Re: [sage-devel] Confusing documentation in element.pyx

2014-12-17 Thread Samuel Lelievre
Vincent wrote: > Sorry to disturb. This discussion should go to the relevant ticket... > You are right. I have now voiced my concerns there: http://trac.sagemath.org/ticket/17480 -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe fro

Re: [sage-devel] Confusing documentation in element.pyx

2014-12-17 Thread Vincent Delecroix
Hello, Sorry to disturb. This discussion should go to the relevant ticket... Vincent 2014-12-17 12:08 UTC+01:00, Samuel Lelièvre : > More thoughts. > > 0. It's good that you are refreshing this part of the documentation, > if only to update references to Pyrex to references to Cython! > > 1. May

Re: [sage-devel] Confusing documentation in element.pyx

2014-12-17 Thread Samuel Lelièvre
More thoughts. 0. It's good that you are refreshing this part of the documentation, if only to update references to Pyrex to references to Cython! 1. Maybe it would be worth pointing to the subtle ways in which these functions differ in names and in nature: There are three relevant functions

Re: [sage-devel] Confusing documentation in element.pyx

2014-12-17 Thread Clemens Heuberger
Am 2014-12-17 um 10:47 schrieb Samuel Lelievre: > Hello! Your rewording mentions "four relevant functions" > but I can only see three listed. thank you, I forgot to change that after unifying two descriptions. Fixed (http://trac.sagemath.org/ticket/17480 ). > Also, there is a typo: "DD NOT" for

Re: [sage-devel] Confusing documentation in element.pyx

2014-12-17 Thread Samuel Lelievre
Clemens Heuberger wrote: > > Am 2014-12-15 um 12:00 schrieb David Roe: > > The difference is in how cpdef functions interact with Cython vs Python > > classes. If you want to override a cpdef method in a *Python* subclass > then you > > must use def (of course). But in a *Cython* subclass, yo

Re: [sage-devel] Confusing documentation in element.pyx

2014-12-16 Thread Clemens Heuberger
Am 2014-12-15 um 12:00 schrieb David Roe: > The difference is in how cpdef functions interact with Cython vs Python > classes. If you want to override a cpdef method in a *Python* subclass then > you > must use def (of course). But in a *Cython* subclass, you must use cpdef. If > you accidental

Re: [sage-devel] Confusing documentation in element.pyx

2014-12-15 Thread David Roe
The difference is in how cpdef functions interact with Cython vs Python classes. If you want to override a cpdef method in a *Python* subclass then you must use def (of course). But in a *Cython* subclass, you must use cpdef. If you accidentally use def instead in a Cython class then it doesn't

[sage-devel] Confusing documentation in element.pyx

2014-12-09 Thread Clemens Heuberger
In http://www.sagemath.org/doc/reference/structure/sage/structure/element.html#how-to-define-a-new-element-class , I find the descriptions of _add_ very confusing, because it seems to imply that there are two versions of _add_, one of them as a "def" and the other as a "cpdef". As it stands, it