Sequel vs. RDBI vs. DBI

So it looks like RDBI hasn’t been touched in a while and DBI
appears dormant, so I’ve been looking at Sequel . I like what
I’ve seen so far and I’m just wondering if this would be
considered “the way to do database programming in ruby”
(if there is one) ?

I moved from DBI to RDBI a while back and would like to
stick with something that’ll be around and supported
along with the language.

Thanks,


Jim H.
IT Security Engineer
UF Information Technology
Office of Information Security and Compliance

Jim H. wrote in post #1099800:

So it looks like RDBI hasn’t been touched in a while and DBI
appears dormant, so I’ve been looking at Sequel . I like what
I’ve seen so far and I’m just wondering if this would be
considered “the way to do database programming in ruby”
(if there is one) ?

I moved from DBI to RDBI a while back and would like to
stick with something that’ll be around and supported
along with the language.

Thanks,


Jim H.
IT Security Engineer
UF Information Technology
Office of Information Security and Compliance

I’m not prepared to state that Sequel is THE way to do database
programming in Ruby". But it’s a solid choice and is continuously
updated. I use it daily in a variety of environments: SQL Server or
SQLite; desktop or server; MRI or JRuby.

David

I would really suggest using Sequel before (or instead of) ActiveRecord.
Its smaller, lighter, and (if you’re used to sql) closer to the bone.

I found I had serious problems trying to understand Sequel’s syntax
after using ActiveRecord, and the lack of #find* helpers like
ActiveRecord
has threw me for a loop.

As I started digging deeper into Sequel, and started throwing out
ActiveRecord it started getting a bit easier, but I do have to say that
working with Sequel was far harder than it had to be because of hitting
ActiveRecord first. I got spoiled with the easier syntax and fooled into
thinking it was the ‘better’ of the two. I got that totally wrong.

I really like Sequel compared to A/R, but just today I’ve hit a problem
that as near as I can tell is a problem with Sequel: It fails when
attempting to use transactions with MS-SQL Server (which sadly, is the
database I have to use).

I can use ‘isql’ from the Linux command line and do transactions with
SQL-Server just fine. I can use Sequel with transactions against
PostgreSQL, but Sequel and transactions against SQL-Server… no-uh-uh…

I’m very bummed about this because I really liked everything else I’ve
seen, but this is likely a deal-breaker for me.

+1 for Sequel. Use it in production with several apps and have
absolutely no complaints.

Bryan

Joe Meirow wrote in post #1124887:

I really like Sequel compared to A/R, but just today I’ve hit a problem
that as near as I can tell is a problem with Sequel: It fails when
attempting to use transactions with MS-SQL Server (which sadly, is the
database I have to use).

I can use ‘isql’ from the Linux command line and do transactions with
SQL-Server just fine. I can use Sequel with transactions against
PostgreSQL, but Sequel and transactions against SQL-Server… no-uh-uh…

I’m very bummed about this because I really liked everything else I’ve
seen, but this is likely a deal-breaker for me.

Sequel fully supports transactions on most adapters that connect to
Microsoft SQL Server (tinytds, jdbc/sqlserver, jdbc/jtds, odbc). I
think the only adapter that doesn’t support them is ado with the default
provider, since that doesn’t provide a consistent connection object (a
problem with win32ole, not with Sequel).

Thanks,
Jeremy

A happy retraction. I’ve found the source of my problem and now can “+1”
Sequel.

(FWIW, I was not setting the :db_type to ‘mssql’ in the connection)