On Nov 3, 8:59 am, Florent Hivert <florent.hiv...@univ-rouen.fr> wrote: > Hi there, > > > > On Mon, Nov 01, 2010 at 05:30:55AM -0500, Jason Grout wrote: > > On 11/1/10 12:23 AM, William Stein wrote: > > >http://sagemath.blogspot.com/2010/10/how-to-referee-sage-trac-tickets... > > > Nice. I'll have more comments later, but I did notice that this: > > > from sage.misc.misc import deprecation > > deprecation("function_name is deprecated, using new_function_name instead!") > > > came out in formatting wrong (the second line looks like this in my > > firefox (3.6.12, osx snow leopard): > > > deprecation("function_name is deprecated, using new > > > Also, you may want to put in the Sage version number to give people > > a clue about when to remove the function (see the second argument to > > deprecation). > > Actually there is a more handy tools for this precise purpose namely > deprecated_function_alias: > > sage: from sage.misc.misc import deprecated_function_alias > sage: g = deprecated_function_alias(number_of_partitions, > ... 'Sage Version 42.132') > sage: g(5) > doctest:...: DeprecationWarning: (Since Sage Version 42.132) g is > deprecated. Please use number_of_partitions instead. > 7 > > This also works for methods:: > > sage: class cls(object): > ... def new_meth(self): return 42 > ... old_meth = deprecated_function_alias(new_meth, > ... 'Sage Version 42.132') > sage: cls().old_meth() > doctest:...: DeprecationWarning: (Since Sage Version 42.132) old_meth > is deprecated. Please use new_meth instead. > 42 > > Cheers, > > Florent
Very nice - I wasn't aware of that decorator. I have been working with a few things regarding decorators in Sage, so I noticed some shortcomings with deprecated_function_alias: 1) Following Trac #9907, deprecated_function_alias should perhaps be moved to sage.misc.decorators. 2) Without the wraps-decorator, deprecated_function_alias will seem to swallow all introspection information on the decorated callable (actually, sage_wraps should then be used to allow for some Sage- specific introspection, this in turn requires Trac #9919). 3) Currently, any decorator (even one using sage_wraps) will disrupt documentation in Sphinx. This is fixed in #9976 and requires no change in the decorators, however. Perhaps a Trac should be opened to fix these things under the assumption of #9907 and #9919, but I don't have time to do it myself right now (I'm procrastinating terribly as it is)... -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
