Alex Rousskov wrote:
> SourceLayout: adaptation/{icap,ecap}, take 1
> 
> Moved src/ICAP into src/adaptation/icap.
> Moved src/eCAP into src/adaptation/ecap.
> 
> Renamed ICAP source files from ICAPFoo.{cc,h} to Foo.{cc,h}.
Still not sure what the effects on windows will be.
But with recently clearing up the automake issues, the non-windows are fine.
> 
> Placed ICAP names into Adaptation::Icap namespace, renaming ICAPFoo to
> Adaptation::Icap::Foo.
> ------------------------------
> 
> I am posting this for review and to provide context for the following
> question.
> 
> Before ICAP and eCAP directories were moved into adaptation/, we had
> Adaptation and Ecap namespaces. We now have
> 
> 	Adaptation
> 	Ecap
> 	Adaptation::Icap
> 
> This intermediate state is inconsistent. What should I implement as the
> final set of namespaces? The choices are
> 
> A) Flat: Adaptation, Ecap, Icap
>    This option makes most adaptation names "shorter".
> 
> B) Scoped or nested: Adaptation, Adaptation::Icap, Adaptation::Ecap.
>    This option better reflects the nested directory structure and scope.
> 
> 
> When answering this question, please keep in mind that we will
> eventually face similar questions when polishing authentication and
> storage code:
> 	Auth, Auth::Basic, Auth::Negotiate versus Auth, Basic, Negotiate
> 	Fs, Fs::Diskd, Fs::Coss versus Fs, Diskd, Coss
> 
> I tend to favor (B). What do you think?
Yes on changing the current state.
Yet to find anything problematic. So +-0 from me on the choice. Nesting 
to match the directory include path sounds okay and 'feels right'.
bb:tweak   debugs needs to use HERE macro
Amos
-- Please be using Current Stable Squid 2.7.STABLE6 or 3.0.STABLE13 Current Beta Squid 3.1.0.5Received on Wed Feb 25 2009 - 09:05:55 MST
This archive was generated by hypermail 2.2.0 : Thu Feb 26 2009 - 12:00:04 MST