Marshall Hampton wrote: > I'm not disagreeing, I just don't know how to quickly change that. If > someone can give me some tips I will at least patch the spkgs. > > -Marshall
There may be no quick fix, though the code did not look very large, so I doubt it would be a huge job to do it properly. The stages I would take would be: 1) Remove the copies to /dev/null. 2) Add -Wall to see all warning reported. 3) Build with the warnings displayed. 4) Contact the original author saying you would like to get his/her package into Sage, but some are objecting since the warnings were hidden. Point out what warnings you get. 5) Sort out why the compiler is complaining. That will take some time I expect. Though you have the advantage the package is not very large. Newsgroups are a good place to get help on how best to re-write code in a portable way that avoids warnings. You could ask on gcc-help, saying "this bit of code generates a warning, but I'm not sure why" Dave > On Jul 31, 4:21 pm, William Stein <wst...@gmail.com> wrote: >> On Fri, Jul 31, 2009 at 2:04 PM, Dr. David Kirkby >> <david.kir...@onetel.net>wrote: >> >> >> >> >> >>> Marshall Hampton wrote: >>>> I agree, that doesn't sound good. At the moment, I just want to check >>>> out the sandpile functionality, so I don't think I will wade in and >>>> try to improve glpk, or bug the author to do so. >>>> On the positive side, I think I now have packages that install >>>> correctly, at least on my own mac. They are at: >>>> http://www.d.umn.edu/~mhampton/4ti2.p0.spkg<http://www.d.umn.edu/%7Emhampton/4ti2.p0.spkg> >>>> http://www.d.umn.edu/~mhampton/glpk.p0.spkg<http://www.d.umn.edu/%7Emhampton/glpk.p0.spkg> >>>> i.e. I have overwritten my previous broken versions. >>>> This is also now trac ticket #6663 (http://trac.sagemath.org/sage_trac/ >>>> ticket/6663). >>>> -Marshall >>> I'm not a mathematician, don't have a clue what this does, so I am >>> probably looking at this from a very different point of view to most. >>> But I don't think it's a good idea to include code that hides warnings. >>> Again, it's a personal thing but when I look at web sites, like Wolfram >>> Research's, which has 42 errors: >>> http://validator.w3.org/check?uri=www.wolfram.com&charset=%28detect+a... >>> it always makes me wonder how seriously quality is taken. >>> In contrast the Sage site has zero errors: >>> http://validator.w3.org/check?uri=http%3A%2F%2Fwww.sagemath.org%2F&ch... >>> Mathematicians I've worked worth have always paid a lot of attention to >>> detail - far more than I think engineers tend to. If someone covers up >>> their compiler errors, it makes me wonder whether sufficient attention >>> to detail is applied elsewhere. >>> If someone like WRI, Maplesoft etc wanted to try to point out the >>> disadvantages of Sage, showing how we hide warnings would be like giving >>> them ammunition to blow us up with. >>> I'm sure a lot of people will disagree with me, but personally I would >>> avoid adding things to sage that rely on code that is built like that. >> I agree with you. It is difficult to disagree with such a natural technical >> way to improve quality. >> >> -- William > > > --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
