I’ve started working on Pushl in earnest now, and one thing that was really bugging me about this is that anything which polls feeds and entries would really benefit from having client-side cache control working. Which was a big missing feature in Publ.
The short version: for any given view it figures out (pessimistically) what’s the most recent file that would have affected the view (well, within reason; it only looks at the current template rather than any included templates, which is pretty difficult to do correctly) and uses that to generate an ETag (via metadata fingerprint) and a Last-Modified time (based either on the file modification time or the time the entry was actually published).
There’s probably a few corner cases this misses but in general this makes client-side caching of feeds and such work nicely.
I found a few more annoying bugs that were shaken out from the whole PonyORM transition, as well as a couple of bugs in the new shape functionality. There’s probably a few more of these bugs lurking in the codebase (I mean, in addition to the existing bugs I know about), but here’s what’s changed:
Image shape bugs:
Fix some FileNotFound handling on images (so shape errors propagate correctly)
Make img_class and class work correctly per the documentation
Did you know that CSS3 has a style called shape-outline? It’s pretty neat, it makes it so that a floated object gets a shape based on the alpha channel of its specified image. But it’s kind of a pain to set up; in plain HTML it looks something like this:
and if you want a different shape mask for your image than its own alpha channel, you have to do a bunch of stuff like making sure that the image sizes are the same and whatever.
For a number of reasons, I have replaced the backing ORM. Previously I was using peewee, but now I’m using PonyORM. The primary reason for this is purely ideological; I do not want to use software which is maintained by someone with a track record of toxic behavior. peewee’s maintainer responds to issues and feature requests with shouting and dismissive snark; PonyORM’s maintainer responds with helpfulness and grace. I am a strong proponent of the latter.
PonyORM’s API is also significantly more Pythonic, and rather than abusing operator overloads for clever query building purposes, it abuses Python’s AST functionality to parse actual Python expressions into SQL queries. Seriously, look at this explanation of it and tell me that isn’t just amazing.
Updated code to use the current Flask cache-control API
Only set cache-control for responses that don’t have a natural cache response
Entry IDs and UUIDs are now semi-stably generated, in order to prevent (or at least reduce) problems like the last time
Publ itself is stable enough (and enough has changed since v0.1.0) that I felt that a minor version bump was a reasonable thing to do.
Anyway! While Publ has been running quite nicely on my website, I’d love to see more people actively using and developing it. This site in particular needs a lot of attention and probably reworking; my other top priorities are:
A better installation/deployment guide
Proper test coverage (rather than manual smoke tests)
So, sorry to anyone who was subscribed to the RSS feed for this and got spammed with v0.1.24 release announcements. I made a mistake and pushed a version of the entry that didn’t have a canonical ID assigned yet, and as a result, every time Heroku spun up, it assigned a new ID. This is something that’s happened before and I really ought to do something about it.
Three things come to mind:
Figuring out how to always make IDs get assigned in an idempotent manner (hard to do correctly)
Don’t run on Heroku so the assignments persist between executions (easy)
Add a pre-push hook to the repo that verifies that all entries alread have an assigned ID (???)
2 seems like the easiest approach for now, so that’s what I’ll probably do.
This server is now running Publ v0.1.21, which I believe will be the final beta version; all of the release blockers have been taken care of. And as a bonus I also implemented the ability to include template fragments, which allows templates' content to change formatting based on the category of the formatted entries. This is particularly useful on Atom feeds, where you might want to have different entry formats for each category. I’m already using it on busybee, to provide full-content feeds in ways that make sense for the different site sections.
I will play with this for a few days and if I’m happy with it I’ll be releasing v0.2.0, which will be the first beta release. In the meantime I’ll also look into packaging up busybee’s templates for other people to use as a starting point, because I know that setting up a new site right now can be incredibly overwhelming!
The amount of stuff I’m having to fix in Publ to support beesbuzz.biz is diminishing rapidly! Here’s what’s happened since 0.1.18:
Improved the Path-Alias redirection logic; now it will do a 301 Permanently Moved for inbound Path-Aliased requests, and if a Path-Alias points to an entry with a Redirect-To it will redirect directly to that URL instead (and it will be a 302, same as the old Redirect-To behavior)
Pagination can now be weekly; you can use entry.archive(paging='week'), and a ?date= view parameter ending in _w will provide a weekly view instead.
Better default formatting for view.range, and an addition of a week format parameter there
So aside from the Dreamhost issues, I would like to share what’s new in the latest version as part of my big “get my website online” push:
PERFORMANCE: Improved the threading mechanisms around image renditions for better stability and performance
BUGFIX: Made markup tagging work consistently between image types
FEATURE: view.link() now allows overriding category
FEATURE: You can now mark an entry with an Entry-Status of DELETED or GONE, which results in a 410 error instead of a 404 (be the envy of your web-developer nerd friends!)
FEATURE: Error templates will automatically use the x00 error code as a potential fallback (e.g. error code 503 will also fall back to a template for 500)
FEATURE: Entry titles can now have Markdown in them, and it usually works most of the time!
BUGFIX: Now when running in debug mode you don’t end up with two watchdog threads
FEATURE/QUALITY: Refactored the way error pages are handled, and now if you get a 404-type error on a page while the index is being asynchronously scanned, it’ll turn into a 503 with a Retry-After disposition in case it’s just something that hasn’t been indexed yet
UX: View pagination URLs now all use id as the query parameter rather than a miasma of contextually-dependent start, last, or first which made no sense anyway
One of the overarching reasons I decided to build Publ the way I did was in order to take advantage of Dreamhost’s support for Passenger WSGI. I was expecting that to be the primary means of hosting my main site (which is way too big for a Heroku instance) and given how smoothly things were working with this site on Dreamhost I figured it wouldn’t be a big deal.
However, there was a huge monkey wrench thrown into things when I switched my site’s configuration over to Passenger; despite all of my configuration being exactly the same between publ.beesbuzz.biz and beesbuzz.biz, the rendition cache on beesbuzz.biz was getting its permissions set wrong, and there was some rather weird behavior with how it was making the temporary files to begin with.
In investigating this I attempted to upgrade my packages on publ.beesbuzz.biz, and all h*ck broke loose.
Oh, it’s been a little while since I’ve posted an update, hasn’t it? That’s
because I’ve been very busy building the templates for my personal site! In
doing so I’ve greatly improved the way that view refinements worked, fixed a few
silly bugs with image search paths on templates, and also added an easier way to
specify background images in CSS, via the image().get_css_background() method.
I’m really excited to be able to bring my first fully-realized Publ site to the
public; I hope it gets other people interested in what a flexible publishing
system allows them to do!
These things are already making my new personal site look way better and easier
to use! I feel like I’m almost ready to flip the switch.
Oh, and I also improved the getting started guide, including
adding basic setup instructions for Linux and Windows. Not that I’ve gotten Publ
to run on Windows, yet, but documenting how to get the environment set up is
the first step, right?
There’s an old joke in programming, that the two hardest things to do are naming things, cache invalidation, and off-by-one errors. But this doesn’t pay sufficient respect to one of the other hardest things, namely handling date and time.