Beagle+Xesam: Weekly Report

Hello All,
Sorry about the delay ... here's the update from this week:

With respect to Xesam, the proposal to have a single "field" value per
query ( There are
some other changes there that I will need to implement if agreed upon,
like blocking searches being dropped.

While this progresses, I am now addressing the TextCache as it stands
today. Since the current TextCache includes a lot of small files, we
end up wasting a lot of disk space (each file that is less than a disk
block in size occupies the entire disk block anyway). For example, on
my machine, the space taken is 219 MB, while the size of the data
itself is only 117 MB. I am trying to see if we can find a more
efficient means of storing this data.

I've written some test code to query for the TextCache values of a
number of Uris. I can measure the time taken by profiling (if this is
prohibitive, I can just time the runs).

This next week I should be able to have an alternative implementation
going (I'm grappling with the lack of documentation for Sqlite's C#
bindings right now).

Arun Raghavan

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]