Posted Wednesday, November 28 at 11:33 PM (12 days ago)
This entry marks the release of Publ v0.3.9. It has the following changes:
more_text and related functionality to image sets (an example being visible over here)
- Improved and simplified the caching behavior (fixing some fiddly cases around how ETags and last-modified worked, or rather didn’t)
I also made, and then soon reverted, a change around how entry IDs and publish dates were automatically assigned to non-published entries. I thought it was going to simplify some workflow things but it only complicated the code and added more corner cases to deal with, all for something that doesn’t actually address the use case I was worried about. So never mind on that.
(What happened to v0.3.8? I goofed and forgot to merge the completed
more_text et al changes into my build system first. Oops.)
See below for more on the caching changes.
Posted Wednesday, November 28 at 9:18 AM (12 days ago)
I just released Pushl v0.1.3, which adds some minor performance optimizations and a bug fix.
Originally I was hoping to have a major performance optimization, in the form of having rewritten Pushl from thread-per-connection to async operation, but unfortunately I ran into a bunch of problems with it. Mostly that I was running into a “too many open files” error and I couldn’t figure out what was causing a descriptor leak. I have the work-in-progress branch online if anyone wants to take a look at it.
Anyway, the reason I went down this route is because I added WebSub subscriber support to my fork of Feed-On-Feeds, which makes it so that WebSub-enabled RSS and Atom feeds will push their updates to your reader instead of having to wait for a polling interval.
You can read more about some of my other thoughts on a blog entry that quickly devolves into a rant, if you’re so inclined.
Posted Wednesday, October 24 at 7:59 PM (2 months ago)
I’ve released v0.3.7, which just fixes a few issues around transaction management and overall indexing performance. Namely:
- The indexer locks individual entries as it’s working on them
- If an entry is being worked on, watchdog will ignore it
- Cleaned up a couple of transaction failures that can occur due to PonyORM’s optimistic locking behavior
Posted Saturday, October 13 at 10:01 PM (2 months ago)
I just released v0.3.6 of Publ, which just allows it to work with databases other than SQLite. In particular this is part of testing more advanced heroku deployment options.
Right now I’m primarily focusing on improving the documentation, especially the quickstart guide, since people are finally showing interest in Publ but aren’t quite sure where to begin!
Posted Friday, October 12 at 10:01 PM (2 months ago)
I’ve now released v0.3.5 of Publ. Changelog:
- Add support for listing deleted entries (accessible via
- Improved how the last-modified/etag reference was determined (also fixing a nasty bug where a site might crash if a file is deleted)
- Fixed a minor shaping bug
I’ve also updated the sample site templates with all of the changes that have happened since, uh, June, and also included some sample content so it’s easier to get started with it.
Posted Thursday, October 11 at 5:41 AM (2 months ago)
I’ve gotten Pushl to the point that I’m confident in releasing it as a full alpha and having a cron job run it every 15 minutes. Whew.
I’ve also added some
h-entry markup to the Publ site templates so that hopefully the notifications appear at least somewhat reasonably elsewhere.
I really need to redo these site templates now that I have a better idea of how they go together. Also the quickstart guide could be a lot better.
Posted Tuesday, October 9 at 6:53 AM (2 months ago)
I finally got around to releasing a very rough prototype of Pushl to pypi. It only sends out WebSub notifications for now (does anyone even use those?), but I’ll work on actually implementing WebMention soon.
Also, recently someone pointed out to me fed.brid.gy which makes it easy to turn a static site into an ActivityPub source. At some point I’ll experiment with setting up Publ for this; it looks like it’s just a matter of adding a couple of additional route rules to Publ, so that will probably go into an advanced configuration guide if I ever get around to making such a thing. (Or it could actually be added to Publ directly but there isn’t much of a reason for that, IMO.)
Posted Friday, October 5 at 8:28 AM (2 months ago)
Turns out I never actually tested the
If-Modified-Since handler, because if I had I’d have seen the glaring exception it threw.
Oh well, that’s fixed now. I think.
Posted Tuesday, October 2 at 6:16 AM (2 months ago)
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.
Well, I finally implemented it, and I’m pretty happy with how I did it.
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.
Posted Sunday, September 30 at 4:00 AM (3 months ago)
So, for the last couple of days I’ve been playing with some of the IndieWeb concepts, in particular Webmention. Spurred on by a helpful thread with Kevin Marks, I took some time to actually do a rough implementation of outgoing Webmentions, and also did some of the work to set up the
h-entry microformats on my main site.
As far as I can tell, it works great, but I’m also not going to actually merge this to master or push it to production. Read on to see why!
Posted Tuesday, September 25 at 9:55 PM (3 months ago)
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)
class work correctly per the documentation
- PonyORM bugs:
Posted Friday, September 21 at 5:58 AM (3 months ago)
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:
<img src="/path/to/image.png" width="320" height="320"
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.
Posted Wednesday, September 19 at 7:46 PM (3 months ago)
Version 0.3.0 is now released, with the change from peewee to PonyORM.
As a result of this change you’ll have to do two things to your config file:
- The database configuration format has changed slightly
- Any existing databases have to be manually deleted/dropped/etc.; unfortunately PonyORM doesn’t provide a mechanism for deleting tables not under its control
Everything else should work identically as before.
Posted Wednesday, September 19 at 9:27 AM (3 months ago)
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.
Posted Wednesday, September 12 at 8:27 AM (3 months ago)
Earlier today I pushed v0.2.2, which was a minor configuration fix for the markdown library to support table syntax.
Then I pushed v0.2.2.1 which was another configuration fix to fix that fix (oops).
Then in deploying updates I upgraded to Python 3.7, and promptly discovered that Publ doesn’t actually work on Python 3.7, so just now I fixed that, and released the fixes as v0.2.3.
Posted Wednesday, August 22 at 6:48 AM (4 months ago)
Just a couple of minor fixes in this release:
Posted Friday, July 6 at 10:57 PM (5 months ago)
A few changes since v0.1.24:
- 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)
Posted Thursday, June 28 at 3:01 AM (6 months ago)
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.
Posted Thursday, June 28 at 2:55 AM (6 months ago)
- The image rendition cache now gets periodically purged; the default is to delete renditions which haven’t been used in the last week (this can be disabled)
entry.title can now accept the
no_smartquotes parameter, which is necessary in Atom feeds
entry.card now uses the same Markdown extensions as
Posted Thursday, June 7 at 4:30 AM (6 months ago)
I neglected to mention that I set Publ to beta status in v0.1.22, which was a minor bugfix release, rather than moving to 0.2 like I previously stated. The changes for 0.1.22 were:
- Fixes to category
- Added support for regex path-alias hooks (this is configured on the Python/WSGI side, and has been working quite nicely over on beesbuzz.biz)
- Fixed a dumb bug in the cache-control headers
And then the changes for 0.1.23:
- Enable automatic smart-quote substitutions (this is the default setting, and can be overridden by passing
- Improve the handling of last-modified times on entries (now there’s a
Last-Modified header which only gets set when you want it to be)