Matthew Draper

Yep, not sure how much time I'll have in the next week or two, but failing that, I'm sure we can practice some Conference Driven Development. :+1:

> A better approach would be to add this to your db/seed.rb file Data transformation is really not appropriate for seeds, IMO. Depending on application-defined models is unwise, but this...

Sorry, I've been distracted with conference+travel. I haven't looked at the diff yet, but just to cut through some of the meta-conversation: * I want this change for the prepared-statements...

So I've been pondering this, and I find myself drawn in somewhat the opposite direction I pushed with #48261... 🤔 I kinda like this: > ```ruby > Post.with(posts_with_comments: Post.where("comments_count >...

I don't expect those to have the same `references_values`; my expectation would be that `post: { id: post }` will produce a query that needs the posts table joined, whereas...

The original addition was here: * #30007 * #30016 I don't recall whether there is now, or was historically, a difference in how Bundler will resolve a gemfile given the...

Spellings aside, the fact the `file:` syntax locks us so rigidly to a precise version (per 707b5cb60616c86cd554ada2ef8754dd0ab5e72a) is IMO, Quite Unfortunate.

> SELECT queries are idempotent No, they're not. _Most_ select queries are idempotent, but it is not safe for us to assume users aren't running any of the forms which...

The trouble is `User.where("myfunc(id)").load` could be calling a SQL function that modifies the database. (I think MySQL has fewer ways of doing this; other databases have more.) The only time...

My very vague theory is that we might be able to track whether a query is "tainted" by app-controlled query content during the same / a similar walk of the...