On Feb 11, 2009, at 2:58 PM, Florent Hivert wrote: > Dear All, > > The patch is close to be ready to submission. However I'm not > sure if the > two last answer was from this mail or another one. > >> As you suggested in your previous e-mail, there is a last chance >> that can save >> us: what about creating a new exception say MatrixFormatError >> which inherits >> both ArithmeticError and ValueError. It has the advantage of >> preserving the >> backward compatibility, to be explicit, and if needed one can >> later on issue a >> deprecation warning and remove the inheritance from ArithmeticError. >> >> class MatrixFormatError(ArithmeticError, ValueError): >> def __init__(self, value): >> self.value = value >> def __str__(self): >> return repr(self.value)+"(Warning: the inheritance of >> MatrixFormatError from AritmeticError is deprecated)" >> >> But it's probably not worth it... What do you think ? > > I had a (visual) chat with Nicolas on that point, and I end up > with a > better proposal. There are a lot of places trough sage where two > size has to > match. Here are some examples: > - the number of row and column of one or two matrix(ces) for > multiplication, > inversion, determinant, characteristic polynomial... > - the number of row and column and the number of variables of a > polynomial > when a matrix acts on polynomials. > - the size of two permutations to multiply them. > - The size of a permutations and the number of variable of a > polynomial. > ... > > It seems that all those case are similar and should raise an specific > error. Maybe a SizeMismatchError. We should use the previous trick of > inheriting both from ArithmeticError and Value error to temporarily > assure > backward compatibility. And change the thing definitely during a X. > 0 release > as suggested by Martin. > > I can also put my patch as such (only raise Arithmetic Error for > matrices) and > prepare another one for the next X.0 release.
I would rather just raise an arithmetic error (which seems more appropriate than a ValueError) rather than creating a new one, unless there's a clear benifit to making a specific SizeMismatchError. - Robert --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
