On Saturday, July 16, 2016 at 1:28:32 AM UTC+8, Evan Digby wrote:
>
> I can't reproduce this in go playground (yet), but under what
> circumstances would/could/should:
>
> nss := []namespace.Namespace{
> append(ns, g2message.NamespaceWin...),
> append(ns, g2message.NamespaceSpend...),
> }
>
> Act differently than:
>
> ns1 := append(ns, g2message.NamespaceWin...)
> ns2 := append(ns, g2message.NamespaceSpend...)
>
> nss := []namespace.Namespace{
> ns1,
> ns2,
> }
>
> ns is a parameter on the function (actually, it's a parameter on the
> function that generated the function that is being called, and ns is a
> closure), and is currently:
>
> namespace.New("localhost")
>
> The definitions in the "namespace" and "g2message" packages are:
>
> type Namespace []string
>
> var NamespaceWin namespace.New("win") // New is making Namespace{"win"}
> var NamespaceSpend namespace.New("spend")
>
> I would think under no circumstances; however, that's not my experience.
>
> Result of fmt.Println(nss) on the first example (expected):
>
> [[localhost win] [localhost spend]]
>
> Result of fmt.Println(nss) on the second example:
>
> [[localhost spend] [localhost spend]]
>
I can't reproduce it also. Can I confirm the input ns is totally the same
for the two callings?
>
> It's not running in a loop, separate goroutine, or anything. Those lines
> are all that change between the two results. I'm at a loss. I will keep
> trying to reproduce in go playground.
>
>
> Thanks,
>
>
> Evan
>
--
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
