release 0.4.0

[suhabe]

It would be helpful to bump gqlgen to 0.15.0 and to release a new version of genqlient.

[benjaminjkraft]

Thanks for the poke! I've been meaning to cut a release but haven't gotten around to it. Some things we need to do:

• Land Allow the use of typename to be combined with bind. #138 (edit: or more likely, land something to explicitly disallow the interaction, now Reject the use of both typename and bind #172), as it will be a breaking change if it comes after a release (cc @csilvers)
• Ideally fix On main, there is no error if the schema file cannot be found #146, since that's a sorta-regression
• Ideally if someone at Khan or another big codebase using genqlient has a chance to test recent main branch, that would be great (cc @csilvers @dnerdy @StevenACoffman)
• Fill in changelog (I think several things are missing)
• Do the release!

[benjaminjkraft]

By the way, what's the reason to bump gqlgen? It's not really used by the production-facing parts of genqlient, so I'm not sure if there's a reason for us to force folks to upgrade.

[csilvers]

Land Allow the use of typename to be combined with bind. #138 (or decide we're not going to)

I think last discussion I was leaning toward "decide we're not going to" -- I'm fine with that.

Ideally if someone at Khan or another big codebase using genqlient has a chance to test recent main branch, that would be great.

We're currently using a52e556, I can try using main. We're in crunch time porting graphql-gateway to go (!) but I'll see if I can squeeze this in.

[csilvers]

I think last discussion I was leaning toward "decide we're not going to" -- I'm fine with that.

In particular, I think the best course is to check for typename with bind and explicitly disallow it (which work has not yet been done).

[benjaminjkraft]

Ok, thanks! I won't get to do the release until at least next week anyway, and if you don't have time it's not a huge deal since I think at least two other users have tested #155 which is the biggest change since a52e556.

[csilvers]

I did manage to test the new genqlient on our codebase. All the tests pass at least! We won't be able to try it out in production until monday at the earliest though.

[csilvers]

@benjaminjkraft : we've been using the new genqlient in production for a few days now with no reported issues.

[benjaminjkraft]

Thanks! I'll get on the rest of the items above tomorrow if I have time, or next week if not!