michaeldiamond has a proposal to replace the polymorphism currently
used for literalizing Sequel::SQL::* objects with a conditional
implementation.  I'd like to get some thoughts on this:

http://github.com/michaeldiamond/sequel/commit/4b9c4f327007be99f78d9ceee5d60c3ace1ab666

Personally, I think it does make the internals cleaner, at the expense
of flexibility and ease of extension.  Currently, it's fairly easy for
a user (or future extension) to add their own Sequel::SQL::Expression
subclass, and just add a literalizing method to the Sequel::Dataset
class.  This change will make it difficult to do that, as you'd need
to override the literal method, and since it is in the class itself
(instead of a module included in the class), that's not easy to do
cleanly without aliasing (which I try to avoid).

I'm willing to hear other's opinions on this, but I'm currently
leaning against it.  Other than making the internals a little cleaner,
can anyone else think of advantages to this approach?

Jeremy
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"sequel-talk" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/sequel-talk?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to