[Veusz-discuss] ANN: Veusz - change of website and soon closure of mailing list and other Gna services

[Jeremy Sanders]

Dear all

gna.org have announced that the website is closing: 
https://mail.gna.org/public/project/2016-11/msg1.html  - there's no 
final date listed, but I've had to take action as I only noticed last week.


I have migrated the website to:
  https://veusz.github.io/
I've changed the website quite a bit. Please feel free to send me pull 
requests for improvements.


The old website was listed in lots of places and will shortly break. 
PLEASE - if you have a link to veusz somewhere on the web, update it 
ASAP to the new address. Any unix package maintainers should update the 
website in the packaging.


I have also changed to using github's release service to enable binary 
downloads and the download page now points to there.


We will shortly lose the mailing list. I get the feeling that people 
don't find it very useful. Therefore we will migrate to the veusz 
subreddit, https://www.reddit.com/r/veusz/, unless someone wants to step 
up and start a new mailing list.


If you want to get veusz release announcements, you can use the 
facilities on github, or subscript to the news page on the new veusz 
homepage: https://veusz.github.io/news/atom.xml


The bug tracker on gna will also be going. If there's a bug on the old 
tracker that you care about, please enter a new one on the github tracker.


Best wishes

Jeremy



___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Transparency on images with nan values and log scaling

[Jeremy Sanders]

Hi Mark

On 04/05/17 04:22, Mark Durre wrote:

I’m displaying images with nan values; normally these show as
transparent and display the background color of the graph. However, when
I change the scaling to log the blank fill become the lowest value color
on the colormap, which could be anything, and is also problematic when
the lowest value on the scale is relevant (causing confusion). My
solution has been to create another dataset with log values and display
that with linear scaling, while also having the original image on the
same graph but hidden, then having a colorbar linked to the old image.
Complicated I know! i.e. on the one graph


I've hopefully fixed this in this change. I'll put it in the new release 
shortly.


https://github.com/jeremysanders/veusz/commit/5cdde90367051ae74d3965c56efad4a18638928d

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] Veusz 1.25.99 beta

[Jeremy Sanders]

Please try out the Veusz 1.26 beta (1.25.99) here:

http://download.gna.org/veusz/testing/

This includes support for colour themes for automatically colouring 
widgets. Select the theme under the document widget settings. You can 
override colours in the Custom Definition dialog where new colours can 
be defined. The theme colours are "theme1", "theme2" etc...


Old documents are loaded with a default "black" theme which looks the 
same as old Veusz versions. New documents get a standard theme (settable 
in the preferences dialog box).


Other changes:

 * Add color themes and automatic coloring of widgets
 * Custom color definitions
 * Improved custom definition dialog box
 * Fix loading translator via command line
 * Show horizontal error lines/fills in key
 * Use iMinuit if available for fitting
 * Fix for dataset plugins returning nD datasets
 * Add editable automatic axis range, including support for shrinking 
range (-X%), percentage ranges (X-Y%) and limits (Y%)


Please report any bugs, etc, ASAP.

Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] ANN: Veusz subreddit for discussion and questions

[Jeremy Sanders]

Dear Veusz users

I've created a forum/subreddit for Veusz on the reddit website. This is 
an experiment to see whether a web forum would be better than the 
mailing list to ask questions and discuss the software.


If you'd like to participate, please go to:

https://www.reddit.com/r/Veusz/

Best wishes

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Creating dataset from expression with numpy command

[Jeremy Sanders]

On 08/12/16 12:06, Ilya Kostanovskiy wrote:


I am trying to create a new dataset out of existing 2D dataset using
standard numpy command. I would like to sum along row and column. From
the control line the command

sum([[1,2],[2,3]], axis=0) or sum([[1,2],[2,3]], axis=1)

works and gives normal output. If I use it as an expression on the menu
for creating a new dataset then it fails:

sum(/2d_datasetname/_data,axis=0).

It only works if I remove axis property and sums along the rows.

Does someone know the trick how to use this numpy expression properly?


I think this is because the environment functions are executed in is 
different from the command line. The sum function seems to be the 
standard python sum, not the numpy function.


As a workaround, you can add a custom import for sum from numpy in the 
Custom Definitions dialog box. This seems to work ok for me (though I 
haven't tried the _data subscript.


I'll try to remember to fix this, but file a bug to remind me if you 
have time.


Cheers

Jeremy



___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] ANN: Veusz 1.25

[Jeremy Sanders]

I'm pleased to announce Veusz 1.25. Please see the release notes below.

Jeremy
Veusz 1.25
--
http://home.gna.org/veusz/

Veusz is a scientific plotting package.  It is designed to produce
publication-ready Postscript, PDF or SVG output. Graphs are built-up
by combining plotting widgets. The user interface aims to be simple,
consistent and powerful.

Veusz provides GUI, Python module, command line, scripting, DBUS and
SAMP interfaces to its plotting facilities. It also allows for
manipulation and editing of datasets. Data can be captured from
external sources such as Internet sockets or other programs.

Changes in 1.25:
 * Add n-dimensional dataset support
 * Add GetColormap command to return RGBA values
 * Add Colormap sequence plugin for choosing colors of widgets
 * Fill above/below setting shown in key for xy widget
 * Add Covariance widget for plotting covariance ellipses
 * Add lineup, linedown, lineleft and lineright markers
 * Add lineup, linedown and lineextend arrows

Bug fixes:
 * Fix pick indices if non-finite values in data
 * Fix drag and drop of files into Veusz on Windows
 * Raise exception if functions in console request input, avoiding hang
 * Initialise unsafe_mode, in case Veusz used in PyQt embedding
 * Fix crash if setting default setting in default setting window
 * Fix crash in Clone() command on python3

Features of package:
 Plotting features:
  * X-Y plots (with errorbars)
  * Line and function plots
  * Contour plots
  * Images (with colour mappings and colorbars)
  * Stepped plots (for histograms)
  * Bar graphs
  * Vector field plots
  * Box plots
  * Polar plots
  * Ternary plots
  * Plotting dates
  * Fitting functions to data
  * Stacked plots and arrays of plots
  * Nested plots
  * Plot keys
  * Plot labels
  * Shapes and arrows on plots
  * LaTeX-like formatting for text
  * Multiple axes
  * Axes with steps in axis scale (broken axes)
  * Axis scales using functional forms
  * Plotting functions of datasets
 Input and output:
  * EPS/PDF/PNG/SVG/EMF export
  * Dataset creation/manipulation
  * Embed Veusz within other programs
  * Text, HDF5, CSV, FITS, NPY/NPZ, QDP, binary and user-plugin importing
  * Data can be captured from external sources
 Extending:
  * Use as a Python module
  * User defined functions, constants and can import external Python functions
  * Plugin interface to allow user to write or load code to
 - import data using new formats
 - make new datasets, optionally linked to existing datasets
 - arbitrarily manipulate the document
  * Scripting interface
  * Control with DBUS and SAMP
 Other features:
  * Data filtering and manipulation
  * Data picker
  * Interactive tutorial
  * Multithreaded rendering

Requirements for source install:
 Python 2.x (2.6 or greater required) or 3.x (3.3 or greater required)
   http://www.python.org/
 Qt >= 4.6 (free edition)
   http://www.trolltech.com/products/qt/
 PyQt >= 4.5 (SIP is required to be installed first)
   http://www.riverbankcomputing.co.uk/software/pyqt/
   http://www.riverbankcomputing.co.uk/software/sip/
 numpy >= 1.0
   http://numpy.scipy.org/

Optional requirements:
 h5py (optional for HDF5 support)
   http://www.h5py.org/
 astropy >= 0.2 or PyFITS >= 1.1 (optional for FITS import)
   http://www.stsci.edu/resources/software_hardware/pyfits
   http://www.astropy.org/
 pyemf >= 2.0.0 (optional for EMF export)
   http://pyemf.sourceforge.net/
 PyMinuit >= 1.1.2 (optional improved fitting)
   http://code.google.com/p/pyminuit/
 dbus-python, for dbus interface
   http://dbus.freedesktop.org/doc/dbus-python/
 astropy (optional for VO table import)
   http://www.astropy.org/
 SAMPy or astropy >= 0.4 (optional for SAMP support)
   http://pypi.python.org/pypi/sampy/

Veusz is Copyright (C) 2003-2015 Jeremy Sanders <jer...@jeremysanders.net>
 and contributors.
It is licensed under the GPL (version 2 or greater).

For documentation on using Veusz, see the "Documents" directory. The
manual is in PDF, HTML and text format (generated from docbook). The
examples are also useful documentation. Please also see and contribute
to the Veusz wiki: https://github.com/jeremysanders/veusz/wiki

If you enjoy using Veusz, we would love to hear from you. Please join
the mailing lists at

https://gna.org/mail/?group=veusz

to discuss new features or if you'd like to contribute code. The
latest code can always be found in the Git repository
at https://github.com/jeremysanders/veusz.git.
___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Changing default color list?

[Jeremy Sanders]

On 17/11/16 08:53, Jeremy Sanders wrote:


import veusz.setting
veusz.setting.Color._colors = []


This also requires the line at the end:

veusz.setting.controls.Color._colors = veusz.setting.Color._colors

This appears to work for me.

Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Changing default color list?

[Jeremy Sanders]

Hi Josh

On 16/11/16 19:36, Josh Taillon wrote:


Is there any way to easily change the default color list in Veusz? I can
see that the list is defined in setting.py
(https://github.com/jeremysanders/veusz/blob/master/veusz/setting/setting.py#L1374),
but I'm not sure if there's any convenient way to override this.


I'm planning to add some colour theme capability some time in the future.

At the moment, the only way which might change this would be to add a 
plugin to hack the values, e.g.


import veusz.setting
veusz.setting.Color._colors = []

(this can go in a python text file and can be loaded by adding it in the 
setting dialog box).


I haven't tested this idea, though, and it'll break if the code changes.

Best wishes

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Splitting data

[Jeremy Sanders]

On 14/11/16 20:03, André Luiz Regis Monteiro wrote:


If I use "fit function" in this "new vector" (( V[150:200], V[300:350],
V[700:800] )), are all that points considered as a single vector?


Yes - it should.

Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Splitting data

[Jeremy Sanders]

Hi André

You can either:

1. Add multiple xy plotters for each range (you can change the settings 
of them simultaneously by selecting them all). Copy and paste is helpful.


2. Plot expressions like

hstack(( V[150:200], V[300:350], V[700:800] ))

hstack is a numpy function which joins arrays.

3. Create a new dataset from an expression (Data->Create Dataset), and 
use an expression like above. Plot this dataset.


Hope this helps

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] I not find a Veusz compatible with Ubuntu 16.10 ...

[Jeremy Sanders]

I've now added a yakkety release. I hope it works ok!

Jeremy

On 08/11/16 08:48, Adriano Bassignana wrote:

My Ubuntu version is: yakkety

Now (2016-11-08T08:48:00) in the PPA list is not present this version:

https://launchpad.net/~jeremysanders/+archive/ubuntu/ppa

veusz   1.24-0~ppa1~xenial1 Jeremy Sanders (2016-05-05)
veusz   1.24-0~ppa1~wily1   Jeremy Sanders (2016-05-05)
veusz   1.24-0~ppa1~trusty1 Jeremy Sanders (2016-05-05)
veusz   1.24-0~ppa1~precise1Jeremy Sanders (2016-05-05)

I think is necessary recompilate the code for 16.10 :(

Thanks for all :)

Adriano

In data martedì 8 novembre 2016 07:37:39 CET, GONZALO RODRIGUEZ PRIETO ha
scritto:

Hello Adriano,

I do not understand. The PPA is perfectly maintained for the Ubuntu
16.10 by Jeremy himself (thank a lot for all the effort, by the way), as
per today. I just have checked now. See in this link how to add the PPA
to your computer list, and then you should be able to install it
immediately:

https://launchpad.net/~jeremysanders/+archive/ubuntu/ppaCheers,
Gonzalo

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss





___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Possibel bug with Latex interpreter.

[Jeremy Sanders]

Hi Fred

I see - I think that idea has come up before. The main problem is that 
Veusz's output is created by the Qt engine, so it's not in the right 
format to be recognised by LaTeX. Maybe it would be possible to write a 
Qt graphics engine which could be used by LaTeX, but I'm not sure how 
hard that would be.


Best wishes

Jeremy

On 24/10/16 16:13, Fred Labrosse wrote:

Hi Jeremy,

I wasn't suggesting to actually use LaTeX as part of veusz, more to export
files that LaTeX would process.  inkscape can export a LaTeX file that is
processed by LaTeX containing all the text in the right position and also
includes a pdf (saved at the same time without the text) also put in the
right place.

The drawback is that the text in the GUI is LaTeX stuff which often uses a
lot more space than the final rendered text, making positioning an
iterative process.  This also means that the GUI never shows the final
product, but that is very much in line with the LaTeX philosophy anyway!
;-)

It was an idea.  So far I have not been limited by the capability of veusz
when it comes to LaTeX rendering.

Cheers,

Fred


On Monday 24 October 2016 16:05:45 Jeremy Sanders wrote:

Dear Fred

I have not used LaTeX as it is a massive dependency to include. I also
could not see an easy way of rendering its output in the program, unless
I write or include a PDF or DVI interpreter.

If anyone wants to work on this and has a good idea how to achieve it,
please let me know.

By the way, Veusz includes MathML support, so complex equations are
supported if they are converted from latex to mathml.

Cheers

Jeremy

On 24/10/16 12:04, Fred Labrosse wrote:

Jeremy,

Just had a thought, which maybe has crossed many minds before,
including
yours!  Why not actually using LaTeX to do the interpretation?  This
would obviously only work for some of the exports (using perhaps a
mechanism similar to that of inkscape, currently buggy by the way).

Just my 2p.

Fred

On Friday 21 October 2016 21:18:00 Jeremy Sanders wrote:

Dear Gonzalo

On 21/10/16 13:37, GONZALO RODRIGUEZ PRIETO wrote:

I am using this excellent program for a long time, and doing a graph
I
have found this behaviour, that I am not sure to qualify as bug. Here
it
goes: When in the label of a graph I write Latex mathematical
commands,
they are interpreted correctly, but for the commands "\left" or
"\right", which left unchanged. Am I doing something wrong? (which is
probable) or did I discovered a feature/bug by change? To clarify my
problem, I attach an empty Veusz graph with problem in the Y axis.
Notice that the rest of the code is correctly interpreted, only the
command "\right" is left.


I'm afraid the problem is that the latex interpreter only supports a
very limited subset of latex. \left and \right aren't supported by the
interpreter.

Best wishes

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss



Un o’r 4 prifysgol uchaf yn y DU a’r orau yng Nghymru am fodlonrwydd
myfyrwyr. (Arolwg Cenedlaethol y Myfyrwyr 2016)
www.aber.ac.uk

Top 4 UK university and best in Wales for student satisfaction
(National Student Survey 2016)
www.aber.ac.uk

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss




Un o’r 4 prifysgol uchaf yn y DU a’r orau yng Nghymru am fodlonrwydd myfyrwyr.
(Arolwg Cenedlaethol y Myfyrwyr 2016)
www.aber.ac.uk

Top 4 UK university and best in Wales for student satisfaction
(National Student Survey 2016)
www.aber.ac.uk

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Possibel bug with Latex interpreter.

[Jeremy Sanders]

Dear Fred

I have not used LaTeX as it is a massive dependency to include. I also 
could not see an easy way of rendering its output in the program, unless 
I write or include a PDF or DVI interpreter.


If anyone wants to work on this and has a good idea how to achieve it, 
please let me know.


By the way, Veusz includes MathML support, so complex equations are 
supported if they are converted from latex to mathml.


Cheers

Jeremy

On 24/10/16 12:04, Fred Labrosse wrote:

Jeremy,

Just had a thought, which maybe has crossed many minds before, including
yours!  Why not actually using LaTeX to do the interpretation?  This would
obviously only work for some of the exports (using perhaps a mechanism
similar to that of inkscape, currently buggy by the way).

Just my 2p.

Fred

On Friday 21 October 2016 21:18:00 Jeremy Sanders wrote:

Dear Gonzalo

On 21/10/16 13:37, GONZALO RODRIGUEZ PRIETO wrote:

I am using this excellent program for a long time, and doing a graph I
have found this behaviour, that I am not sure to qualify as bug. Here
it
goes: When in the label of a graph I write Latex mathematical commands,
they are interpreted correctly, but for the commands "\left" or
"\right", which left unchanged. Am I doing something wrong? (which is
probable) or did I discovered a feature/bug by change? To clarify my
problem, I attach an empty Veusz graph with problem in the Y axis.
Notice that the rest of the code is correctly interpreted, only the
command "\right" is left.


I'm afraid the problem is that the latex interpreter only supports a
very limited subset of latex. \left and \right aren't supported by the
interpreter.

Best wishes

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss




Un o’r 4 prifysgol uchaf yn y DU a’r orau yng Nghymru am fodlonrwydd myfyrwyr.
(Arolwg Cenedlaethol y Myfyrwyr 2016)
www.aber.ac.uk

Top 4 UK university and best in Wales for student satisfaction
(National Student Survey 2016)
www.aber.ac.uk

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Possibel bug with Latex interpreter.

[Jeremy Sanders]

Dear Gonzalo

On 21/10/16 13:37, GONZALO RODRIGUEZ PRIETO wrote:


I am using this excellent program for a long time, and doing a graph I
have found this behaviour, that I am not sure to qualify as bug. Here it
goes: When in the label of a graph I write Latex mathematical commands,
they are interpreted correctly, but for the commands "\left" or
"\right", which left unchanged. Am I doing something wrong? (which is
probable) or did I discovered a feature/bug by change? To clarify my
problem, I attach an empty Veusz graph with problem in the Y axis.
Notice that the rest of the code is correctly interpreted, only the
command "\right" is left.


I'm afraid the problem is that the latex interpreter only supports a 
very limited subset of latex. \left and \right aren't supported by the 
interpreter.


Best wishes

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Annotating of spectroscopy data

[Jeremy Sanders]

Dear Ilya

On 17/10/16 10:24, Ilya Kostanovskiy wrote:


Do you think it is a good option to consider to hide axes related shapes
(labels, lines etc) which are not in the selected axes range? In my
understanding if one uses an axes related positioning, then there is no
need for a labels or forms outside of the selected plotting range?


Does the "clip" formatting option help? This should show only features 
within the container graph.


Cheers

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Doubt about axis-scalling

[Jeremy Sanders]

On 09/10/16 00:16, André Luiz Regis Monteiro wrote:


My original scale starts in -30e-9 to 30e-9 (seconds).

I changed  -30 to 30 using 1e9 scale in Formating Veusz. But I really
like to start in zero to 60 (-30+30 to 30+30). Unfortunately, I don´t
know how to do it.


Unfortunately there's no easy way to add an offset to the data. It would 
be a great thing to add, though. Options are:


1. Add 30e-9 to all the things you plot. You can create new datasets 
with the constant added (use Data->Operations->Add constant) or plot the 
dataset +30e-9 (this is harder if you have error bars).


2. Add an "axis-function" widget, linked to the other axis you want to 
change. Set the linked property, choose the linked axis and set the 
function to be 't*1e9+30' (don't use scale). Set the direction to be the 
same as the original axis. Hide the original axis.


Hope this helps

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Control export resolution from command line

[Jeremy Sanders]

Warrick

On linux you can use xvfb-run to run the window so you don't see it.

xvfb-run -a --server-args "-screen 0 640x480x24" thecommand

Alternatively, you could use the embedding interface from python

import veusz.embed as veusz

e = veusz.Embedded(hidden=True)
e.Load('in.vsz')
e.Export('out.png', dpi=300)

Cheers

Jeremy


On 29/09/16 10:55, Warrick Ball wrote:

Jeremy,

Thanks!  That's good enough for now.  Is there any way to suppress the
window that pops up?  I tried adding the --quiet flag to no avail.

Cheers,
Warrick



Warrick Ball
Postdoc, Institut für Astrophysik Göttingen
wb...@astro.physik.uni-goettingen.de
+49 (0) 551 39 5069

On Wed, 28 Sep 2016, Jeremy Sanders wrote:


Dear Warrick

On 28/09/16 09:40, Warrick Ball wrote:


I'm finding that the PNG exported this way is much lower resolution (100
DPI, I think) whereas I've tried to set the default to 300.  i.e.  If I
open a new Veusz window and hit File/Export, the DPI in the dialog box
is already set to 300 DPI.

Can I demand 300 DPI from the command line?


It would be a good idea to fix this to use the default or last-used
settings, or add options to the command line. It would be handy to
have a bug to keep track of this.

I think a workaround is to add some text to the vsz file such as

Export('test.png', dpi=300)
import sys
sys.exit()

And do

veusz --unsafe-mode input.vsz


e.g.

cp input.vsz temp.vsz
cat >> temp.vsz <


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Fwd: Some symbol and filling the curves

[Jeremy Sanders]

Dear Sabya

On 15/09/16 20:43, Sabya Sachi Chatterjee wrote:


 I sent a request few days back and I have faced
this problem many times  in physics that is filling the curve between
two different curves. can you please tell me if there is solution to
this in veusz.


There are currently two options:

1. Use the function widget to fill under one function. Use the function 
widget to fill under the second using the background colour (white).


This has disadvantages: if the functions cross this won't work, it will 
plot over other widgets.


2. Create a new range dataset over the range you want plotting. 
Data->Create Dataset->Value or Range: enter a number of steps, then 
enter the range as e.g. 1:10.


Add an XY plotter and plot something like

0.5*(func1(xds)+func2(xds),0.5*(func1(xds)-func2(xds))

Where func1 and func2 are your functions of the created dataset (xds 
here), e.g.


0.5*(xds**2+xds**3),0.5*(xds**2-xds**3)

Switch the xy plotter error style to fillvert. Turn of the plotting 
symbols and line, if required. This draws the filled region as error bars.



Otherwise, please extend the function plotter to plot between functions. 
I welcome contributions.


Best wishes

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Connector line requested

[Jeremy Sanders]

On 15/09/16 19:52, M. Ignacio Monge García wrote:

PS: could you post a test version for windows?


http://download.gna.org/veusz/testing/

It's a bit of a pain making binaries. It's a shame there's no automatic 
daily generation.


J

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] X-Y plots

[Jeremy Sanders]

Dear Martin

On 05/09/16 23:14, Martin Valencia wrote:


I am trying to make a X-Y plot for chemical concentrations of rock
samples. The y axis is the concentration in ppm and the x axis is for
the chemical elements (55 in total). I am doing the plot adding the
samples one by one, but the problem is that I have a dataset of more
than 100 samples. Is there a way to plot all them a at time, instead of
one by one? I would appreciate much your help…


In addition to the console, as Ben has already discussed, you can use 
the Tools->Widgets->Clone for datasets menu item to clone xy widgets.


Unless I have misunderstood you and you're trying to plot one data point 
per xy widget? (which of course you shouldn't).


Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Capture data sets

[Jeremy Sanders]

On 23/08/16 11:23, Simon Kleinknecht wrote:


It would be super comfortable and handy if the dialog of the capture
function would be similar to the import dialog and just adding the
required buttons and settings to define start/end etc.
Is this a huge effort to implement?


Unfortunately it's not so easy, as the standard import routines aren't 
suited to interpreting continuous data streams. They mostly work by 
collecting the entire input and then creating the datasets.


It's possible that it wouldn't be too hard to add a method to the import 
classes to optionally add streaming support. I'd have to think about 
that. Obviously it would only work for text-like formats, like CSV.


The only format the current capture dialog supports is the veusz 
standard format, which is whitespace-separated columns of data. The 
descriptor is just a list of space-separated names.


Cheers

Jeremy




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] Euroscipy talk

[Jeremy Sanders]

Dear all

I will be giving a talk at EuroSciPy 2016 about Veusz on Friday 26th August:

https://www.euroscipy.org/2016/schedule/sessions/33/

I'll put up a copy of my talk when it's done.

Cheers

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Icons disappeared from menus and Formatting tabs

[Jeremy Sanders]

Dear Kazuyoshi

On 22/07/16 04:42, Kazuyoshi Furutaka wrote:


As the subject above said, the icons of menu items as
well as Formatting tabs disappeared from veusz 1.24,
when installed by PIP or via fedora rpm (fedora 24, x86_64).

The binary in
http://download.gna.org/veusz/Binaries-Linux-i386/veusz-1.24-linux-i386.tar.gz
runs w/ the icons visible.

Any clue to fix this?


This usually happens if qt can't find the image/iconengine plugins 
responsible for rendering the svg/png icons.


These don't seem to be separately packaged on fedora, so I'm not sure 
what's going on. It looks ok with the fedora package for me.


It might be due to an invalid qt configuration file. You could try 
loading qt designer or qt linguist to see whether their icons work ok.


Hope this helps

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Qt drops PostScript output

[Jeremy Sanders]

On 17/07/16 18:12, Michael Clerx wrote:

Thanks for the heads-up! I've never used ps, but often use eps. Is eps
output affected?


Unfortunately, yes.

There are tools to covert PDF to EPS, but I don't know if they are good 
enough.


Cheers

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] Qt drops PostScript output

[Jeremy Sanders]

Hi -

Emmanuel has noticed that Qt 5 has dropped Postscript support - see 
https://github.com/jeremysanders/veusz/issues/152


If we're going to move to Qt 5 soon (Qt 4 is no longer supported), we'll 
need to use drop PostScript, use some sort of external library or 
implement our own output.


How many people are using PostScript?

Thanks

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] adding ellipses to plots

[Jeremy Sanders]

Dear Johan

On 06/15/2016 11:02 AM, Johan Asplund wrote:

I’m trying to add ellipses to a veusz figure like the one in the
attached “plot.png”. The ellipses in plot.png was created using
ordielllipse from the R package vegan.


The existing ellipse widget can do something like this. The main problem 
is getting your data into the right form. The ellipse centres can be 
specified in axes coordinates, but unfortunately the width and height 
and have to be in fractions of the plot width. The angle is in degrees. 
This is a pain to compute from your covariance matrix.


However, I think it'd be a good idea to have a direct way of plotting 
ellipses based on covariance matrices. I was thinking of doing something 
similar for my own work. I assume it's mostly a matter of computing the 
eigenvectors and plotting an ellipse/diamond/box aligned along these 
vectors.


I'll have a think about this. Maybe separate datasets for the items in 
covariance matrix, or using the datasets themselves plus the 
non-diagnonal entry.


Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Fill area under curve

[Jeremy Sanders]

On 06/08/2016 02:21 PM, Michael Clerx wrote:

Thanks for the hard work Yan, but I'd like to turn this into a feature
request!


Thanks - I'll add this soon. I'm currently restructuring the xy plotter 
code, so I'll try to incorporate it there.


I'm implementing "Transformations" which can be applied to the data when 
plotting. This avoids creating one-off linked datasets and makes it easy 
to apply to lots of datasets. They can transform both x and y (plus 
labels, colours and sizes).


One can do something like

AddX(1),MulY(2),Thin(4)

I'm also thinking about how to hand plotting multiple datasets at once 
in a single plotter.


Thanks

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Problems with contours plot

[Jeremy Sanders]

Hi

On 05/23/2016 05:13 PM, Marcelo Chiapparini wrote:


I am a newbie with Veusz. I've already installed v1.24 on a Mac running
OSX 10.11.4. I need to plot a data set as contours. The data are written
in three columns in a plain text file. When importing the data file, I
named the three columns as "x" "y" and "z".


The contour and image widgets only plot 2D datasets. You have to convert 
your three datasets to a 2D dataset. If x and y are regularly gridded, 
you can do this in Data->Create 2D dataset.


Hope this helps

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Displaying images

[Jeremy Sanders]

Hi Fred

On 05/23/2016 07:01 PM, Fred Labrosse wrote:


I'm trying to display images with a y axis going vertically down since
pixel (0,0) is at the top left.  I can invert the loading of the data for
the image to be displayed correctly, but I can't make the y axis display
the values correctly (0 at the top, positive going down).


You can invert the direction of the axis by manually entering a minimum 
which is greater than the maximum. Alternatively you can swap the 
minimum and maximum axis position settings to reverse the direction 
(0->1 to 1->0).


Cheers

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Image to pdf conversion

[Jeremy Sanders]

On 11/05/16 23:24, Michael Clerx wrote:


I've used Veusz to create some nice 2d-density plots using the "image"
widget.
However, when I export my graph to PDF the colours change considerably.
I'm using the bluegreen colormap (inverted).

Has anybody else experienced similar problems?


I don't see the problem with the mandelbrot example with bluegreen on 
linux, after exporting to PDF. There is some smoothing introduced, but 
the colours look ok.


What version and platform are you using?

Thanks

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Histogram data isn't saved

[Jeremy Sanders]

On 11/04/16 14:16, Michael Clerx wrote:

I'm not quite sure! Fedora calls it 1.23-2, but I can't see if that's
directly related to 1.23.2

Is there some way I can tell from the code? (The splash screen etc. only
state 1.23)


You're running 1.23, not 1.23.2. In Fedora/RedHat/SUSE the -suffix 
indicates the package revision of the package, not the version of the 
software.


Fedora don't encourage packagers to push recent versions of packages for 
existing releases (see their guidelines), so as packager I only push the 
latest version to the development version.


If you file a bug I can push a newer version, but I'm away for the next 
week or so.


Alternatively, you can use the binary version from the website.

Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] String constants and current date in labels

[Jeremy Sanders]

Hi Frans

On 04/03/2016 12:34 PM, Frans Ooms wrote:


I would like to display user defined constants (e.g. a string) in a
label widget.
Defining the constants is easy, creating a label on a page also, but how
to display the constant in the label? As far as I can see, the label
only accepts "Label text" directly.

It would also be interesting to be able to show the current date,
filename etc.
Are there e.g. accessible environment variables that can be used for this?


I'm afraid labels can't currently include any dynamic content, though it 
sounds a good idea.


One suggestion would be a tools plugin, which would replace the text of 
a label with the text you want,


e.g.

Root.page1.label1.val = 'my text'

Cheers

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] bar chart

[Jeremy Sanders]

On 03/25/2016 11:01 AM, Michael Weiner wrote:


It would be best if Veusz could simply plot only the non-missing values
directly.  "NaN" works but breaks the line on xy plots.  Microsoft Excel
handles this well with the "NA()" value combined with the following option.

Select data -> Hidden and Empty Cells -> Show empty cells as gaps


Veusz is designed to break lines with NaN values.

Maybe it would be a good idea to add an option to not do that.

I could envisage supporting two different types of NaN - a datagroup 
breaking value and a missing value. This would be possible as there are 
so many different NaN values allowed. The main problem is how you 
support importing these different types of NaN.


Cheers

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Move to Qt 5?

[Jeremy Sanders]

Hi Daniele

Python 2/3 compatibility isn't too difficult to keep in the codebase. 
The main problem is testing. The built-in tests mostly cover the 
rendering and internal code, but not the GUI. It's hard to remember to 
try out every bit of new GUI code on both Python 2 and 3.


Qt4 and 5 might be difficult to keep in the same codebase without 
filling it with spaghetti - that's why I have two branches. I haven't 
investigated whether a total merge is achievable.


Is there any need to keep Qt4? It's essentially unsupported, so there 
won't be any bug fixes.


With both Python 2/3 and Qt 4/5, there are four different possibilities 
for testing, which is really too much.


Cheers

Jeremy

On 21/03/16 09:18, Daniele Paganelli wrote:

Just found this interesting article about what the python scientific
community feels as "standard":
https://www.webucator.com/blog/2016/03/still-using-python-2-it-is-time-to-upgrade/

The community is large and populated by many small and tightly
inter-dependent niches, where py version is a negligible detail. Many of
us still has one or two niche packages which still forbids upgrade to 3
(in my case, I have py-twisted...).

I think Veusz did a very good job supporting both 2 and 3. I agree that
compiled packages can be moved to Py3/Qt5, if you think it is tested and
stable enough.

But, would it be worth to throw away all the work Jeremy did to support
both, by dropping support of py2/qt4, given the community seems still
divided 50/50?

Jeremy, can you quantify/exemplify the amount of burden involved in
keeping compatibility, as the qt5 branch becomes the default?


Il 17/03/2016 21:15, Christian Ambros ha scritto:

So, are there any decision yet about finally switching to standards?
PyQt5 is worth it.

cheers,
Christian
-- "A little learning never caused anyone's head to explode!" "Ein
wenig Lernen hat noch niemandens Kopf zum Explodieren gebracht!"


On Tuesday, February 23, 2016 9:44 AM, Daniele Paganelli
<ve...@mythsmith.it> wrote:


Hi all,
I am relying on current Veusz, Py2 and Qt4 for Misura package
(tainstr/misura.client).
I know that also Python(x,y) distribution is still Py2 and Qt4. I rely
on that distribution to quickly install a complete environment on Windows.

Anyway, I am not voting to keep Veusz behind! Go ahead and move to Qt5
and Py3 whenever you like: it will be a good testing phase for me.

Until I can schedule some time to upgrade, my contributions to Veusz
will probably be to the Py2/Qt4 branch.

There is no technical reason to stick with those versions: only lack of
time. It's a complex system and Py/Qt is just a detail for me.

Regards,
Daniele

Il 23/02/2016 09:19, Jeremy Sanders ha scritto:
> Hi all
>
> As an additional question, how many people are relying on Veusz Python
> 2 compatibility?
>
> There would be some advantages in dropping the Python 2 support, such
> as the reducing costs in maintenance and testing.
>
> Cheers
>
> Jeremy
>
>
> On 23/02/16 05:55, Christian Ambros wrote:
>> Hi Jeremy,
>>
>> I'm waiting for a long time for this step. And in some cases I'm using
>> pretty old laptops which do not have problems with Qt5 either. Even
that
>> old T42 Thinkpad of my parents, who where teachers runs fine. They are
>> now supporting their school as consultants during their retirement and
>> came into situations where they had to do statistical work. They are
>> running the Qt5 version you offered some month ago and are pretty happy
>> with it. This shows that there are no problems with old hardware.
>> Since 3D still isn't an option, and switching to matplotlib was denied
>> by you earlier because you didn't want to depend on such, but standard,
>> libs, there won't be any problem wit Qt5 in any direction.
>>
>> So I welcome this change and encourage you to do this step, as one in
>> the right direction since Qt5 is standard since 2012.
>> I disagree with Benjamin about depending on LTS version, because this
>> behavior doesn't cause enough pressure on the developers to be up to
>> standards when they are published. Take a look at python3 which came as
>> standard in 2010, which was six years from now and even today there are
>> plenty of developers who think that python3 has more drawback than
>> advantages which isn't true if you go through that process of updating.
>> Those people are just afraid of learning something new. They try to
>> stick to a level but don't want to evolve themselves. With Qt5 it's the
>> same. I've been writing programs with both 4.8 and 5.5 and changing to
>> five was the best decision I made. Things got that much easier and
>> faster that even my boss couldn't believe it.
>>
>>
>> cheers,
>> Christian
>> -- "A little learning never caused anyon

Re: [Veusz-discuss] Filtering based text values

[Jeremy Sanders]

Hi Frans

On 20/03/16 15:51, Frans Ooms wrote:


So, I am probably doing something wrong in the expressions with strings...?
Or is it maybe related to the NumPy 'where' function?
Is there maybe a workaround?


Veusz uses lists of strings for text datasets, not numpy arrays. You 
could try working around this by wrapping the dataset name with array() 
to convert to a numpy array.


Maybe Veusz should switch to numpy string datasets, but I haven't 
investigated how hard this would be and what the effect on memory 
consumption would be.


Hope this helps

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] Move to Qt 5?

[Jeremy Sanders]

Dear all

Qt 4 is no longer supported and Qt 5 is included in many linux 
distributions.


Would there be any objections if I merge the Qt 5 branch into the main 
tree? It shouldn't affect binary distributions (though I'll need to 
update my setups).


Thanks

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Changing tick labels format - how can I do?

[Jeremy Sanders]

Hi André

On 02/11/2016 07:26 PM, André Luiz Regis Monteiro wrote:


I´m plotting data in x axis from 0 to 0.1 in seconds.

I need to show

0 . to . 100
time [ms]  instead,

but, when I choose %VE format , it shows me
0mto100m

How can I take "m" off?


%VE format adds the suffix. The standard %Vg format (or Auto) doesn't 
show it. Is there a reason you switched to %VE?


Cheers

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] line plot with text variable as x?

[Jeremy Sanders]

Hi Tim

On 02/09/2016 05:28 PM, Tim Gaines wrote:

I'm new to the list today, and though I hope to plot some research results 
later, my first goal is to plot some personal blood pressure measures as a 
function of month (a text variable).  I haven't been able to make it happen.  
If I turn the month labels into a number sequence (1 to 12) it works, but I 
want to extend the plot over several years and don't want to keep counting up 
the months.  Have I missed something?



Another way to do this is to use a text dataset as labels. You can 
create a new text dataset containing Jan Feb Mar, etc... In your point 
plotter choose this dataset in the labels setting. Then change the 
horizontal axis mode to be labels. You can hide the text labels by each 
point under the plotter labels formatting tab.


Hope this helps

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Layering of images

[Jeremy Sanders]

Hi Mark

On 02/01/2016 04:30 AM, Mark Durre wrote:


Perhaps what is needed is to just numerically add the RGB colors from
each layer (obviously clipping at 255), with the user adjusting the
colors by choice of colormap (e.g. blue, green and red, but could be
others) and the Min/Max values and scaling mode.


The layering is currently just what you get with Qt if you use different 
opacity for the filled regions. It's what you get if you just hold 
materials with different opacities in front of each other.


A more sophisticated image widget with layers could add the layers in 
different ways (e.g. as RGB components), but this would have to be done 
within the widget.


Cheers

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] ANN: Veusz 1.23.2

[Jeremy Sanders]

verbankcomputing.co.uk/software/pyqt/
   http://www.riverbankcomputing.co.uk/software/sip/
 numpy >= 1.0
   http://numpy.scipy.org/

Optional requirements:
 h5py (optional for HDF5 support)
   http://www.h5py.org/
 astropy >= 0.2 or PyFITS >= 1.1 (optional for FITS import)
   http://www.stsci.edu/resources/software_hardware/pyfits
   http://www.astropy.org/
 pyemf >= 2.0.0 (optional for EMF export)
   http://pyemf.sourceforge.net/
 PyMinuit >= 1.1.2 (optional improved fitting)
   http://code.google.com/p/pyminuit/
 dbus-python, for dbus interface
   http://dbus.freedesktop.org/doc/dbus-python/
 astropy (optional for VO table import)
   http://www.astropy.org/
 SAMPy or astropy >= 0.4 (optional for SAMP support)
   http://pypi.python.org/pypi/sampy/

Veusz is Copyright (C) 2003-2015 Jeremy Sanders <jer...@jeremysanders.net>
 and contributors.
It is licensed under the GPL (version 2 or greater).

For documentation on using Veusz, see the "Documents" directory. The
manual is in PDF, HTML and text format (generated from docbook). The
examples are also useful documentation. Please also see and contribute
to the Veusz wiki: https://github.com/jeremysanders/veusz/wiki

If you enjoy using Veusz, we would love to hear from you. Please join
the mailing lists at

https://gna.org/mail/?group=veusz

to discuss new features or if you'd like to contribute code. The
latest code can always be found in the Git repository
at https://github.com/jeremysanders/veusz.git.
___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Export to PS/PDF without date/time information

[Jeremy Sanders]

Hi Michael

On 12/20/2015 02:48 PM, Michael Clerx wrote:


I have a bunch of figures that I export to EPS and PDF using a script.
Each time I regenerate the figures, git indicates they have changed.
Inspecting the EPS files shows the only change is a comment about the
creation date. Inspecting the PDF is more troublesome, as git treats it
as a binary file, but I'm assuming the same thing happens.

Is there any way to disable creation time information in the export so
that the figures only differ when the actual contents have changed?


The PS and PDF output is created by Qt's QPrinter code. Unfortunately it 
doesn't look like it's possible to program it to disable the creation 
date in its output.


It might be possible to delete the date in the postscript file by 
modifying it (Veusz has to do this to make the page size correct, so it 
could be stripped out). For PDF it's trickier, as you say it's a binary 
file. There are some tools available which will let you dump the decoded


There's a page here which suggests that it might be possible to add a 
filter to git to do the stripping:


http://stackoverflow.com/questions/16058187/make-git-ignore-the-date-in-pdf-files

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Back ground display for images with "nan" data

[Jeremy Sanders]

Hi Mark

I've tried an example 2x2 plot with a single nan pixel and it always 
seems to be transparent, with normal and inverted colour maps. I've 
tried the heat map. You should see the background colour of the graph, 
unless the background of the graph is hidden.


If the graph has no background, for bitmap formats you should get the 
colour set in the export dialog box (Bitmap background). This is 
transparent by default (for PNG). I think you get the non transparent 
version for other bitmap formats. In my test I got transparent and white 
for PNG and JPG.


For PDF, the transparent area should render in your PDF viewer as the 
normal page colour, as nothing should be drawn there.


Send me an example, if there's still a problem.

Thanks

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Integration

[Jeremy Sanders]

Hi Jeff

On 10/21/2015 08:27 PM, Jeffrey Steplowski wrote:


I am using your program for my physical chemistry class. I was trying to
find out how to integrate an exponential decay graph so I could find the
area underneath the graph, but I'm having a lot of trouble. Could you
assist me on how I can go about finding the area under the curve?


You could try plotting trapz(datasety, datasetx) - see 
http://docs.scipy.org/doc/numpy/reference/generated/numpy.trapz.html


This should estimate the area using trapezoid integration. There's also 
the cumsum function for estimating using rectangles (if you have 
regularly spaced data).


Cheers

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Group plots for color automation

[Jeremy Sanders]

Dear Weiguo

On 10/20/2015 10:59 PM, Yin, Weiguo wrote:

Can many plots be grouped together and assigned different colors or line styles 
automatically?


Unfortunately there isn't a good way to make groups which automatically 
get assigned colors or styles.


You can select the same style by selecting multiple plotters in the tree 
simultaneously.


You can set ranges of colors or random colors for all the plotters 
within a document or graph by looking under the menu Tools->Colors


It would be nice to add a sophisticated plotting widget for plotting 
multiple groups of data and assigning properties together, but I haven't 
come up with a good way of doing that.


Thanks

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] monitoring live data

[Jeremy Sanders]

On 18/10/15 17:31, Carl Wolff wrote:


I like to use Veusz for monitoring live data (update interval 1 sec,
maximal 10 separate figures to monitor). In this case Veusz must show the
actual figures and previous values over some timewindow (e.g. one hour).
Data older than the timewindow is to be disregarded.

Is this possible with Veusz? I think I have to write a plugin, or embed
Veusz in our python data acquisition package.


Veusz has a capture dialog, which may be useful to you. Have you looked 
at this? It can read data from an application or socket and plot the 
previous N values.


Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Accidentally Scrolling in Drop Down Boxes

[Jeremy Sanders]

On 02/10/15 09:26, Simon Kleinknecht wrote:


I noticed this on my Windows 8.1 (Veusz 1.23.1), so I guess I cannot
change any theme or behavior. I checked it again and it's possible to
scroll through values in every list in the Formatting and Properties
windows.


It is possible to change this from the veusz console, if you want to see 
whether other themes are better. At the console type


import veusz.qtall as qt
qt.qApp.setStyle(qt.QStyleFactory.create("XXX"))

where XXX is the style, which on Windows (for me) allows Windows, 
WindowsXP, WindowsVista, Motif, CDE, Plastique and Cleanlooks as valid 
options.


Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] (1) how to access Export() function in plugin? (2) how to reload plugin from command line?

[Jeremy Sanders]

Dear Douwe

On 08/18/2015 01:05 PM, douwe van der veen wrote:

1. The Export() call is not working. From the file 
...\veusz\document\export.py, line 152, I seem to provide three variables to it instead 
on the single one on the command line: doc, filename, and pagenumber. I only provide filename here. 
How do I obtain this 'doc' value? Can I also access this thing from the command line?


Have you tried using interface.Export(...) instead of Export(...)? 
Inside a plugin, the command interface are methods of the interface 
object, not plain functions. You shouldn't need to pass the doc item - 
this is internal. The interface object is defined in 
veusz/document/commandinterface.py



2. When I modify my plugin, I have to close and re-open Veusz in order to load 
the plugin's latest version. Is there a way to force reload of the plugin from 
the command line?


There's no support for reloading plugins, unfortunately.

Hope this helps

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Embedded BoxPlot

[Jeremy Sanders]

Hi Dennis

On 07/15/2015 01:53 PM, Zink, Dennis wrote:

your tool seems  to be perfect for my issues . Thank you for that.

For evaluating many results I’m trying to create boxplots automatically.
First I don’t need an embedded Box. The vsz-File is important for me. Is
it possible to build a vsz-File in Python without showing the graph window?


Yes, you can use Embedded(Name, hidden=True) to hide the window.


I tried to write a short test script (see below). But I can’t get a
proper output. For the version below I get the ValueError: API 'QDate'
has already been set to version 1


What type of system are you running on?  I don't get that message (which 
is a bit worrying), but got


Traceback (most recent call last):
  File test.py, line 41, in module
BoxPlot_1.values.val=[y_test[1],y_test[2]]
  File /home/jss/code/veusz/veusz/embed.py, line 476, in _setVal
self._ci.Set(self._path, val)
  File /home/jss/code/veusz/veusz/embed.py, line 316, in runCommand
return self.sendCommand( (self.winno, cmd, args[1:], args2) )
  File /home/jss/code/veusz/veusz/embed.py, line 309, in sendCommand
raise retobj
veusz.utils.utilfuncs.InvalidType

I think you may need a str() around the list, as this setting has a text 
type. It runs to the end with this change.


The best way to learn the various settings for Veusz is to make plots in 
the GUI and look at the produced vsz files. I was working on an proper 
API documentation generator, but this is at the bottom of my priority 
list at the moment. The Veusz source code documents the settings as 
well, see e.g. 
https://github.com/jeremysanders/veusz/blob/master/veusz/widgets/boxplot.py 
(look for addSettings).


Hope this helps

Jeremy



___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Number Format and Axes Visualized

[Jeremy Sanders]

On 10/07/15 14:22, André Luiz Regis Monteiro wrote:

Please, I am new in Veusz.
I would like to know:

1-  how can I change a number format, from 2,34 (separate with comma) to
2.34 (separate with dot) to show it in the axes?


This is changed by the locale settings. Change on a per-document basis 
by clicking on the document in the tree, then clicking English locale 
in properties. Alternatively, to force it always go to Preferences and 
choose Override system locale settings to show Veusz in US English.



2 - how can hide right and upper axes (only to show x and y axes)?


Disable the Auto mirror format option for each axis. You may want to 
go to the graph and hide the border, too.


Hope this helps

Jeremy




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] position of bars in a graph

[Jeremy Sanders]

Dear Anna

On 07/03/2015 07:03 PM, Anna Burniol wrote:


I am having troubles with a bar graph. i have 4 sets of data that go with
the same x axis and two y axis. I can plot them in two y axis, but the
bars are overlapped. I tried playing a bit with the option position of
the x axis, but I actually do not really understand it. I do not manage to
make it work.
Any suggestions?


I'm not quite sure what you're doing, but you could try leaving blank 
(use nan) values alternately for each dataset so that the bars don't 
overlap, e.g.


ds1:

1
nan
2
nan
3

ds2:
nan
2
nan
4
nan

Alternatively, the positions setting gives the X position of the bars. 
You could try using Positions 1,3,5,7 for one and 2,4,6,... for the 
other. You may need to change Bar fill to make the bars narrower.


Hope this helps

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Export image quality higher than 300 dpi

[Jeremy Sanders]

Hi Otini

On 02/07/15 17:05, Otini wrote:

Most of the journals we publish in these days require the quality
of black and white line art to be a minimum of 900 dpi.  Is there
a way in to get to a minimum of 900 dpi (either from within Veusz
or with third party software)?


Recent versions of Veusz with the new export dialog box (1.23+) allow 
you to enter a bitmap DPI value up to  DPI. It seems to work ok for 
me using values of 900. Veusz should remember the previous value used. 
If you want to go past , a value can be passed to the Export() 
command line interface command, but I imagine there are likely to 
problems allocating enough memory.


Hope this helps

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] ANN: Veusz 1.23.1

[Jeremy Sanders]

Dear Fabian

On 06/25/2015 06:13 PM, Fabian Braennstroem wrote:


I just tried to install the lastest binary on a opensuse 13.1.
When I start veusz, I get the error that it is missing libSM.so.6, but
actually
this is in /usr/lib64.
Do you have an idea, how I can correct this.


The Veusz binary is 32 bit. I suspect you only have the 64 bit version 
of libSM.so.6. I'm not familiar with SuSE, but there are probably 32 bit 
X11 compatibility libraries which can be installed.


I believe there are also native openSUSE packages available which are 
likely a better option

http://software.opensuse.org/download.html?project=sciencepackage=python-veusz

Unfortunately I only have time to make 32 bit linux and windows versions 
of Veusz. If anyone is interested in helping with this, please let me know.


Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] ANN: Veusz 1.23.1

[Jeremy Sanders]

Dear Simon

Simon Kleinknecht wrote:

 Hi Jeremy, thanks for the new version. Great job!
 
 Once again I would like to ask about the global color behavior of a xy
 widget. Is it your intention that it does not control the color of the
 marker border?

Yes - that's correct. I thought it looked better with a black border. You 
can hide the border if it's a problem (unless you're using the line 
markers).

 Moreover, I think it would be nice if the sub-colors (plot line, marker
 fill, etc) pick up the global color again when selecting again the
 global color from a custom one. Right now the link between the global
 and the sub-colors breaks once you have changed the sub-colors. What do
 you think about it?

That was intentional again, though it would be difficult to implement any 
other behaviour with the current settings system in Veusz. You can click on 
the label, then Reset to default to restore the link.

Best wishes

Jeremy



___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Veusz 1.23.1 test release

[Jeremy Sanders]

On 06/13/2015 10:38 PM, Jeremy Sanders wrote:

I had to make some changes due to a bug in loading previous documents.
If you downloaded the testing link before receiving this email, please
update to the latest version in that directory again.

It looks like even small changes need a proper beta release.


Sigh. There was another problem, but only in the linux binary release. 
Please download again if you're using that.


Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] Veusz 1.23.1 test release

[Jeremy Sanders]

Dear all

Given some bugs in the histogram and exporting code, I've decided to 
make a 1.23.1 release. As this is a bug fix release, I won't make a beta 
test. Instead, I've put the to be released version here:


http://download.gna.org/veusz/testing/

Please let me know if there are any *new* problems in this release. I'll 
announce it in a few days if there are none. The changes are as follows:


Changes in 1.23.1:
 * Enable compression in exported files (Benjamin K. Stuhl)
 * Add option to thin error bars plotted
 * Fix label colour in contour labelling
 * Remove use of deprecated Qt functions
 * Fix clone widget tools plugin
 * Fix EPS/PDF crash with broken axis widget
 * Do not crash filter plugin if using date datatype
 * Avoid crashes in normalise, subtract minimum and divide maximum plugins
 * Avoid crash in image widget
 * Avoid issue in identifying spacing in 1D-2D conversion
 * Add Gnuplot 2D data import plugin (Joerg Meyer)
 * When loading data from import plugins, provide default values

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Veusz 1.23.1 test release

[Jeremy Sanders]

On 06/13/2015 11:42 AM, Jeremy Sanders wrote:


Given some bugs in the histogram and exporting code, I've decided to
make a 1.23.1 release. As this is a bug fix release, I won't make a beta
test. Instead, I've put the to be released version here:

http://download.gna.org/veusz/testing/


I had to make some changes due to a bug in loading previous documents. 
If you downloaded the testing link before receiving this email, please 
update to the latest version in that directory again.


It looks like even small changes need a proper beta release.

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Formatting individual words and symbols in axis labels?

[Jeremy Sanders]

Dear Steve

On 12/06/15 01:38, Steven Vanek wrote:

Hi, just wondering if it is possible to format individual words or parts
of text as part of axis labels?  I am working only in the GUI at this
point and I only see the option to change font or italics on the entire
label.

e.g., I would like to use superscripts in units like g mL -1


Veusz supports general formatting using LaTeX-like codes in the text. 
For instance, to do superscript you can do


g mL^{-1}

and subscripts like

x_{i}

You can also do selective bold, italic, underline, e.g. \bold{text} 
\italic{text} and \underline{text}, even \font{Arial}{my text}. Many 
symbols are supported like this (e.g. \alpha), but you can also directly 
paste in unicode symbols.


The manual lists the supported codes.

Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] date-type data filtering

[Jeremy Sanders]

Dear Sergey

Thanks for the report. The crash is fixed with this commit
https://github.com/jeremysanders/veusz/commit/0207311d4b43ff9006faaef2a14e2422c97603e9

Unfortunately, the filtered dataset doesn't preserve the type of the 
date type (but can still be plotted as dates), due to a design flaw in 
how dataset plugins work.


You can instead use the new Data-Filter dialog, which doesn't use the 
filtering plugin. It should also handle filtering text.


Best wishes

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] API problem between SIP and qmml

[Jeremy Sanders]

On 10/06/15 11:35, brunetto wrote:

Hi!
This morning my Veusz 1.23 installed from source and working yesterday
evening can't start.
I'm using Kubuntu 14.04LTS, up-to-date.

The error is

RuntimeError: the sip module implements API v11.0 but the qtmml module
requires API v11.1

I checked the upgraded packages and no one is related.


This is a version mismatch between SIP and Veusz. There must be an older 
version of SIP on the computer than Veusz was built with. I don't know 
how this could happen if you're using the Ubuntu-provided SIP, as the 
API should never go backwards. Are you sure there isn't another SIP on 
the computer (e.g. try locate sip.so)?


Jeremy



___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] 3D plots progress

[Jeremy Sanders]

Dear all

I'm making some progress on the 3D plotting functionality. I've got a 
reasonably stable plotting backend. At the moment the code can plot 3D 
points and function-based surfaces and lines. It's missing 3D data 
surface plots. There's also some work needed to get axis labels and 
numbers to appear in the right place.


Here's an example image to see what's there:

http://imgur.com/4DmOOsw

The code implements a straightforward Painter's algorithm renderer 
(which isn't accurate for objects with overlap in depth) and a much more 
accurate BSP (binary space partitioning) code. BSP is much slower and 
tends to make large output. It also blows up with many solid surfaces as 
the memory usage scales badly. I'm thinking of later adding something 
else which will be in between in terms of speed and output accuracy.


Grab the threed branch if you're interested (don't expect any continuing 
compatibility in saved files when using 3D plots, however).


Best wishes

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Clone for datasets functional?

[Jeremy Sanders]

This seems to fix the problem. I'll include the fix in the next release.

https://github.com/jeremysanders/veusz/commit/754728b624318a8f95dbcd9e16f3ff91649d2aae

Thanks

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Clone for datasets functional?

[Jeremy Sanders]

Dear Douwe

On 04/06/15 22:47, douwe van der veen wrote:


Observed behavior:
After applying widget, I see 19 additional XY graphs are created: good!
Yet, all have as y-axis data L01_data. The widget names are adjusted
though if I select build new names from ds


Unfortunately it looks like the cloning code is broken. You're correct 
in describing how it should work. I don't know why, but I'll have a look.


Thanks for reporting the problem

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Load a Veusz file, create a figure and store it from Python

[Jeremy Sanders]

Dear Michael

On 01/06/15 21:43, Michael Clerx wrote:


 1. Load an existing Veusz file
 2. Create an EPS plot
 3. And store it somewhere


In addition to Gilles's answer, you can do

From the command line to export directly from a vsz file
$ veusz --export out.eps in.vsz

You can also use the Python embedding interface to load existing, plots, 
modify the data and re-export, e.g.


import veusz.embed
e = veusz.embed.Embedded()  # use hidden=True not to show window
e.Load('test.vsz')
# can call e.SetData(), e.ImportFile(), e.ImportFileCSV(), etc... here
e.Export('test.eps')

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] using a variable as a scaling fr an axis and in a text field

[Jeremy Sanders]

Dear Yan

On 01/06/15 14:08, Yan Grange wrote:


So basically for my axis function I have a function in the form of (t/4e6)*100. 
Is there any way I can replace the 4e6 by something like sum(dataset_a)?


Unfortunately they code which evaluates the axis function doesn't do the 
dataset substitution, unlike other places which evaluate expressions. I 
assume making a new t_new, which is t/t.max() is no use?


I can't think of a workaround for this, as the evaluation code is pretty 
strict on what's accepted. Perhaps it's possible to do something 
horrible, like add add a python module to the custom dialog to import 
veusz, and extract the data out of window.document.data. It would be 
nice to remove this limitation.



The next question would be: can I actually also put a text box that essentially 
puts this value on the plot somewhere (so basically adding a parameter 
calculated based on the data to a text box)?


Again, this isn't possible. It would be a nice feature to allow 
expansion of Python expressions within text labels, though.


The only work around would be a Tools Plugin that you would have to 
manually run to modify the document to set both of these.


Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] How to add text labels to an x-axis?

[Jeremy Sanders]

On 19/05/15 16:20, Martin Pescador wrote:


I wish to plot some values in a time series (i.e. one value for each
month of a year) as points connected by a line. I can do this fine by
plotting y versus a vector that contains 1, 2, 3 etc.. But I cannot
figure out how to easily add text labels to the x-axis (e.g. Jan., Feb.,
etc.). I'm sure there is an easy way and I just can't see it!



It is possible to import date information, to get this kind of behaviour 
more automatically (though you need to change the axis mode to date).


To plot labels on an axis, you need to set the label option of an xy 
plotter to point to a text dataset. You then need to set the axis mode 
to be labels. You probably also want to hide the labels on the xy 
plotter themselves (in the formatting tab).


Unfortunately it's not very discoverable.

Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Automatic export

[Jeremy Sanders]

Dear Henrik

On 12/05/15 18:54, Henrik Rydberg wrote:

Is there an easy way to automate export with overwrite functionality?
(pdf/svg and so on). I'm thinking about functionality like the Reload
where you are able to set a timer... automatically reload every XX seconds.


Probably the only sensible option would be to use dbus.  A dbus request 
could be sent every XX seconds to export the document. Dbus is pretty 
powerful - for instance you can update the data in documents from other 
programs using this. Dbus might only work on linux or unix, as I'm not 
sure what the state of support is on Windows or Mac. I can give some 
help if this sounds useful.


Unfortunately tools plugins can only do a set of operations once. 
Perhaps it would be useful to extend them to allow persistent calls.


The plugin interface could be used to hack into the main veusz code, but 
this wouldn't be easy.


Cheers

Jeremy




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Zero axis

[Jeremy Sanders]

On 12/05/15 20:49, Gilles Zumbach wrote:

First of all, a big THANK YOU to Jeremy for this great tool!

Then, a request. A feature that I miss repeatedly is the possibility to add a
zero axis, namely a line (and more generally with ticks and labels) at a
position given as a coordinate in a perpendicular axis. Its length would
extend across the graph.


Thanks for the suggestion, Gilles. I'll keep it in mind.

Cheers

Jeremy



___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] The postscript glyph problem

[Jeremy Sanders]

On 04/24/2015 07:03 PM, Tim Channon wrote:

I am creating a small data presentation as a PDF, 179 pages in total, 6
plots and two images per data page. There is a small amount of textual
content.

The pages are created by Veusz as PDF files.

These are combined into a single PDF (10.8MB) by Ghostscript 9.15

Unfortunately this trips a known issue with Postscript, reported as
glyph errors. Oddly this is only for a few of all the PDF being combined.


I don't know how to get ghostscript to remap characters. I've 
experienced the same problem myself. To fix it I wrote a script to merge 
the PDF files with pypdf2. See the attached python script.


Jeremy
#!/usr/bin/env python

# merge PDF files using PyPDF2

from __future__ import division, print_function

import argparse
import PyPDF2

def main():
parser = argparse.ArgumentParser(description='Merge PDF files')
parser.add_argument('outfile', help='Output file')
parser.add_argument('infile', nargs='+', help='Input files')

args = parser.parse_args()

outpdf = PyPDF2.PdfFileMerger()

for filename in args.infile:
print('Opening', filename)
with open(filename, 'rb') as inf:
inpdf = PyPDF2.PdfFileReader(inf)
outpdf.append(inpdf)

print('Writing', args.outfile)
with open(args.outfile, 'wb') as outf:
outpdf.write(outf)

if __name__ == '__main__':
main()
___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Multiline text

[Jeremy Sanders]

On 23/04/15 17:13, Tim Channon wrote:

There seems to be no obvious way of creating multi-line text, no line
break method.

How is this done?


\\ is a line break

I should also add a way to enter this as a normal line break.

Jeremy



___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] ANN: Veusz 1.23

[Jeremy Sanders]

On 04/22/2015 06:10 PM, Yubaraj Malakar wrote:

After upgrading to new version ie 1.23, image plot for 2d dataset in log
scale color mode did not show anything except whole graph filled with lowest
color. After tricking with (2d dataset+1), it worked fine. Is this a bug?


No - it's a deliberate change. The definition of log has changed in the 
image widget to actually plot the log. The old definition was adding a 
constant to allow 0 values to be plotted. This not quite log behaviour 
broke other parts of the program (e.g. colorbars).


I probably should have highlighted this change better as it does change 
the behaviour of the program.


Thanks

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Veusz-future: Plans on 3D-Plots

[Jeremy Sanders]

Hi Christian

On 04/22/2015 12:01 AM, Christian Ambros wrote:


are there any plans on adding 3D-Plots into veusz?
Using matplotlib or pyqtgraph it shouldn't be that hard to add.


I did do quite a lot of work looking at this (see the threed branch). I 
had a basic vector renderer working, but it's hard to get all the 
details right. My aim was to get proper vector output, but it seems to 
be a hard problem. Many programs just output bitmaps (e.g. OpenGL) or 
only work in limited situations.


The main problem with my renderer was to properly get the order of 
overlapping polygons correct. This is important if you want to plot 
multiple surface functions which may intersect and could be transparent.


Matplotlib doesn't really fit into the veusz way of doing things as it's 
not object oriented (I did look at re-basing veusz on matplotlib at some 
point). It would also be a massive dependency. The matplotlib code also 
looked rather limited in what you could plot and wasn't a general renderer.


I may have a look back at my code and see whether I can make it more 
robust. If anyone knows of a robust general algorithm for depth sorting 
triangles please let me know. I could use a bsp tree, though that would 
be terribly slow and make a huge number of triangles.


Best wishes

Jeremy




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Possible to add custom tutorial?

[Jeremy Sanders]

Hi Martin

On 04/21/2015 01:18 PM, Martin Pescador wrote:


In the first instance, is this possible? If so, can you point me in the
right direction so that I can investigate whether it lies within my
technical abilities?


The tutorial code is defined in this source file:

https://github.com/jeremysanders/veusz/blob/master/veusz/windows/tutorial.py

You'd need to make your own classes based on TutorialStep. At the moment 
the code just starts with StepIntro. It won't be too hard to do if you 
just want to show some instructions. More involved interactive tutorials 
will be harder.


If you're building from source then it should be possible to replace the 
code easily enough. If you want to use the binary, it would be more 
difficult. It may be possible to hack in some code using the plugin 
system, e.g.


import veusz.windows.tutorial as tutorial
from veusz.windows.tutotorial import TutorialStep

class MyStep(TutorialStep):
  def __init__(self, mainwin):
...

tutorial.StepIntro = MyStep

Cheers

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Version 1.23 Export

[Jeremy Sanders]

Hi Mark

The greyscale option appears to work ok for me with PDF. Are you sure 
there's a problem? You could try switching it on and off again.


Thanks

Jeremy


On 04/21/2015 01:13 PM, Mark Durre wrote:

Hi Jeremy,

The Greyscale option on the new Export page doesn't seem to produce
greyscale images, still colour for PDF, EPS and PS.

Cheers,

Mark Durré
Ph.D. Student
Centre for Astrophysics and Supercomputing
Swinburne University of Technology
PO Box 218, Hawthorn
Victoria 3122, Australia
Email: mdu...@swin.edu.au








On 19/04/15 8:00 PM, veusz-discuss-requ...@gna.org
veusz-discuss-requ...@gna.org wrote:


Send Veusz-discuss mailing list submissions to
veusz-discuss@gna.org

To subscribe or unsubscribe via the World Wide Web, visit
https://mail.gna.org/listinfo/veusz-discuss
or, via email, send a message with subject or body 'help' to
veusz-discuss-requ...@gna.org

You can reach the person managing the list at
veusz-discuss-ow...@gna.org

When replying, please edit your Subject line so it is more specific
than Re: Contents of Veusz-discuss digest...


Today's Topics:

   1. ANN: Veusz 1.23 (Jeremy Sanders)


--

Message: 1
Date: Sat, 18 Apr 2015 14:31:38 +0200
From: Jeremy Sanders jer...@jeremysanders.net
To: veusz-annou...@gna.org
Cc: veusz-discuss@gna.org veusz-discuss@gna.org
Subject: [Veusz-discuss] ANN: Veusz 1.23
Message-ID: 55324eaa.4020...@jeremysanders.net
Content-Type: text/plain; charset=utf-8; Format=flowed

I'm please to announce the release of Veusz 1.23. Please see the release
notes below for more details of what's included.

There are plans that this could be the final (major) release using Qt4
and Python2. Future releases would move to Qt5 and Python3. Please let
me know if this is a problem.

Jeremy
-- next part --
Veusz 1.23
--
http://home.gna.org/veusz/

Veusz is a scientific plotting package.  It is designed to produce
publication-ready Postscript, PDF or SVG output. Graphs are built-up
by combining plotting widgets. The user interface aims to be simple,
consistent and powerful.

Veusz provides GUI, Python module, command line, scripting, DBUS and
SAMP interfaces to its plotting facilities. It also allows for
manipulation and editing of datasets. Data can be captured from
external sources such as Internet sockets or other programs.

Changes in 1.23:
* Add new export dialog box which can export multiple pages and
   modify the export options
* Add new dataset filtering dialog
* Add cubehelix() functional colormap
* Add -stepN suffix for colormaps to make arbitrary numbers of steps
* Fix incorrect colors in log images and log color scales
* Fix unsafe commands not being run

Minor changes
* Fix incorrect use of None in (x,...) pattern
* Catch crash if plotting nan/inf value in log space
* Fix getData in dataset plugin for dimensions=2
* Catch error in too large float to date time conversion
* Catch disappeared file during import
* Index error fixed in pickable
* Catch error in data edit dialog if 2d dataset size changes
* If root widget is selected with others, do not error on hide
* Fix undo for dataset histogram with a single output dataset
* Fix error resizing ellipse with a tuple width, height or position
setting
* Only use finite values in histogram
* Rewrite Line/FillSet setting controls for internal consistency and to
fix
   new style extended fills
* Do not crash with log date-time axes
* Also ignore non-finite values when fitting with minuit
* Avoid syntax error with invalid colormap
* Updates to setup.py and desktop files
* Recreate dataset now works if dialog hasn't been opened already
* Restore dock layout when using Python3
* Fix undo after loading stylesheet/custom definitions
* Support unicode example filenames
* Clip bezier lines to avoid problems with log axes

Features of package:
Plotting features:
  * X-Y plots (with errorbars)
  * Line and function plots
  * Contour plots
  * Images (with colour mappings and colorbars)
  * Stepped plots (for histograms)
  * Bar graphs
  * Vector field plots
  * Box plots
  * Polar plots
  * Ternary plots
  * Plotting dates
  * Fitting functions to data
  * Stacked plots and arrays of plots
  * Nested plots
  * Plot keys
  * Plot labels
  * Shapes and arrows on plots
  * LaTeX-like formatting for text
  * Multiple axes
  * Axes with steps in axis scale (broken axes)
  * Axis scales using functional forms
  * Plotting functions of datasets
Input and output:
  * EPS/PDF/PNG/SVG/EMF export
  * Dataset creation/manipulation
  * Embed Veusz within other programs
  * Text, HDF5, CSV, FITS, NPY/NPZ, QDP, binary and user-plugin importing
  * Data can be captured from external sources
Extending:
  * Use as a Python module
  * User defined functions, constants and can import external Python
functions
  * Plugin interface to allow user to write or load code to
 - import data using new formats

Re: [Veusz-discuss] Veusz as a slave

[Jeremy Sanders]

On 20/04/15 02:47, Tim Channon wrote:


I find the export() command is fragile anyway, won't accept all the
mentioned parameters. I have also managed to trip an MS C++ lib
exception. (nothing like an idiot to find cracks)


The following works ok on Linux for me. It's possible I haven't tested 
the Windows support thoroughly enough, however. I don't have a system to 
hand to check at the moment.


test.vsz:

To(Add('page'))
To(Add('graph'))
Add('function', function='x')
To('..')
To('..')
To(Add('page'))
To(Add('graph'))
Add('function', function='x**2')
To('..')
To('..')
Export('out.pdf', page=[0,1])
Quit()


At the command line:

veusz --listen  test.vsz


Can you send the exception if this doesn't work (and/or your current 
exception)?


Thanks

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] ANN: Veusz 1.23

[Jeremy Sanders]

Hi Daniele

Thanks for the comments. Py2 support is fairly easy to maintain for the 
moment. I think a Qt5 move may have to be made soon as Qt4 isn't really 
supported any more. It's probably ok as it is on Linux/Unix and Windows, 
but Mac is a bit of a moving target.


I believe there may be some bugs in Qt4 with recent Mac systems, e.g. 
Yosemite, but I don't have one to confirm. For instance, can anyone tell 
me on Yosemite whether the modal dialog boxes (e.g. Preferences) work 
ok? I've just read that such dialogs might not work.


Thanks

Jeremy

On 04/20/2015 10:37 AM, Daniele Paganelli wrote:

Hi Jeremy,
Great news!
About dropping py2/Qt4 support.
Some popular pre-built scientific python distributions like Python(x,y)
and Enthought Canopy are still bound to py2.
Anaconda and WinPython offer py3 but not Qt5.

These users might not be very happy if you drop py2/Qt4 support.

Distributions are slowly moving (as Anaconda and WinPython already offer
both 2/3, they might add qt5 in near future), but I fear this is because
some underlying scientific packages are in their turn still py2-only.
This would exclude even more users.

My 2 cents,
Daniele

ps: I also wrote a lot of code importing veusz modules and Qt4, and I
will not have time for migration in the next year or so.
Il 18/04/2015 14:31, Jeremy Sanders ha scritto:

I'm please to announce the release of Veusz 1.23. Please see the
release notes below for more details of what's included.

There are plans that this could be the final (major) release using Qt4
and Python2. Future releases would move to Qt5 and Python3. Please let
me know if this is a problem.

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss



___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Contour label problem

[Jeremy Sanders]

Hi Mark

This is fixed here:
https://github.com/jeremysanders/veusz/commit/fb7cfe72a3736e2291415f58acdf3f0d73d20a92

Thanks for the report.

Jeremy

On 04/18/2015 02:46 AM, Mark Durre wrote:

Hi Jeremy,

I've encountered a problem with contour labels – these seem to ignore
the colour setting and are always black; please the attached example
with data. This is for both 1.21 and the new 1.23 (beta).

Kind regards,

Mark Durré
Ph.D. Student
Centre for Astrophysics and Supercomputing
Swinburne University of Technology
PO Box 218, Hawthorn
Victoria 3122, Australia
Email: mdu...@swin.edu.au mailto:mdu...@swin.edu.au




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss



___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] ANN: Veusz 1.23

[Jeremy Sanders]

I'm please to announce the release of Veusz 1.23. Please see the release 
notes below for more details of what's included.


There are plans that this could be the final (major) release using Qt4 
and Python2. Future releases would move to Qt5 and Python3. Please let 
me know if this is a problem.


Jeremy
Veusz 1.23
--
http://home.gna.org/veusz/

Veusz is a scientific plotting package.  It is designed to produce
publication-ready Postscript, PDF or SVG output. Graphs are built-up
by combining plotting widgets. The user interface aims to be simple,
consistent and powerful.

Veusz provides GUI, Python module, command line, scripting, DBUS and
SAMP interfaces to its plotting facilities. It also allows for
manipulation and editing of datasets. Data can be captured from
external sources such as Internet sockets or other programs.

Changes in 1.23:
 * Add new export dialog box which can export multiple pages and
   modify the export options
 * Add new dataset filtering dialog
 * Add cubehelix() functional colormap
 * Add -stepN suffix for colormaps to make arbitrary numbers of steps
 * Fix incorrect colors in log images and log color scales
 * Fix unsafe commands not being run

Minor changes
 * Fix incorrect use of None in (x,...) pattern
 * Catch crash if plotting nan/inf value in log space
 * Fix getData in dataset plugin for dimensions=2
 * Catch error in too large float to date time conversion
 * Catch disappeared file during import
 * Index error fixed in pickable
 * Catch error in data edit dialog if 2d dataset size changes
 * If root widget is selected with others, do not error on hide
 * Fix undo for dataset histogram with a single output dataset
 * Fix error resizing ellipse with a tuple width, height or position setting
 * Only use finite values in histogram
 * Rewrite Line/FillSet setting controls for internal consistency and to fix
   new style extended fills
 * Do not crash with log date-time axes
 * Also ignore non-finite values when fitting with minuit
 * Avoid syntax error with invalid colormap
 * Updates to setup.py and desktop files
 * Recreate dataset now works if dialog hasn't been opened already
 * Restore dock layout when using Python3
 * Fix undo after loading stylesheet/custom definitions
 * Support unicode example filenames
 * Clip bezier lines to avoid problems with log axes

Features of package:
 Plotting features:
  * X-Y plots (with errorbars)
  * Line and function plots
  * Contour plots
  * Images (with colour mappings and colorbars)
  * Stepped plots (for histograms)
  * Bar graphs
  * Vector field plots
  * Box plots
  * Polar plots
  * Ternary plots
  * Plotting dates
  * Fitting functions to data
  * Stacked plots and arrays of plots
  * Nested plots
  * Plot keys
  * Plot labels
  * Shapes and arrows on plots
  * LaTeX-like formatting for text
  * Multiple axes
  * Axes with steps in axis scale (broken axes)
  * Axis scales using functional forms
  * Plotting functions of datasets
 Input and output:
  * EPS/PDF/PNG/SVG/EMF export
  * Dataset creation/manipulation
  * Embed Veusz within other programs
  * Text, HDF5, CSV, FITS, NPY/NPZ, QDP, binary and user-plugin importing
  * Data can be captured from external sources
 Extending:
  * Use as a Python module
  * User defined functions, constants and can import external Python functions
  * Plugin interface to allow user to write or load code to
 - import data using new formats
 - make new datasets, optionally linked to existing datasets
 - arbitrarily manipulate the document
  * Scripting interface
  * Control with DBUS and SAMP
 Other features:
  * Data filtering and manipulation
  * Data picker
  * Interactive tutorial
  * Multithreaded rendering

Requirements for source install:
 Python 2.x (2.6 or greater required) or 3.x (3.3 or greater required)
   http://www.python.org/
 Qt = 4.6 (free edition)
   http://www.trolltech.com/products/qt/
 PyQt = 4.5 (SIP is required to be installed first)
   http://www.riverbankcomputing.co.uk/software/pyqt/
   http://www.riverbankcomputing.co.uk/software/sip/
 numpy = 1.0
   http://numpy.scipy.org/

Optional requirements:
 h5py (optional for HDF5 support)
   http://www.h5py.org/
 astropy = 0.2 or PyFITS = 1.1 (optional for FITS import)
   http://www.stsci.edu/resources/software_hardware/pyfits
   http://www.astropy.org/
 pyemf = 2.0.0 (optional for EMF export)
   http://pyemf.sourceforge.net/
 PyMinuit = 1.1.2 (optional improved fitting)
   http://code.google.com/p/pyminuit/
 dbus-python, for dbus interface
   http://dbus.freedesktop.org/doc/dbus-python/
 astropy (optional for VO table import)
   http://www.astropy.org/
 SAMPy or astropy = 0.4 (optional for SAMP support)
   http://pypi.python.org/pypi/sampy/

Veusz is Copyright (C) 2003-2015 Jeremy Sanders jer...@jeremysanders.net
 and contributors.
It is licensed under the GPL (version 2 or greater).

For documentation on using Veusz, see the Documents directory. The
manual is in PDF, HTML and text format (generated

Re: [Veusz-discuss] Veusz 1.22.99 beta

[Jeremy Sanders]

Dear Fabian

On 14/04/15 10:40, Fabian Brau wrote:


I create a dataset using a parametric equation (in the data panel, those
data are expression type). Then, I do a plot point. I save the file and
open it. I right-click on the dataset in the data panel and in the menu
there is no edit. There is an edit data instead where I see the number
generated by the parametric equation but I cannot access the panel where the
parametric equation was initially written.


Thanks for the report. I've already spotted this and fixed the bug for 
the next beta or final release.


I've also almost finished implementing a native (non plugin) data 
filtering feature. I think I may include this in the next release, but I 
probably should make another beta to make sure there aren't any glaring 
bugs, as quite a bit of the dataset code has had to be reorganised.


Cheers

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] How to create ternary contour plot with Veusz?

[Jeremy Sanders]

On 04/09/2015 02:14 PM, nxkryptor nxkr wrote:

Can such plots be created with Veusz?


I'm afraid there's no support for contours on ternary plots.

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] Qt5 branch

[Jeremy Sanders]

If anyone is interested in trying out a Qt5 version of Veusz, there's a 
branch available here:

 https://github.com/jeremysanders/veusz/tree/qt5

I haven't found any problems with it so far. The only thing I might do 
is to rename all the qt4 symbols to qt, but this would make a big diff.


Does anyone have any objections to moving to Qt5 for the next release 
(after 1.23)? The only problem I know is that for X11, native drawing 
support has been dropped, so everything is done in big bitmaps and makes 
X11 network support perhaps sub-optimal.


We could also move to only supporting Python 3 at the same time to 
reduce workload. This would cause problems for people writing plugins or 
using the embedding module. However, it might be possible to make the 
embedding module able to work across the Python2/3 boundary. The other 
problem is that the EMF export is Python 2 dependent, though I have a 
fork of pyemf here which appears to work:

 https://github.com/jeremysanders/pyemf

Cheers

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] Veusz 1.22.99 beta

[Jeremy Sanders]

Hi -

I've made a beta release for 1.23 (called 1.22.99). Please download and 
try it from here:


http://download.gna.org/veusz/testing/

I'll release it in a few days if there are no problems.

The release notes are below.

Thanks

Jeremy


Veusz 1.23
--
http://home.gna.org/veusz/

Veusz is a scientific plotting package.  It is designed to produce
publication-ready Postscript, PDF or SVG output. Graphs are built-up
by combining plotting widgets. The user interface aims to be simple,
consistent and powerful.

Veusz provides GUI, Python module, command line, scripting, DBUS and
SAMP interfaces to its plotting facilities. It also allows for
manipulation and editing of datasets. Data can be captured from
external sources such as Internet sockets or other programs.

Changes in 1.23:
 * Add new export dialog box which can export multiple pages and
   modify the export options
 * Add cubehelix() functional colormap
 * Add -stepN suffix for colormaps to make arbitrary numbers of steps
 * Fix incorrect colors in log images and log color scales
 * Fix unsafe commands not being run

Minor changes
 * Fix incorrect use of None in (x,...) pattern
 * Catch crash if plotting nan/inf value in log space
 * Fix getData in dataset plugin for dimensions=2
 * Catch error in too large float to date time conversion
 * Catch disappeared file during import
 * Index error fixed in pickable
 * Catch error in data edit dialog if 2d dataset size changes
 * If root widget is selected with others, do not error on hide
 * Fix undo for dataset histogram with a single output dataset
 * Fix error resizing ellipse with a tuple width, height or position setting
 * Only use finite values in histogram
 * Rewrite Line/FillSet setting controls for internal consistency and to fix
   new style extended fills
 * Do not crash with log date-time axes
 * Also ignore non-finite values when fitting with minuit
 * Avoid syntax error with invalid colormap
 * Updates to setup.py and desktop files

Features of package:
 Plotting features:
  * X-Y plots (with errorbars)
  * Line and function plots
  * Contour plots
  * Images (with colour mappings and colorbars)
  * Stepped plots (for histograms)
  * Bar graphs
  * Vector field plots
  * Box plots
  * Polar plots
  * Ternary plots
  * Plotting dates
  * Fitting functions to data
  * Stacked plots and arrays of plots
  * Nested plots
  * Plot keys
  * Plot labels
  * Shapes and arrows on plots
  * LaTeX-like formatting for text
  * Multiple axes
  * Axes with steps in axis scale (broken axes)
  * Axis scales using functional forms
  * Plotting functions of datasets
 Input and output:
  * EPS/PDF/PNG/SVG/EMF export
  * Dataset creation/manipulation
  * Embed Veusz within other programs
  * Text, HDF5, CSV, FITS, NPY/NPZ, QDP, binary and user-plugin importing
  * Data can be captured from external sources
 Extending:
  * Use as a Python module
  * User defined functions, constants and can import external Python functions
  * Plugin interface to allow user to write or load code to
 - import data using new formats
 - make new datasets, optionally linked to existing datasets
 - arbitrarily manipulate the document
  * Scripting interface
  * Control with DBUS and SAMP
 Other features:
  * Data picker
  * Interactive tutorial
  * Multithreaded rendering

Requirements for source install:
 Python 2.x (2.6 or greater required) or 3.x (3.3 or greater required)
   http://www.python.org/
 Qt = 4.6 (free edition)
   http://www.trolltech.com/products/qt/
 PyQt = 4.5 (SIP is required to be installed first)
   http://www.riverbankcomputing.co.uk/software/pyqt/
   http://www.riverbankcomputing.co.uk/software/sip/
 numpy = 1.0
   http://numpy.scipy.org/

Optional requirements:
 h5py (optional for HDF5 support)
   http://www.h5py.org/
 astropy = 0.2 or PyFITS = 1.1 (optional for FITS import)
   http://www.stsci.edu/resources/software_hardware/pyfits
   http://www.astropy.org/
 pyemf = 2.0.0 (optional for EMF export)
   http://pyemf.sourceforge.net/
 PyMinuit = 1.1.2 (optional improved fitting)
   http://code.google.com/p/pyminuit/
 dbus-python, for dbus interface
   http://dbus.freedesktop.org/doc/dbus-python/
 astropy (optional for VO table import)
   http://www.astropy.org/
 SAMPy or astropy = 0.4 (optional for SAMP support)
   http://pypi.python.org/pypi/sampy/

Veusz is Copyright (C) 2003-2015 Jeremy Sanders jer...@jeremysanders.net
 and contributors.
It is licenced under the GPL (version 2 or greater).

For documentation on using Veusz, see the Documents directory. The
manual is in PDF, HTML and text format (generated from docbook). The
examples are also useful documentation. Please also see and contribute
to the Veusz wiki: https://github.com/jeremysanders/veusz/wiki

Issues with the current version:

  * Due to a bug in the Qt XML processing, some MathML elements
containing purely white space (e.g. thin space) will give an error.

If you enjoy using Veusz, we would love to hear from you. Please

Re: [Veusz-discuss] fitting over a set y range

[Jeremy Sanders]

On 26/03/15 19:51, Lauren Levac wrote:


This works, but I'm fitting ~60 data sets with multiple different fits
and always changing t_i.  Is there a way to have the max fit range
depend on the y values?  Then I could just set it to P_1 and fit all my
curves with that set point.

I've attached a sample of the data I'm using.  I'm fitting it with a
exponential (760-b*exp (-x/a)) from 0-400 torr and 0-700 torr and then
fitting with a line from 0-300 torr.


You could try creating a filtered dataset. You could created filtered 
versions of P and t, using an expression, e.g. P1, and fit these 
filtered datasets.


It's a bit harder if P may go back below 1. You could do something ugly, 
such as using an expression like


arange(len(P))  where(P1)[0][0]

where(P1)[0][0] gives the first element where P  1.

Hope this helps

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Merge two veusz files

[Jeremy Sanders]

Yes. It is also possible to concatenate the vsz files into the same 
file. Unfortunately you have to change the name of the page widgets to 
be unique in the text file for this to work. It also won't solve 
conflicts between the dataset names, etc.


Jeremy

On 25/03/15 23:21, Steve Richardson wrote:

I'm not sure what append does in origin, but you can copy pages or
graphs from one veusz file and paste them in another.

Steve


On Mar 25, 2015, at 12:52 PM, Yin, Weiguo w...@bnl.gov
mailto:w...@bnl.gov wrote:


Hello,

I’m new to Veusz and already found it’s delightful to use. I‘d like to
ask

How can one merge two Veusz files into one, like the “Append” function
in Origin by OriginLabwhich I frequently used?

Best regards,

Weiguo

___
Veusz-discuss mailing list
Veusz-discuss@gna.org mailto:Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss



___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] stops on colormap

[Jeremy Sanders]

Hi Yubaraj

On 03/08/2015 10:12 PM, Yubaraj Malakar wrote:


Is there any way to implement manual color stops instead of default equally
spaced color stops so that one can have better control over color of the
image of 2d dataset?


There's no current way to use non-equal stops directly. You can, of 
course, use lots of equal stops to emulate the non-equal stops, using 
e.g. 256 sets of values. You can interpolate between your non-equal 
stops to map to the fixed spacing.


Alternatively you could transform your data so that equally spaced stops 
can be used.


Using the fixed spacing makes lookup from value to colour very fast.

Hope this helps

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Incorrect logarithmic colorbar

[Jeremy Sanders]

Dear Fabian

On 03/06/2015 01:46 PM, Fabian Brau wrote:


I can send you an example if I was not clear enough.


Ok, I've fixed the problem here:

https://github.com/jeremysanders/veusz/commit/931341adaf09932b1ebf9b3203ee76c493247442

The problem was that the log scaling originally was designed to be 
log-like in order to allow images to be shown with negative values (this 
is like the astronomical software ds9). The behaviour was to use log 
scaling over the full range (min to max), even if min is negative. This 
doesn't work when combined with the colorbar widget.


Now the scaling is purely log, but if anyone was taking advantage of the 
log-like scaling allowing negatives, this change will break their plots.


Best wishes

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Label offset

[Jeremy Sanders]

Dear Simon

On 03/06/2015 05:59 PM, Simon Kleinknecht wrote:


Would it make sense to fix the axis label offset to the axis/frame of
the graph? This would solve the issue.


The at edge setting for the axis label may be what you want - set it 
for all the axes. Ideally Veusz should do this automatically for graphs 
in a grid, however.


Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] how to define color when plot a 2D image

[Jeremy Sanders]

On 06/02/15 04:37, sean wrote:

I major in seismic exploration. Recent I use venuse software to
image seismic wavefront. The software is easy and free to use. But there
is a few default color in 2D image. So I try to define the color by
myself, such as red-white-blue color.

   So I want to know how to define a color by myself.


You can do this in the Edit-Custom definitions dialog box. Choose Add, 
change type to Colormap, enter a name. The definition looks like


((255,0,0), (255,255,255), (0,0,255))

(for red-white-blue)

The values are red, green and blue from 0-255. Veusz interpolates 
between these colours and spaces them equally. You can have as many 
(R,G,B) values as necessary. You can add a 4th value to each triplet, 
which is the opacity. Start stepped colormaps with (-1,0,0).


They are saved with the document. If you want to add them to each new 
document, choose Save from the custom definitions dialog box. You can 
then go to the preferences dialog box to specify a custom definitions 
file which is loaded for each new document.


For the next version of veusz I'm adding functional colormaps (for the
cubehelix scheme) and the ability to step any colormap with N steps.

I hope this helps

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] veusz1.22 windows version crash by big date, ubuntu, debian version of 1.22 run very nice

[Jeremy Sanders]

On 18/12/14 09:18, markus.go...@hamm.eu wrote:


I use veusz 1.22 under linux, hdf5 interface…. It’s nice and very good.
Co-worker in company must use the windows 1.22 version on win 7. This
version crash by using big data, 2GB *.vsz. Is the win veusz a 32bit or
a 64 bit version?


The binaries are just 32 bit versions (except for OSX, which is a fat 
32/64 bit binary), so that's probably why it doesn't work with more than 
2GB.


Thanks

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] veusz1.22 windows version crash by big date, ubuntu, debian version of 1.22 run very nice

[Jeremy Sanders]

Dear Markus

On 18/12/14 09:35, markus.go...@hamm.eu wrote:


Thanks! Can I build a 64bit windows version, or run a native Python
Veusz 1.22 under Windows 7 with Python2.7, Qt…?


It's possible to do either. Getting the dependencies might be difficult 
- you need a Python (2.7 or 3.3+), matching MSVC 2008 (for 2.7), numpy, 
Qt, sip, PyQt, hdf5 and h5py.


I use pyinstaller to build an exe file from these (it doesn't work on 
3.x). There's a pyinstaller configuration file in the github repository 
(this will need a little modification for the system).


However, it might be better to use the 64 bit module binaries built by 
Christoph Gohlke here:

 http://www.lfd.uci.edu/~gohlke/pythonlibs/

I've not tried them before. If you want an exe, you'll need to use 
pyinstaller or a similar tool (e.g. cx_freeze).


Cheers

Jeremy



___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] data values on bars

[Jeremy Sanders]

On 12/12/14 23:38, Birger Ilau wrote:


I need the bottom of each bar (under the x-axis) to have a label
(defined as a dataset names), I can do that by going /x-axis
properties - mode - labels/ and then defining labels from /bar
properties - labels - names/. So far it works fine. But now I need
data values (y-axis values) on top of each bar and I can not find a way
to do that. Only way to see data values on each bar is to put them
instead of my name labels under the x-axis, but I'm sure there's a way
to do it properly.


I don't know of an obvious way to do this. However, you could add a 
label widget to the plot if you can calculate x and y coordinates for 
the labels (these can be datasets), use your numerical dataset for Label 
and set the position mode to axes.


Hope this helps

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] how to define a function by adding other functions?

[Jeremy Sanders]

Dear Kazuyoshi

You can define functions in the Edit-Custom Definitions dialog box. 
Click Add. Change Type to function. Change name to, e.g. func1(x) and 
change definition to e.g. 2*x+1. You can define func2(x) like this. 
sumFunc(x) should be defined func1(x)+func2(x)


Best wishes

Jeremy

On 09/12/14 05:28, Kazuyoshi Furutaka wrote:

Hi?

How can I define a function by adding some
already-defined (user-defined) functions?
e.g.
   func1: 2*x+1
   func2: x*x

   sumFunc: func1 + func2
 (instead of defining (2*x+1)+(x*x) again)

Thanks in advance.

Kazuyoshi
--
Kazuyoshi Furutaka
furutaka _dot_ kazuyoshi _at_ jaea _dot_ go _dot_ jp

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Using multiple datasets

[Jeremy Sanders]

Dear Scott

On 05/12/14 20:23, Scott Vetter wrote:

Just started using Veusz and was wondering if one can import multiple
datasheets in the csv format to plot on the same graph using different
data markers?  I could not seem to find how to do it.  Any help would be
great.


Sorry, but I'm not sure what datasheet means here. CSV only supports a 
single grid of values.


You can of course import multiple CSV files into a Veusz document and 
plot them with different markers. You need to use different names for 
the datasets in the CSV files (the optional names at the top of 
columns), or use the prefix/suffix options when importing to 
differentiate them. You'll see the imported datasets in the dataset 
window (usually on the right), or in the Data-Editor dialog.


You just need to add xy plotting widgets for each set of data you want 
to plot (Insert-Add xy). Set the datsets to plot in the X data and Y 
data settings. Choose the markers and colours for each xy widget.


There are lots of example plots in the help menu. The tutorial is also 
useful if you haven't done this.


Best wishes

Jeremy




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Overlapping labels

[Jeremy Sanders]

Hi Fred

On 10/11/14 12:21, Fred Labrosse wrote:


I'm trying to use scatter plots and label the glyphs using another of the
fields in my data.  However, many of my points are very close (or on top of
each other), which is fine.  The problem is that all the labels end up on
top of each other, making reading them impossible.

I have seen in the archives something about jittering the points
themselves, but nothing about labels.

Can this be done or is that to go on the wish list?


I'm afraid the labelling is not intelligent enough to avoid putting 
labels on top of other labels.


It might be possible to implement some sort of overlap check and to try 
putting labels in alternative positions, but this is a TODO unfortunately.


Thanks

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Using Key with multiple Y axes

[Jeremy Sanders]

Hi

On 30/10/14 23:40, Barry Buternowsky wrote:

My plot has multiple Y axes. I used the Key feature to display the
legend. I have 3 y axes on my plot. One of the left, and two on the
right. Only the item from the first axis (left axis) was shown in the
legend/key box.


Can you send an example of where this fails? Are you sure you don't have 
more than one graph? The multiaxes example provided with Veusz does this 
and is ok.


Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] Veusz memory usage

[Jeremy Sanders]

Hi Barry

Veusz stores data in double precision arrays, so that's 8 bytes for each 
value. If you multiply your array length by 8 and divide by 1048576, 
that should tell you how many megabytes you will use per dataset. Be 
aware that Veusz is currently only compiled for 32 bit for Windows, so 
there will be a limitation of 2GB of memory or so. There's no fixed 
limit on array sizes. From your numbers, there shouldn't be a problem in 
terms of memory.


Importing the data might be a bit slow - I've not tested CSV files that 
long recently. Plotting the points might also be slow. The plotting 
speed should be much faster if you disable the anti-aliasing in the plot 
(right click on the page to see the menu).


Hope this helps

Jeremy

On 18/10/14 23:20, Barry Buternowsky wrote:

Hi,
I am currently using Excel vba to import multiple files. Then I create about 30 
charts with x-axis as date format. Each chart has a left and right y axis. The 
files I import are csv with multiple columns, and some columns could be over 
10 points. I do some processing on the data before charting. Now I am 
finding I am running out of memory. With vba I tried keeping some of the data 
in vba buffers to make it faster, but vba Excel has a bug/limitation in that 
arrays over 65535 points cannot be imported into charts. Therefore all data I 
now store in the worksheet. Would Python/veusz use less memory for this. 
Thoughts?





On Oct 18, 2014, at 11:57 AM, Jeremy Sanders jer...@jeremysanders.net wrote:

I'm pleased to announce the release of Veusz 1.22. Please find the release 
notes. Note that this release fixes serious bugs in the 2D dataset support with 
irregular data points (particularly inverting the coordinates).

There are some changes from the beta, including a couple of bug fixes, improved 
resizing of shapes and a preferences option to add directories to the python 
path.

Jeremy
README
___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss




___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Capturing data set -

[Jeremy Sanders]

On 23/09/14 13:26, Simon Kleinknecht wrote:

I managed to import the data after adding a % at the beginning of line
1; however, I have no idea how to tell Veusz to get the DateTime right.
I end up with one Date and one Time object which I cannot combine. I
changed the date format manually to -MM-DD, my OS default format,
from DD.MM. to enable automatic date object recognition in Veusz. If
you have specified a different format you have to change it again.


Veusz can't really handle a date and time separated by a space when 
capturing. But if you insert a T before the date and time (as in the ISO 
specs), then it is treated as a single item e,g,


2000-01-01T12:00:00

This seems to work for me. You can also make the first line descriptor 
DateTime B1_1 B1_2..., so you don't have to fill the datasets in.


The capturing code could be improved. I don't know how widely it is used.

Best wishes

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] rescording a data set automatically

[Jeremy Sanders]

On 21/09/14 19:13, Simon Basyuk wrote:

I have created a set of multiple points on a graph, by some means. Now,
I use the Read data points on the graph option to get coordinates of a
small set of points, by sliding cursor over them. Can I write these
coordinates to a data set automatically, while sliding the cursor? (I
can not program with Python to build a script for that).


I'm afraid there's no way to do that automatically. However, you can 
enable picking to the clipboard (in preferences), paste the data 
somewhere else and re-import it into Veusz.


Best wishes

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] Importing FITS fils

[Jeremy Sanders]

Hi John

On 04/09/14 21:15, Josh Fuchs wrote:


I have recently begun using Veusz. I am not able to import FITS files
and can't figure out why, despite plentiful searching online. Below is a
screenshot of what I get when I select a FITS file to import. As you can
see, Veusz will not allow me to Import or adjust the columns. I feel
like I am missing something simple. I do have astropy installed.


The FITS import only allows the import of binary/ascii tables and 1D/2D 
images. Unfortunately, the data file you have there consists of a 3D 
image, which Veusz doesn't know how to handle. It would be useful in the 
future to add some sort sophisticated treatment of FITS files (such as 
allowing slicing), to allow this sort of data to be imported.


Best wishes

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

[Veusz-discuss] Veusz 1.21.1 pre-release

[Jeremy Sanders]

Hi -

I'm going to make a Veusz 1.21.1 release soon. I've uploaded it to
http://download.gna.org/veusz/testing/ , so you can get early access.

Please let me know whether there are any major regressions in this 
release soon. It's mostly a bug fix release. The relevant changes are


Changes in 1.21.1
 * Add swap option on keys, swapping symbol and text

Bug fixes
 * Fix missing labels if colorbar out of plot
 * Preserve tags when reloading datasets
 * Handle nan/inf/-inf when as value in UI
 * Remove dependence on pyqtconfig in build system
 * Handle errors properly in 2D CSV read
 * Do not crash if selecting siblings of document widget
 * Fix accidental modification of data by line widget
 * Improve PDF bounding box rounding to prevent chopping near edge
 * Fix function plotter in embedding
 * Fix astropy version checking
 * Fix crash for error bar in CSV with blank previous column

Best wishes

Jeremy

___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss

Re: [Veusz-discuss] embed Examples

[Jeremy Sanders]

On 15/08/14 13:01, Simon Schopferer wrote:

I would greatly appreciate any help, as I am very impressed with Veusz
and would like to use it for my work (however this would require working
scripting/embedding interface).


Unfortunately it looks like there was a bug in the embedding interface, 
which didn't set up the environment properly, breaking expression 
evaluation. Thanks for the bug report.


I've fixed it here:
https://github.com/jeremysanders/veusz/commit/55341dfb1f94fcafe1e6c0213127499710d761d9

I'm planning to make a 1.21.1 release soon, which will have this fix.

Thanks

Jeremy


___
Veusz-discuss mailing list
Veusz-discuss@gna.org
https://mail.gna.org/listinfo/veusz-discuss