[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Gnucap-devel] waves branch and wave copy constructor
From: |
al davis |
Subject: |
Re: [Gnucap-devel] waves branch and wave copy constructor |
Date: |
Sat, 7 Mar 2015 14:01:45 -0500 |
User-agent: |
KMail/1.13.7 (Linux/3.2.0-4-amd64; KDE/4.8.4; x86_64; ; ) |
On Saturday 07 March 2015, Felix Salfelder wrote:
> Hi Al.
>
> On Sat, Mar 07, 2015 at 11:14:23AM -0500, al davis wrote:
> > Look at the unstable branch. There is an updated
> > dispatcher. This issue is not fixed there, but it explains
> > what is happening.
>
> ok, i missed these commits. :/
>
> > Proper fix is that the simple one is the only one.
>
> i do not see how the current constructor could work for
> non-static dispatchers (as it does not initialize _map
> "cleanly").
>
> are you suggesting to unconditionally initialize _map, and
> shortcut/remove the link-order sanity check eventually (and
> cross fingers instead)?
yes, in constructor
sanity check could be only in install()
> somehow i think _map should be just a map (rather than map*)
> and the sanitycheck should be done another way...
probably.
sanity check could be a flag, but then it would not still work.
The old implementation works even if construction order is
improper, which would be lost with the change you suggest.
Not sure if that is bad or not.