Skip to content


Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP


Korma is not simple #64

technomancy opened this Issue · 8 comments

7 participants


It complects the generation of SQL and connection pooling.

This causes problems like #62; I would like to use the SQL generation facilities without running into bugs in the hidden connection pooling implementation that can't be disabled.


See also #56 (a hackish way to define connection pool) and #33.

In general, I think that korma should not have code to create connection pool at all. Or at least creating the connection pool should be extra option, and the default should be not to create connection pools. Creating a connection pool is not a job for ORM like/sql generator library.


I agree that connection pooling should be a separate library. Digging through the code, it seems that the :pool value for a korma db is simply passed to jdbc/with-connection. The latest jdbc/with-connection is already flexible enough to recognize various types of database definitions including connection pooling instances.

Having the connection pooling built in doesn't seem to add anything and only makes it hard to work around if you don't need or want it.


Agree. jdbc/with-connection already accepts a variety of db-spec. For example, it accepts a simple DATABASE_URL string, whereas korma requires us to parse the string. This makes it awkward to work with Heroku's environment variable based workflow. Please fix this, make it possible to establish a connection using jdbc db-spec directly.


I can give this a whirl if it's not already happened?


Name suggestions for the connection pooling library?


korma-0.3.0-RC1 includes changes to help with this. Your feedback is needed to help make the API ideal for the most people.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Something went wrong with that request. Please try again.