Yes, this functionality would be useful in a general setting (probably in the ntheory module). It would also be useful to have a class, ContinuedFraction, which can help with the abstractions of the continued fraction representation, especially the infinite ones. There is actually a lot of cool arithmetic you can do on continued fractions. The class would also handle printing.
Of course, you shouldn't spend too much time away from the things you planned to do for GSoC, but if you find some free time, maybe you could look into it. Otherwise, just open an issue. Aaron Meurer On Fri, Jun 28, 2013 at 11:56 AM, Thilina Rathnayake <[email protected] > wrote: > > Code segment I proposed should be changed as below. Otherwise it fails > when sqrt()'s are not involved. > > > def continued_fraction(a, b): > p = Wild('p') > q = Wild('q') > r = Wild('r') > > a = Add(a) > > match = a.match(p + r*sqrt(q)) > # and do the processing according to whether > # q = 0 or not > > > > On Fri, Jun 28, 2013 at 10:00 PM, Thilina Rathnayake < > [email protected]> wrote: > >> One approach would be to have three inputs to continued_fraction() >> function >> corresponding to the general form of a quadratic irrational fraction (a + >> sqrt(b)) / c >> We can set b = 0 for other cases. But I think it is not a good approach. >> >> Another approach would be to keep the current interface of >> continued_fraction() >> unchanged and pattern match to identify sqrt() s involved in the >> numerator. >> >> def continued_fraction(a, b): >> p = Wild('p') >> q = Wild('q') >> match = a.match(p + sqrt(q)) >> # and do the processing according to whether >> # q = 0 or not >> >> >> >> On Fri, Jun 28, 2013 at 8:23 PM, Thilina Rathnayake < >> [email protected]> wrote: >> >>> Yes, that is where it should go. I don't think it will be that hard to >>> combine. >>> There is a Java code <http://www.alpertron.com.ar/CONTFRAC.HTM> that is >>> used in Dario Alpern's website. >>> >>> I need the continued_fraction representation of quadratic irrational >>> numbers >>> in the Diophantine equation module. I can implement it in diophantine >>> module >>> but I think that does not belong there. We should implement it in ntheory >>> and then use it in Diophantine module. Shall I start on this? >>> >>> >>> >>> On Fri, Jun 28, 2013 at 8:49 AM, Stephen Loo <[email protected]> wrote: >>> >>>> I have considered this, in quantum module, the continued_fraction >>>> function is only work for rational number a/b = continued_fraction(a, b), >>>> but my function above is work for float and depending on the precision. I >>>> don't know how to combine into single function, and it is not suggested to >>>> place in quantum module, it should be in ntheory. >>>> >>>> Stephen >>>> >>>> Thilina Rathnayake於 2013年6月28日星期五UTC+8上午10時58分36秒寫道： >>>>> >>>>> >>>>> Thanks guys for the help. Shouldn't we add this functionality to the >>>>> current continued_fraction or >>>>> add a new function to SymPy so user can use it directly? >>>>> >>>>> >>>>> >>>>> On Fri, Jun 28, 2013 at 7:45 AM, Stephen Loo <[email protected]>wrote: >>>>> >>>>>> def continued_fraction(r): >>>>>> while True: >>>>>> i = int(r) >>>>>> yield i >>>>>> if i == r: >>>>>> break >>>>>> r = 1 / (r - i) >>>>>> >>>>>> >>>>>> Thilina Rathnayake於 2013年6月28日星期五UTC+**8上午4時16分34秒寫道： >>>>>> >>>>>>> Hi all, >>>>>>> >>>>>>> Is there a function in sympy which gives the continued fraction >>>>>>> representation >>>>>>> of quadratic irrationalities like, (sqrt(17) + 5 ) / 4? >>>>>>> >>>>>>> I found an implementation of a continued fraction function in >>>>>>> Shor's algorithm: >>>>>>> http://docs.sympy.org/dev/**modu**les/physics/quantum/shor.**html<http://docs.sympy.org/dev/modules/physics/quantum/shor.html> >>>>>>> >>>>>>> But it doesn't seem to give correct results when irrationalities are >>>>>>> involved. >>>>>>> >>>>>>> >>> from sympy.physics.quantum.shor import continued_fraction >>>>>>>> >>> continued_fraction(sqrt(17)+5, 4) >>>>>>>> [2, 4] >>>>>>>> >>>>>>> >>>>>>> Actual answer should be 2, [3, 1, 1] with the numbers inside [] >>>>>>> repeating forever. >>>>>>> Any help would be appreciated. >>>>>>> >>>>>>> Regards, >>>>>>> Thilina. >>>>>>> . >>>>>>> >>>>>> -- >>>>>> 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+un...@**googlegroups.com. >>>>>> To post to this group, send email to [email protected]. >>>>>> >>>>>> Visit this group at >>>>>> http://groups.google.com/**group/sympy<http://groups.google.com/group/sympy> >>>>>> . >>>>>> For more options, visit >>>>>> https://groups.google.com/**groups/opt_out<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. >>>> For more options, visit https://groups.google.com/groups/opt_out. >>>> >>>> >>>> >>> >>> >> > > > Thilina Rathnayake, > > Undergraduate, > > Dept. of Comp. Sci. > > UoM. > [image: > LinkedIn]<http://s.wisestamp.com/links?url=http%3A%2F%2Flk.linkedin.com%2Fpub%2Fthilina-bandara-rathnayake%2F35%2F36a%2F50%2F> > “"A casual stroll through the lunatic asylum shows that faith does not > prove anything." - Friedrich > Nietzsche<http://s.wisestamp.com/links?url=http%3A%2F%2Fwww.quotesdaddy.com%2Fquote%2F477550%2Ffriedrich-nietzsche%2Fa-casual-stroll-through-the-lunatic-asylum-shows-that> > ” Get this email app! > <http://s.wisestamp.com/links?url=http%3A%2F%2Fwww.wisestamp.com%2Fapps%2Fquotes%3Futm_source%3Dextension%26utm_medium%3Demail%26utm_term%3Dquotes%26utm_campaign%3Dapps> > > > -- > 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. > 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. For more options, visit https://groups.google.com/groups/opt_out.
