you have not done this before 
> you will probably trip over some missing -devel packages (furnishing header 
> files for python, gtk, etc), but once all that is straightened out it should 
> go smoothly.
>
> Eric
>

On Fri, 2006年10月27日 at 11:58 +0200, Trond Danielsen wrote:
> 2006年10月27日, Robert Elsner <ml...@re...>:
> > Hello Everybody,
> >
> > I use the usual scipy + numpy + matplotlib combination for my daily
> > work. Unfortunately after switching to Fedora Core 6 (from FC5) I am
> > unable to compile matplotlib 0.87.6 on my system.
> >
> 
> I am sure you allready know this, but matplotlib 0.87.6 is available
> from fedora extra.
> 
> Best regards,
> Trond
> 
Yes I noticed that 0.87.6 is available from fedora-extras for FC6. But
unfortunately this one is unable to work with numpy 1.0 final. But
thanks nonetheless ;).
And as Chuck pointed out - he managed to compile from svn. I tried the
new 0.87.7 and that one finally compiled.
I ran some quick tests and it seems to work well.
So thanks for your help @Chuck,Trond.
Robert