Hi Ron,
On 19/8/16 20:24, Ron Savage wrote:
Hi Nigel
On 20/08/16 01:51, Tim Bunce wrote:
On Fri, Aug 19, 2016 at 10:09:45AM -0400, Nigel Horne wrote:
On 8/19/16 9:56 AM, Tim Bunce wrote:
On Fri, Aug 19, 2016 at 09:30:32AM -0400, Nigel Horne wrote:
[snip]
But I wonder about the name. "DBD::XML" seems to be a bold name,
implying that it's _the_ DBI interface for data stored in XML files.
Of course the same kind of issue applies to many other drivers,
so it's not a major concern, but does seem worth dicussing.
I'm more than happy to entertain other names if you have any
suggestions.
As a rule, we have DBI (there's just one) and anything it uses
(DBI::*), and then we have DBIx::* for all our additions in thefield.
That is, we all agree to explicitly avoid using DBI::*. Even drivers
loaded by DBI are called DBD::Pg, DBD::mysql, DBD::SQLite, etc.
This usage of $ABCx or $AbcX is a wide-spread convention. E.g.
Marpa::* is exclusively used by Jeffrey Kegler, and all ours are
MarpaX::*. But there are many, many modules called ${Something}x::* as
add-ons to $Something.
And yes, for DBI and other it's 'x', elsewhere it's often 'X'.
For clarification, are you saying that my driver should be called
DBDx::XML? The reason I ask is that all the DBDx::* modules I could
find seem to be extensions to DBI or things that use DBI, rather than
DBI drivers or backends.
Also would that confuse with the existing DBIx::XML code? That does
something entirely different.
-Nigel
