[MacPorts] #30521: ipython update to version 0.11
MacPorts
noreply at macports.org
Thu Aug 4 08:53:45 PDT 2011
#30521: ipython update to version 0.11
--------------------------------------+-------------------------------------
Reporter: Deil.Christoph@… | Owner: akitada@…
Type: update | Status: new
Priority: Normal | Milestone:
Component: ports | Version:
Keywords: | Port: py25-ipython, py26-ipython, py27-ipython
--------------------------------------+-------------------------------------
Comment(by michelle.lynn.gill@…):
Replying to [comment:11 Deil.Christoph@…]:
> Replying to [comment:10 michelle.lynn.gill@…]:
> > I've just installed the python26 version of ipython 0.11 with the
variants +scientific+parallel.
>
> How? Did you patch the py26-ipython Portfile using the diff I attached
to this ticket?
> I didn't try that myself, I'm waiting for the ipython Portfile
maintainer to answer / update.
Yes, I applied the patch to the existing iPython26 in a local port
repository because there are parallel computing features new to version
0.11 that are useful to my research. I realize the port maintainer needs
to commit a final version of the port, but it seems prudent to highlight
the port select issue now so it can be dealt with if necessary when such a
commit occurs.
> > I cannot install the qtconsole variant as Qt will not build for me on
Lion.
>
> There are three open issues for building Qt on Lion at the moment, we
just have to wait:
>
> https://trac.macports.org/wiki/LionProblems
I am aware of the link describing port issues involving Lion. This is what
I was referring to.
> > Also, please do not delete the scientific variant. There are
definitely some who use its capabilities. Thanks.
>
> I just don't see why installing http://dirac.cnrs-
orleans.fr/ScientificPython/ (which is port pyXX-scientific) should be a
variant of the ipython port.
> I've never used it, but as far as I can see from the web page it is just
some python package that is not related to ipython?
> There are hundreds of such packages, some much more useful (e.g. numpy,
scipy, matplotlib) than ScientificPython (which seems to be dead since
2007), which are also not ipython variants.
> If the variant is removed you can still use ScientificPython as you
always have, just install the pyXX-scientific port instead of selecting
the ipython scientific variant.
I understand how to install pyXX-scientific independently and import it as
a module in the iPython environment.
One reason the scientific variant exists is that one of the iPython-
provided profiles, ipythonrc-physics, uses ScientificPython. I also know
of at least one secondary project which uses this profile, so there may be
(minor) downfield compatibility issues. Whether or not the existence of
this profile and the possibility of other compatibility problems warrants
the creation of a variant is a philosophical discussion, but clearly
someone thought so at one point.
However, I have just noticed that as of version 0.11, several of the other
modules utilized by ipythonrc-physics are on [http://ipython.org/ipython-
doc/rel-0.11/whatsnew/version0.11.html#deathrow deathrow], so maybe these
concerns will become moot if the profile is broken for other reasons.
Regardless of what the iPython maintainer decides, the reason for the
existence of the scientific variant is not as arbitrary as it may seem.
--
Ticket URL: <https://trac.macports.org/ticket/30521#comment:12>
MacPorts <http://www.macports.org/>
Ports system for Mac OS
More information about the macports-tickets
mailing list