On Friday, July 11, 2014 9:36:22 AM UTC-7, Rodrigo Rosenfeld Rosas wrote:
>
>  I'm curious. How do you decide when to stop supporting some Ruby version?
>
> 1.8.7 has reached EOL from the MRI team and it's no longer supported by 
> newer Rails releases for a while already. It also makes it harder for 
> newcomers to contribute code as they might not even know that the new Hash 
> syntax is not supported by Ruby 1.8, for example. And it's also responsible 
> for things that wouldn't make sense otherwise, like the case for overriding 
> Object#id, which no longer is defined since Ruby 1.9.
>

I'll remove support for 1.8.7 when it becomes a burden to support it (with 
proper deprecation, of course).  I see a significant benefit in continuing 
to support it (allowing people running ruby 1.8.7 to update to newer Sequel 
versions), so there would have to be a significant cost before I would 
remove it.  So far, supporting ruby 1.8.7 has not held me back in adding 
any feature I wanted to add, so I haven't seen a significant cost.

Much of the ruby ecosystem still supports ruby 1.8.7, including most of the 
database drivers that Sequel uses.  Current versions of pg, mysql2, 
sqlite3, oci8, ibm_db, and sqlanywhere still support it, for example 
(tinytds dropped support). When the common databases drivers stop 
supporting ruby 1.8.7, that will signal to me that it may be time to drop 
support in Sequel.
 

> Anyway, if I understood correctly it's totally fine for me to add an 
> undef_method(:id) to my model without any side effect on Sequel overall, 
> right?
>

Sequel doesn't call the method internally, so that's probably fine.

Thanks,
Jeremy

-- 
You received this message because you are subscribed to the Google Groups 
"sequel-talk" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sequel-talk.
For more options, visit https://groups.google.com/d/optout.

Reply via email to