We should also hopefully be in shape if not now pretty soon to release pretty much whenever, so it should hopefully be possible to just rely on the latest version, and if you need something from git, to just push a new version out.
I say "if not now pretty soon" because it seems that currently there are several cracks in the Travis testing system, and it seems that test failures have leaked through all of them. The ones I currently know of - Travis doesn't fail if warnings are printed to stderr - Travis is installed before it is tested, so files that are not installed are not tested (I'm not sure if any of these remain). - NumPy is always installed by Travis, so there are test failures that only show up in that case are not seen (I think the current failures that I'm thinking of may have been fixed). - Travis doesn't test the slow tests. - Travis doesn't test with the cache disabled. Once we get these fixed, and https://github.com/sympy/sympy/pull/2333 is finished, let's do another release. Aaron Meurer On Fri, Aug 16, 2013 at 12:24 PM, Jason Moore <moorepa...@gmail.com> wrote: > Ok, I've started this repo: https://github.com/PythonDynamics/PyDy > > I'll be working on getting it into shape so that we can have one installer > for the PyDy distribution and it will contain SymPy mechanics and the new > visualization package. > > You'll be able to do things like: > > from pydy import mechanics > from pydy import visualization > > or > > from pydy import ReferenceFrame, VisualizationFrame > > We'll need to think about how to setup the name space. > > I may use git submodules or may not and just use the setuptools dependency > stuff to keep track of things. I can see how the git submodule is useful for > having an up-to-date dev version of an external library, but I'm not that > sold on them. We can always pin versions of PyDy to versions of SymPy and it > be expected that the master branch of PyDy is dependent on the master branch > of SymPy. I don't see a lot of reason to pollute the git repo with the > submodule. I typically use virtual environments and setuptools to handle > this kind of stuff. > > > Jason > moorepants.info > +01 530-601-9791 > > > On Tue, Aug 13, 2013 at 2:52 PM, Aaron Meurer <asmeu...@gmail.com> wrote: >> >> It's worth pointing out that the key to it being active is not so much >> whether or not you yourself plan to work on it, but rather how much of >> a community you are able to build around it. If you really encourage >> contributions from the people who have shown interest, then I think it >> will be successful even outside of SymPy. >> >> Aaron Meurer >> >> On Tue, Aug 13, 2013 at 12:48 PM, Ondřej Čertík <ondrej.cer...@gmail.com> >> wrote: >> > On Tue, Aug 13, 2013 at 12:02 PM, Jason Moore <moorepa...@gmail.com> >> > wrote: >> >> I don't there is any way to know how active the development will be. We >> >> presented it at two conferences for the first time in the last couple >> >> of >> >> months and have some people showing interest because of that. But for >> >> now it >> >> is just the core devs side project and we use it in our research. >> > >> > As to sympy.mechanics, my recommendation, based on my experience with >> > PyDy so far, is to wait till the late fall, >> > like November and see how the development goes. If it stays active, >> > then by all means >> > you should maintain your own project. Also so that you can include more >> > things, >> > for example some compiled code, and so on. >> > If, on the other hand, the development slows down, as it has in the >> > past between the summers, >> > it might be a good idea to keep it in sympy for now. >> > >> > As to PyDy Viz, by looking at the code >> > (https://github.com/PythonDynamics/pydy-viz), I would probably keep it >> > outside. >> > You can put sympy as a git submodule, and Travis will automatically test >> > it. >> > >> > It really depends on how active the development is --- if it is >> > active, then it is easy to keep sympy.mechanics working with recent >> > versions of sympy (in the git submodule). If it is not very active, >> > then it might be easier to let the rest of sympy team >> > to maintain sympy.mechanics. Even things like Python 3 support, you >> > get this for free by being part of sympy. >> > But if you are active, then it is not a big deal to do all this work >> > yourself. >> > >> > Ondrej >> > >> > -- >> > You received this message because you are subscribed to the Google >> > Groups "PyDy" group. >> > To unsubscribe from this group and stop receiving emails from it, send >> > an email to pydy+unsubscr...@googlegroups.com. >> > To post to this group, send email to p...@googlegroups.com. >> > Visit this group at http://groups.google.com/group/pydy. >> > For more options, visit https://groups.google.com/groups/opt_out. >> > >> > >> >> -- >> You received this message because you are subscribed to the Google Groups >> "PyDy" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to pydy+unsubscr...@googlegroups.com. >> To post to this group, send email to p...@googlegroups.com. >> Visit this group at http://groups.google.com/group/pydy. >> For more options, visit https://groups.google.com/groups/opt_out. > > > -- > You received this message because you are subscribed to the Google Groups > "PyDy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to pydy+unsubscr...@googlegroups.com. > To post to this group, send email to p...@googlegroups.com. > Visit this group at http://groups.google.com/group/pydy. > For more options, visit https://groups.google.com/groups/opt_out. -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to sympy+unsubscr...@googlegroups.com. To post to this group, send email to sympy@googlegroups.com. Visit this group at http://groups.google.com/group/sympy. For more options, visit https://groups.google.com/groups/opt_out.
