13:11 <GitHub9> [sequel] jeremyevans pushed 5 new commits to master: http://git.io/2PJNQg

13:11 <GitHub9> [sequel/master] Add a server_block extension for scoping database access inside a block to specific shards - Jeremy Evans

13:11 <GitHub9> [sequel/master] Add another connection pool spec for reentrant #hold - Jeremy Evans

13:11 <GitHub9> [sequel/master] Allow Database#server_opts to take a Hash - Jeremy Evans

13:40 <GitHub53> [sequel] jeremyevans pushed 1 new commit to master: http://git.io/m4LVpw

13:40 <GitHub53> [sequel/master] Add example code in RDoc to ease integration of server_block/arbitrary_servers extensions - Jeremy Evans

18:39 <wardrop> Does sequel have an escape method, for escaping strings to make them safe to use as value in a query? I know TinyTDS has such as method, but I was hoping Sequel would have a portable method that works across databases. Does anyone know?

18:39 <wardrop> Btw, I aware of the string substituion, but sometimes it's more conveniant to escape inline, than pass in named parameters.

18:40 <jeremyevans> wardrop: #literal is the method you want

18:40 <jeremyevans> wardrop: It's defined on both Dataset and Database#

18:41 <jeremyevans> wardrop: Just FYI, Sequel uses escaping by default, it doesn't use parameters unless you specifically set that up

18:42 <wardrop> jeremyevans: Thanks Jeremy. Yes, it's only in the context of dynamic query building that I have the requirement for #literal (which is exactly the method I was after, thanks).

18:43 <wardrop> I use Sequel's own query methods wherever possible, only falling back to raw SQL where it either makes the code much clearer, or where there's no other option.

21:06 <pvh> jeremyevans: i would like you to create a new library which manages migrations and schema changes

21:07 <pvh> jeremyevans: and i would like you to name that library Trilogy

23:10 <jeremyevans> pvh: OK. What's the rationale?

23:58 <pvh> jeremyevans: first, the name sells itself

23:58 <pvh> jeremyevans: second, the actual process of migrating databases is insanely perilous

23:58 <pvh> jeremyevans: there's so much domain knowledge required. for example, setting a "default" on a column causes a table rewrite

23:59 <pvh> how would a developer ever find that out except by failing?