Mailman 3 October 2001 - SciPy-User

plotting enhancements and copyright
by Jochen Küpper 01 Nov '01

01 Nov '01

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Cc'ed to the scipy lists. Should also make it into the archive. This message is to inform the users and developers od SciPy what's going on in politics. The plot_window class (an earlier version of which I send to this lists) or respective changes to plot_canvas and some further changes to scipy.plt won't get included in scipy 'cause of politcal issues. This is sad as it provided a first implementation of zooming and an *useful* interface to use scipy.plt as wxWindow in other applications. See below for more. Because this is new for most people I cite a little more. Eric, eric> 4) Do you feel strongly about the inclusion of the copyright? For clarification: This is refering to a line at the top of the respective source file: "#Copyright (C) 2001 Jochen Küpper" >> Yes. The code is copyrighted by me anyway, so why not tell everybody? eric> Ok. Unfortunately, I am not interested in encumbering the eric> existing code with another copyright for a single or even eric> multiple additional features. This sort of thing is almost eric> never done because it can lead to major hassles in the future eric> trying to work with the 20 people who have added feature patches eric> and bug fixes to a file over its lifetime. What problems are you thinking of here? Changing the license? What else? eric> So I guess the best thing to do then if for you to derive a new eric> plot_canvas class and add/override the methods you need to and eric> place it in a new module. That's what I did to begin with. eric> You can then copyright this module. Yeah. I guess I just put it into jkext, so I can everything the way I like it, not restricted by scipy rules. (I do not want to argue about these rules. I think there have to be rules and the ones that are there are appropriate for the project from my point of view.) The module is GPL'ed. eric> I guess you need to include the Enthought/BSD copyright also if eric> any or my original code is in the modules and you plan on eric> distributing it separately. Yes. eric> As for how it is handled in SciPy, I'm not sure. We can include eric> your module, but the zoom capability is also likely to be folded eric> back into the original class once someone has the time to sit eric> down and work on the graphics code (with someone having to redo eric> your work). It isn't hard, as I showed. I actually spent more time to fit it into scipy than writing the stuff in the first place:( And this while I was learning wxPython. eric> I guess we'll need to add a policy that copyrights for patches eric> and bug fixes will need to be assigned to the original code eric> author or copyright holder. Contributors will be acknowledged in eric> the THANKS file. This is the standard practice in OS projects eric> so I didn't feel the need to make it explicit, but I will now to eric> prevent future confusion on the issue. New modules can retain eric> the authors copyright, as long as it is BSD compliant (or eric> whatever we eventually end up with). I think you are confusing things here. "Copyright" and "license" are two distinct features. Any code I write I do have a copyright on, no matter what the license is. I can give up my copyright i.e. by transfering it to someone else, but unless I explicitely do that I'll own the copyright. The license is whole different story. (I am ok with the BSD style licens, esp. since I see that there is (good) reasoning behind it.) Many people never write the FSF-papers. So far I am one of them. (Not that they would seriously need mine.) For now I am not going to transfer my copyright on BSD licensed code to a company, sorry. I wish SciPy good luck! Bye, Jochen - -- Einigkeit und Recht und Freiheit http://www.Jochen-Kuepper.de Liberté, Égalité, Fraternité GnuPG key: 44BCCD8E Sex, drugs and rock-n-roll -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6-cygwin-fcn-1 (Cygwin) Comment: Processed by Mailcrypt and GnuPG <http://www.gnupg.org/> iD8DBQE74HPBiJ/aUUS8zY4RAkg8AJ9/samJAatCoGs8HtHZiyFYGa2LQQCfT0OY aFLQzbp5IzVZZKaUGovzZFA= =4r2z -----END PGP SIGNATURE-----

4 9

plt
by Jochen Küpper 23 Oct '01

23 Oct '01

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi All, once more my appreciation for a great project! I hope you (we) get this a more-or-less standard package for python! After using the plt-module interactively for some time I want to embed it into a small app. So I need to use it as a workarea within the frame of my application. I managed to use plot_canvas for that but this of course looses most of the functionality (axes, zoom, print, ...) So what is the preferred way to use plt as a widget? Am I overseeing something? Greetings, Jochen - -- Einigkeit und Recht und Freiheit http://www.Jochen-Kuepper.de Liberté, Égalité, Fraternité GnuPG key: 44BCCD8E Sex, drugs and rock-n-roll -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: Processed by Mailcrypt and GnuPG <http://www.gnupg.org/> iD8DBQE707h7iJ/aUUS8zY4RAim8AJwIu8QRDGXzMBGTFGgFFFfigCJtxgCgp7FK TdQZ4+OsmJ6GmJmXL1DSQJw= =A3Nw -----END PGP SIGNATURE-----

2 1

Install Problems (missing g77)
by Hartley, Ed 17 Oct '01

17 Oct '01

I had the same problem it's a missing g77 This doesn't get installed on some Linux flavours by default particularly in my case RedHat7.1 The picture is slightly confused when looking for rpms becuase the rh numbering diverges from gnu which run 2.95.1, 2.95.2, 2.95.3, 3.00, 3.01 so you need an g77 a redhat rpm for the same redhat revision as your gcc. 2.96 in my case. BTW I don't know why the numbering diverges. Hope this helps. Ed Hartley ------ Distributed Multimedia Research Group Computing Department Lancaster University Lancaster UK LA1 4YR Phone +44 (0) 1524 593675 Fax +44 (0) 1524 593608

2 1

installation_help
by Karshi Hasanov 16 Oct '01

16 Oct '01

Hi all, I try to install SciPy and wasn't successfull. Do I have to have the Frotran libs installed to SciPy to work. If " No", then how do I avoid from getting the following masseges: ------------------------------------------------------------------------- skipping special/cephes/isnan.c (build/temp.linux-i686-2.1/isnan.o up-to-date) skipping build/temp.linux-i686-2.1/libcephes.a (up-to-date) running build_flib Traceback (most recent call last): File "setup.py", line 532, in ? include_dirs = ['Numerical/Include'] File "/usr/local/lib/python2.1/distutils/core.py", line 138, in setup dist.run_commands() File "/usr/local/lib/python2.1/distutils/dist.py", line 899, in run_commands self.run_command(cmd) File "/usr/local/lib/python2.1/distutils/dist.py", line 919, in run_command cmd_obj.run() File "/usr/local/lib/python2.1/distutils/command/build.py", line 106, in run self.run_command(cmd_name) File "/usr/local/lib/python2.1/distutils/cmd.py", line 328, in run_command self.distribution.run_command(command) File "/usr/local/lib/python2.1/distutils/dist.py", line 918, in run_command cmd_obj.ensure_finalized() File "/usr/local/lib/python2.1/distutils/cmd.py", line 112, in ensure_finalized self.finalize_options() File "build_flib.py", line 83, in finalize_options raise ValueError, 'Fortran compiler not available' ValueError: Fortran compiler not available ------------------------------------------------------- Thanks

2 1