So if you would like to get in touch, then please do so before the 20th August 2010. After then Beki and I will be getting married (21st August) and going on honeymoon (22nd August – 5th September). I shall be picking up calls and receiving emails again from the 6th September 2010. Of course, you can send me an email during the time that I’m away, but I shan’t answer until I’m back.

Into married-life! It is a very exciting time for me!

SELECT * FROM `information_schema`.`COLUMNS` WHERE TABLE_SCHEMA = "databasename" and COLUMN_NAME = "columnname";

Yes, there is a hidden database schema called “information_schema“, which contains all the metadata from all of your database schemas on that mysql server. Within information_schema is a table called COLUMNS which has details about the columns of each table. By matching the TABLE_SCHEMA variable with your DBN and the COLUMN_NAME with the name that you think it is you’ll find the tables that you’re looking for. This could also quite easily be a LIKE comparator. Yay for metadata analysis!

To give you an example. In Drupal, each node (essentially an object of any type to put it in Object-Oriented terminology) has an nid (roughly equivalent to a universal Object ID to put it in OO terms again). This is called nid in all tables, but there are no explicit foreign key declarations. So we can do this to find all occurrences of the “nid” column, like this:

SELECT * FROM `information_schema`.`COLUMNS` WHERE TABLE_SCHEMA = "drupal" and COLUMN_NAME = "nid";

Some of you may know that Beki and I have recently been looking for a new home, well today we found one but that’s not why I am posting today.

I want to bring to bring to your attention “Property Search Aggregators”. These are websites which grab properties (often both rentals and sales) from various sources (agents and agent indexes) and displays them to the users with some funky search technology and filters. I guess you could call it a “delegator” (to talk in a meta-systems perspective, or even an object-oriented perspective).

These four are of particular interest:

1. Nestoria: an amazingly easy-to-use interface where you search for a location, and it comes up on a google map. It provides dimension filters such as how many bedrooms, and the price range. Plus there are further trivalued logic filters (yes-either-false), for things like Garden and Garage etc. Incredibly easy to use, and a site I recommend greatly. Nestoria is also available for countries other than those in the UK.
2. Globrix: Much more awkward to use than Nestoria, but if you can work through that you’ll find that Globrix is packed with a lot more functionality than Nestoria, with more filters and an interesting “or” based logic search for locations. Quite neat
3. FindAProperty: FindAProperty is advertised in the local newspapers over here in Bristol, England. It is an ok site, not as clean as Globrix and Nestoria, but tends to be a little more up-to-date in terms of availability of properties.
4. PrimeLocation: This aggregator has a nice colour scheme, it works ok, but its a lot better at providing miscellaneous additional information. It also serves a wide variety of other countries.

The one downside to these sites is that they are sometimes not updated with current statuses of the properties. For example, you may find that a rented property listed on one of the above sites is already taken, simply because its not connected to the agents systems in an effective fashion. What we ended up doing was searching on the above sites, looking at the agents sites to see if there were any further information (or an updated status), and also looking at the other properties on the agents websites. That seemed to work quite well, and that’s how we got our new home.

Thanks for reading!

Daniel

“John Smith” <jsmith@example.com>
Harry Bo <hbo@example.org>
Lillian Frog <lfrog@example.co.uk>
<tbell@example.net>

And your task is to get rid of everything that isn’t the email itself?

Well look no more! I have a simple sed command which removes everything up to and including the left arrow for each line. Simply run this linux/unix command on a filename and give it an output filename, press enter, and bobs-your-uncle you have a new file with no names at the start.

sed 's/.*<//g' filename.txt > filename.done.txt

You will still have to open up a file and do a find and replace on the right arrow, but it saves some time! I did this for a file containing lots of email addresses a few months ago, and thought that I’d share it with the world. Hope it helps others out there.

It is true that I used to have a blog called “Daniels Blog“. I hadn’t posted for a while, and it seemed to become a bit redundant.  It will still continue to run in the background, but I’m treating it a bit like an archive now. It will be replaced with this blog, the Vanir Systems Blog.

The mission of this blog is:

1. To update everybody interested with the work done by Vanir Systems, this will be using the “vanirsystems” category.
2. To highlight interesting bits of technical news, along with my opinion, using the “technical” category. This may be on the topic of the Semantic Web / Linked Data, or Artificial Intelligence, Programming or Web Development.
3. To highlight interesting bits of general news, along with my opinion, using the “interesting” category. This may be on the topic of religion, spirituality, theology, philosophy, psychology, or a mixture of science and technology.
4. To provide code snippets in various programming languages, using the “codesnippets” category.
5. To provide reviews or previews of products (hardware, software, etc), using the “reviews” category. If you have anything that you would like me to (p)review then please to get in touch, and I am sure that we may be able to work out a deal.

It will attempt to:

1. Use relevant inbound tags
2. Use outbound links to relevant information
3. Be 100% valid XHTML that is reasonably POSH – wish me luck (maybe with sneaky bits of RDFa or Microformats)

Thank you for reading, and I hope you join me here again soon.

Daniel