FAQ
I have just noticed a change in either testing or math/rand between
go1.2.1 and tip that causes one of my packages to now fail its tests.

Does anyone know anything about what this might be?

To replicate:

go get code.google.com/p/biogo.cluster/kmeans
go test code.google.com/p/biogo.cluster/kmeans

k-means depends on rand, so I tried seeding to 1 just in case this was
causing a problem, but no change - still passes 1.2 and fails tip.

--
You received this message because you are subscribed to the Google Groups "golang-nuts" group.
To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Search Discussions

  • Dan Kortschak at Mar 20, 2014 at 7:07 am
    This appears to be non-monotonic; releases do not exhibit the behaviour
    and unfortunately hg doesn't seem to be able to bisect the problem
    reporting "starting revisions are not directly related".
    On Thu, 2014-03-20 at 17:10 +1030, Dan Kortschak wrote:
    I have just noticed a change in either testing or math/rand between
    go1.2.1 and tip that causes one of my packages to now fail its tests.

    Does anyone know anything about what this might be?

    To replicate:

    go get code.google.com/p/biogo.cluster/kmeans
    go test code.google.com/p/biogo.cluster/kmeans

    k-means depends on rand, so I tried seeding to 1 just in case this was
    causing a problem, but no change - still passes 1.2 and fails tip.

    --
    You received this message because you are subscribed to the Google Groups "golang-nuts" group.
    To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscribe@googlegroups.com.
    For more options, visit https://groups.google.com/d/optout.
  • Dan Kortschak at Mar 20, 2014 at 8:59 am
    Found it. The non-monotonicity in the bisection was due to that commit
    not being included in release as far as I can see.

    The first bad revision is:
    changeset: 18668:3ed9d5c72102
    user: Jeff R. Allen <jra@nella.org>
    date: Wed Dec 18 15:38:53 2013 -0500
    summary: math/rand: Float32/64 must only return values in [0,1)


    --
    You received this message because you are subscribed to the Google Groups "golang-nuts" group.
    To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscribe@googlegroups.com.
    For more options, visit https://groups.google.com/d/optout.

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupgolang-nuts @
categoriesgo
postedMar 20, '14 at 6:41a
activeMar 20, '14 at 8:59a
posts3
users1
websitegolang.org

1 user in discussion

Dan Kortschak: 3 posts

People

Translate

site design / logo © 2021 Grokbase