While currently too busy at work to deliver much Open Source contributions, let's debunk an old habit of PostgreSQL extension authors. It's all down to copy pasting from contrib, and there's no reason to continue doing $libdir this way ever since 7.4 days.

 

Introducing Extensions

October, 21 2010

After reading Simon's blog post, I can't help but try to give some details about what it is exactly that I'm working on. As he said, there are several aspects to extensions in PostgreSQL, it all begins here: Chapter 35. Extending SQL.

 

In trying to help an extension debian packaging effort, I've once again proposed to handle it. That's because I now begin to know how to do it, as you can see in my package overview page at debian QA facility. There's a reason why I proposed myself here, it's that yet another tool of mine is now to be found in debian, and should greatly help extension packaging there. You can already check for the postgresql-server-dev-all package page if you're that impatient!

 

Yes. This pgloader project is still maintained and somewhat active. Development happens when I receive a complaint, either about a bug in existing code or a feature in yet-to-write code. If you have a bug to report, just send me an email!

 

prefix 1.1.0

November, 30 2009

So I had two bug reports about prefix in less than a week. It means several things, one of them is that my code is getting used in the wild, which is nice. The other side of the coin is that people do find bugs in there. This one is about the behavior of the btree opclass of the type prefix range. We cheat a lot there by simply having written one, because a range does not have a strict ordering: is [1-3] before of after [2-4]? But when you know you have no overlapping intervals in your prefix_range column, being able to have it part of a primary key is damn useful.

 

prefix 1.0.0

October, 06 2009

So there it is, at long last, the final 1.0.0 release of prefix! It's on its way into the debian repository (targetting sid, in testing in 10 days) and available on pgfoundry to.

 

Escreen integration

September, 22 2009

After having used elscreen for a long time, I'm now a very happy user of escreen, which feels much better integrated and allows to have one ring of recently visited buffers per screen. Which is what you need when using a screen like feature, really.

 

The function didn't allow for using more than one mailrc file, which isn't a good idea, so I've just added that. Oh and for gnus integration what I need is (add-hook 'message-mode-hook 'mail-abbrevs-setup) it seems... so that if I type the alias it'll get automatically expanded. And to be real lazy and avoid having to type in the entire alias, mail-abbrev-complete-alias to the rescue, assigned to some easy to reach keys.

 

Improving ~/.mailrc usage

September, 07 2009

So I've been adviced to use ~/.mailrc for keeping a basic address book in Emacs, for use within gnus for example. I had to resort to the manual to find out how to use the file aliases when I need them, that is when composing a mail. For the record, here's what I had to do:

 

It seems like debian developers are back from annual conference and holiday, so they have had a look at the NEW queue and processed the packages in there. Two of them were mines and waiting to get in unstable, hstore-new and preprepare.

 

At long last, here it is. With binary versions both for postgresal-8.3 and postgresal-8.4! Unfortunately my other packaging efforts are still waiting on the NEW queue, but I hope to soon see hstore-new and preprepare enter debian too.

 

prefix 1.0~rc2-1

July, 09 2009

I've been having problem with building both postgresql-8.3-prefix and postgresql-8.4-prefix debian packages from the same source package, and fixing the packaging issue forced me into modifying the main prefix Makefile. So while reaching rc2, I tried to think about missing pieces easy to add this late in the game: and there's one, that's a function length(prefix_range), so that you don't have to cast to text no more in the following wildspread query:

 

At long last, after millions and millions of queries just here at work and some more in other places, the prefix project is reaching 1.0 milestone. The release candidate is getting uploaded into debian at the moment of this writing, and available at the following place: prefix-1.0~rc1.tar.gz.

 

The prefix project is about matching a literal against prefixes in your table, the typical example being a telecom routing table. Thanks to the excellent work around generic indexes in PostgreSQL with GiST, indexing prefix matches is easy to support in an external module. Which is what the prefix extension is all about.