Re: beagle implementation in C

I'm immensely excited by this news - even after some serious work on the beagle team's part, the current version in Feisty is still quite a lot larger in terms of resources that anyone would like for a daemon. I do think it's a shame that one of C#'s flagship products is switching, but it seems a lot of momentum has left that initial rush towards Mono anyway.

One small criticism: surely "Ceagle"?

Carl van Tonder

On Sun, 2007-04-01 at 12:21 -0400, Debajyoti Bera wrote:
Hey all,
	Recently there has been an increase of beagle memory related and crash bugs. 
It is increasingly getting difficult to fix such bugs so I (we?) have decided 
to re-implement beagle in C. That will allow us to do precise memory 
allocation and de-allocation leading to better memory behaviour. Also, we 
will be able to debug using good old gdb.
	Implementing the whole of beagle in C will be extremely tedious. Fortunately, 
there exists a C port of lucene (clucene, IIRC) that we can make use of. As 
of now, I will go ahead and implement the core parts. Depending on how it 
goes, the filters and backends will be implemented in C or some scripting 
language (not decided between python and ruby).
	I plan to have a working implemenation at by 06/31/07. Watch the space. I will 
probably also blog about it (if possible with figures comparing regular and 
	This will cut down my time with the regular beagle but I will be checking the 
bugs and mails. It will be a lot of work (and its a bit late for GSoC) so a 
any volunteers will be very helpful.

- dBera

Attachment: signature.asc
Description: This is a digitally signed message part

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