tl;dr; : On 6 January 2016 at 20:48, Nathann Cohen <[email protected]> wrote: > A ticket is in needs_review at #19661 [1], and the branch's main > feature is a new graph constructor. I write here because of a > disagreement Dima and I have about the code: I do not like at all that > most of the code is a *string* that is sent to GAP for evaluation.
There is a call to libgap.function_factory(), which contains (ahem!) GAP code as the parameter. (That's what libgap.function_factory() is for, no?). That is, the essential part of the functionality is a GAP function that is installed into libGAP using libgap.function_factory() The said function is then called to create a Sage object from its output. The reviewer objects to this style, saying that the guts of the GAP computation must be exposed to Sage (with finer libgap calls, I presume). Apart from being harder to write and debug, the style suggested by the reviewer is degrading performance unnecessarily. [...] > > [1] http://trac.sagemath.org/ticket/19661 -- 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 post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
