Hi, On 23 March 2011 12:15, Aaron S. Meurer <[email protected]> wrote:
> Hi. > > On Mar 23, 2011, at 8:31 AM, VPeric wrote: > > > Hello, > > > > I'm a student at the Faculty of Electrical Engineering, Czech > > Technical University in Prague. I'm studying Open Informatics, which > > is a study program orientated towards research and the utilization of > > new technologies. > > > > While I have experience with several languages (notably C, Java, > > Prolog), Python is by far my favorite language. I find the issues > > related to conversion to Python 3 particularly interesting, and would > > like to work on porting SymPy to Python 3. The porting is inevitable > > if SymPy is to survive, and doing it sooner rather than later offers > > advantages that could be helpful in increased adoption (and hence > > increased development), so I consider it a very worthwhile pursuit. > > > > My main problem is whether this is simply "big enough" for a GSoC > > project? While it is listed on the Ideas page, after a cursory glance > > at the code I don't see any major problems here (or I simply missed > > them, of course). I have read about the last time such a conversion > > was attempted and the problems encountered, but as that was over 18 > > months ago, I believe most of those problems can be solved now. > > > > My plan would be to first run in 2.6 with no errors (ie. "Python -3"), > > which shouldn't be too difficult (around 150 errors, most of which > > look trivial), and then try to maintain a dual codebase with automatic > > running of the 2to3 tool when building for Python3 (which is the > > recommended procedure). Problems might arise from the fact that SymPy > > currently supports Python 2.4 or higher, but this would have to be > > solved as it comes up. I hope it will be possible to keep supporting > > 2.4, though limiting it to 2.6 would likely result in cleaner code. > > > > Thank you for your time, > > Vladimir Perić > > So we are planning on dropping 2.4 support after the next release (see > http://code.google.com/p/sympy/issues/detail?id=2135). This should make > it much easier to support Python 3 (indeed, it is already difficult to > support Python 2.4 without supporting Python 3). We do plan on keeping > Python 2.5 support into the future, though. > Leaving Python 2.5 support is important because Python runtime environment in Google App Engine is based on 2.5 and SymPy must run in App Engine. > > Maybe others can comment on whether they think this is a suitable project > or not. Previously, there were quite a few things that needed to be done > for Python 3 support, but many of them have already been fixed. Definitely > one thing that needs to be done is to automate as much as possible the > process of handling Python 3 code. We do not want to have two separate code > bases, so the best way would be if we could just code in Python 2 like we do > now and have everything transferred over to Python 3 automatically. Another > thing that should be considered is setting up buildbots that automatically > run the tests in all versions of Python (including Python 3) to make sure > everything is running OK. > > So I think you would have to consider a Python 3 project to be less about > fixing the code base to work with Python 3 (which should not take very > long), and more about making it as painless as possible for us to support > it. > > Aaron Meurer > > -- > You received this message because you are subscribed to the Google Groups > "sympy" group. > To post to this group, send email to [email protected]. > To unsubscribe from this group, send email to > [email protected]. > For more options, visit this group at > http://groups.google.com/group/sympy?hl=en. > > Mateusz -- You received this message because you are subscribed to the Google Groups "sympy" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/sympy?hl=en.
