[Mono-dev] Patch for some obscure cases in Dictionary (small)
Andreas Nahr
ClassDevelopment at A-SoftTech.com
Fri Jun 13 11:45:08 EDT 2008
OK I found some way to unit-test this. Unfortunately it only tests the
incompatibility but not the problems that result because of it.
Happy Hacking
Andreas
> -----Ursprüngliche Nachricht-----
> Von: mono-devel-list-bounces at lists.ximian.com [mailto:mono-devel-list-
> bounces at lists.ximian.com] Im Auftrag von Andreas Nahr
> Gesendet: Freitag, 13. Juni 2008 17:17
> An: 'Juraj Skripsky'
> Cc: 'mono-devel-list'
> Betreff: Re: [Mono-dev] Patch for some obscure cases in Dictionary
> (small)
>
> Thanks for testing. If nobody from the maintainers objects I'll commit.
> I've been looking for a testcase, but the error is not easily
> reproducible
> without unmanaged code.
>
> Greetings
> Andreas
>
> > -----Ursprüngliche Nachricht-----
> > Von: Juraj Skripsky [mailto:js at hotfeet.ch]
> > Gesendet: Freitag, 13. Juni 2008 14:32
> > An: Andreas Nahr
> > Cc: 'mono-devel-list'
> > Betreff: Re: [Mono-dev] Patch for some obscure cases in Dictionary
> > (small)
> >
> > Hi Andreas,
> >
> > Your changes look good to me, it makes the code cleaner and leaner. I
> > was the one who wrote (or rather re-wrote) the code you're changing,
> > but
> > I guess you need the okay-to-commit from one of the corlib
> > maintainer(s).
> >
> > - Juraj
> >
> >
> > On Thu, 2008-06-12 at 09:50 +0200, Andreas Nahr wrote:
> > > Re-sent without input file.
> > > Please test the patch and tell me if it is ok to apply. Does not
> seem
> > to
> > > break anything for me, but did not test very extensively.
> > > You can get the input file from
> > > http://shootout.alioth.debian.org/download/knucleotide-input.txt
> > >
> > > Andreas
> > >
> > > > While trying to optimize the k-nucleotide benchmark from Computer
> > > > Language Shootout (or whatever) I stumbled across a corner case
> in
> > > > Dictionary.
> > > > Situation is: We are behaving different from MS.Net and current
> > Mono
> > > > behavior is somewhat illogical. But the case itself is surely
> > outside
> > > > of the spec ;) We are changing the indexer itself in the indexer
> > (base
> > > > problem, concrete is a bit more complex).
> > > >
> > > > Anyways this could be patched (it's slightly faster, too).
> > > >
> > > > Happy Hacking
> > > >
> > > > P.S. Am attaching the benchmark application and the input for it
> (<
> > at
> > > > console)
> > > >
> > > > P.P.S. I AM aware that doing stuff this way is asking for trouble
> > ;)
> > > _______________________________________________
> > > Mono-devel-list mailing list
> > > Mono-devel-list at lists.ximian.com
> > > http://lists.ximian.com/mailman/listinfo/mono-devel-list
>
> _______________________________________________
> Mono-devel-list mailing list
> Mono-devel-list at lists.ximian.com
> http://lists.ximian.com/mailman/listinfo/mono-devel-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DictionaryTest.patch
Type: application/octet-stream
Size: 1111 bytes
Desc: not available
Url : http://lists.ximian.com/pipermail/mono-devel-list/attachments/20080613/7cff3db8/attachment.obj
More information about the Mono-devel-list
mailing list