Thank you very much Mr.Bruin for your helpful advice.
I will follow your suggestion and try to start with a stand-alone repository. Thank you once again. Kenta Kobayashi 2023年3月27日月曜日 1:53:58 UTC+9 Nils Bruin: > On Sunday, 26 March 2023 at 05:54:09 UTC-7 kenta kobayashi wrote: > > Question 1: > Would it be appropriate to push these functions as a ‘Geometry and > Topology’ library in Sage? > > > Nowadays, for brand new code that isn't obviously extending or improving > already existing capabilities in sage, it's probably a good idea to start > with packaging the code into a stand-alone repository. It's your choice how > nicely packaged you want to make it: you can just make it a directory that > people can download and then link into sage (at their barest, python > modules really just consist of a directory with python files), or you can > wrap it up in a pip-installable package on github and/or publish it on pypi. > (see for instance https://github.com/nbruin/RiemannTheta for an > "intermediately polished" example) > > It has the advantage that the code is available immediately and that for > the first while you can respond very quickly to bugs and issues that arise > (so SageMath development cycle or review to contend with!). It also allows > for a period to gauge how people actually use the code, which can be quite > different from what you envisaged. > > Once the usage has stabilized a bit, it's worth pushing for inclusion into > the sagemath library itself, so that your code is kept up-to-date with > other changes in the library (over longer time spans this becomes > important). At that point you can archive the original repo with a pointer > to the relevant code in sagemath. > > This process also has the advantage that there is a specific place for you > to point at to show what you've accomplished (for job and grant > applications). > > > Question 2: > I cannot find any SageMath polynomial library that supports polynomials > of rational degrees. > However, elliptic genera are Laurent polynomial series of rational > degrees. > What is the most appropriate representation for this? > > > LaurentPolynomialRing (for finite series) and LaurentSeriesRing > > > Question 3: > In my code, I have redefined the zeta function only for negative integers > due to computation time. > Is it acceptable to use such ad-hoc coding, or should I use the zeta > function of the SageMath library in the code to be published? > > > You could look if this redefinition is already available somewhere in sage > and then you can import it from there. In general, you should probably use > whatever feature is already available that significantly fills your need, > to avoid duplication of effort and also because your (first) version would > probably not handle edge cases and variations of inputs as gracefully as > more mature code does. > > Your redefinition obviously needs to stay confined to your own modules and > then for maintainability it should probably be named "modified_zeta" to > match (so that other people reading your code understand it's not the usual > zeta function). > > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/f7fb620d-c0f6-4dda-9c13-ce1d3a1e1674n%40googlegroups.com.
