joe shaw

Beagle 0.2.3 is out. The big news about this release is that it no longer has the plague of growing memory usage that has been a problem since its inception. I’ll go more into how I tracked that down (using heap-buddy) later.

This release includes a couple new filters: a BMP image filter from Alexander Macdonald; a new video filter which uses mplayer to extract metadata, also from Alexander Macdonald; and a new “external” filter, written by me.

The idea behind the external filter is that you may want to extract data from files — maybe a specialized, local format — but you don’t want to have to learn C# to do it. Today is your lucky day! You’ll need some tool to extract text content from a file. Although Beagle already has PDF filtering, a good example of this is pdftotext from xpdf. You give it a PDF file, and it spits out just the text from the file. To set one of these up, you just add a simple XML blob to /etc/beagle/external-filters.xml:

<filter>
  <mimetype>text/plain</mimetype>
  <extension>.txt</extension>
  <command>cat</command>
  <arguments>%s</arguments>
</filter>

And that’s it! Now any files with mime type “text/plain” or extension “.txt” will be fed into “cat” and the output will be indexed. Of course, we also have plain text indexing too, so that’s not a good example either. But whatever, you get the idea.

All the information you need is included in the external-filters.xml file included.