I don't understand what this prep flag to dsolve does (other than give wrong results, as you've identified). This must have been something Chris added. When is it useful to have prep=False?
Aaron Meurer On Fri, Feb 22, 2013 at 7:39 AM, Manoj Kumar <[email protected]> wrote: > Thanks again on your timely reply. > > 1. I had indeed downloaded the zip file just as a precaution. > > 2. On playing around with the source code , I found out that, it is because > of the unevaluated derivative , that leads to a wrong output. > > But in cases like the above mentioned, I guess there should be an option, by > which we can directly integrate it , something like dsolve(d, f, hint = > 'direct') . > > It maybe inconsequential in cases like above , but in cases like > > Derivative(x* f(x), x , x , x) , where it could give a solution by direct > integration (without preprocessing) ,and preprocessing would give a > NotImplementedError I think it would be helpful. > > This is my attempt at solving the problem, > https://github.com/sympy/sympy/pull/1814 . > > > > > On Fri, Feb 22, 2013 at 7:48 PM, Stefan Krastanov > <[email protected]> wrote: >> >> Hi, >> >> There are a few things that I find unclear, however I do reproduce >> this behavior. >> >> First, there is no need to download any zip files (I guess you did it >> as a precaution). Please just use git so we can use the commit hashes >> and know about what revisions of the code we are talking. >> >> Then, be aware that there is no building/compiling in python. It is >> interpreted language (there are some unimportant detail about .pyc >> caches). If you have changed the code you are using the changed >> version. >> >> Here is what I got: >> >> >>> f = f(x) >> >>> d = Derivative(x * f , x) >> >>> dsolve(d , f) >> Cā >> f(x) = āā >> x >> >>> dsolve(d , f , prep = False) >> f(x) = Cā >> >> I am quite sure that this happened because `d` was constructed as an >> unevaluated derivative. After >> >> d = d.doit() >> >> all works well. >> >> >> I am not sure whether this should be considered a bug. After all sympy >> permits you to construct many unevaluated objects >> (Add(evaluate=False), Mul(evaluate=False), Integral, Derivative, Sum) >> but does not promise to work well with them. >> >> -- >> 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 [email protected]. >> To post to this group, send email to [email protected]. >> Visit this group at http://groups.google.com/group/sympy?hl=en. >> For more options, visit https://groups.google.com/groups/opt_out. >> >> > > > > -- > Regards, > Manoj Kumar, > Mech Undergrad. > BPGC > Blog > > -- > 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 [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/sympy?hl=en. > 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 [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sympy?hl=en. For more options, visit https://groups.google.com/groups/opt_out.
