Weeknotes 2021 WK 3

Released Django 3.2a1

After a couple of weeks building up, I put out the Django 3.2 alpha on Tuesday.

I had one issue left to finish reviewing, which was to allow using non-pytz timezone implementations, such as Python 3.9’s zoneinfo (and backports).

Current plan is to make zoneinfo the default timezone implementation for Django in v4.0. We’ll provide a compatibility shim, which you’ll be able to opt into, if you’re using pytz specific features, that will be removed in Django 5.0.

The issue this week was just to let folks push on if they want to. There may be a few rough edges, but if this is you, you’re automatically in I know what I’m up to club, so… 🤷‍♀️

Big thanks to Paul Ganssle (Python core contributor, and zoneinfo author) for his help in putting this issue on our radar and helping to push it forward.

Please do download the alpha and give it a run against your projects. This pre-release period is vital to ensure that we ship as high a quality release as possible. See the Django 3.2 release notes for what’s new.

Recording for TalkPython

After Will & I went on PythonBytes, which I think was the first public mention of Button, Micheal (Kennedy) was gracious enough to invite us to go on Talk Python to talk about Django Deployments.

We recorded that this week. It was good fun, like a Django Chat away game, but with Will and me moving to a fortnightly schedule, given the ongoing restrictions around the pandemic, and then the holidays and new year, we’ve not actually spoken that much, so it was nice to catch up.

We talked a bit about the Django release process, Will’s superhero like ability to keep his books updated for the latest version, and then deployment.

It was quite nice to get into a bit of depth about why I favour the approach I do, scaling, and where Button fits on the spectrum of deployment options. Michael brought-up the You are not Google article, which I’d forgotten about, and will add (back) to the repertoire.

The episode will be out in a couple of weeks, I think. (I should know the number, but don’t… 🤦‍♂️)

Preparing Early Access for Button

Since the new year, I’ve allocated Saturday mornings as dedicated time to work on Button. I’m calling this #ButtonSaturday — still nascent hashtag — and I’m very much enjoying it.

As I wrote previously, since the pandemic struck it’s been difficult to find the time to make progress. Scheduling #ButtonSaturday has been a boon. I’m only three weeks in but, seeing the various bits I’ve been pottering at come together as I can allocate time to them is, like, a massive out-breath. (Similar, in a different realm was to be observed on Twitter this Wednesday/Thursday, and beyond.)

Current status is the crucial pre-shipping phase of pulling together what works, and working out what can be postponed to be able to release something minimal, yet valuable.

I’m working towards early access around mid-February, a more substantial beta around the end of March, and then filling in the pieces for what I’d think of as a feature complete towards the end of the year.

The spectrum here is from more to less opinionated. The first version is Do it THIS way!. Later versions will offer more options, but that’s not the priority. (The whole reason Button exists is that there are already too many options.)

I’m going to have to start doing more than just the weekly post. I’m having so much fun with Swift, I’d like to talk about that. And I need to do a whole series around Introducing Button — the mailing list likely best for that initially.

As ever on a solo-project, slowly, slowly: there’s only so-much that I get done in the hours.