Re: Mail indexer

[Nils Erik Svangård <nilserik gmail com>]

OK,
I'll try to learn mono by making a licq backend. I'll probably borrow
tons of lines of code from the other backends, what do I write on the
top of every file I produce. I have checked the source and everything I
have seen i copyright novell. Since Im not an employee there I probably
should put something else there :)
/nisse

On 9/20/05, Joe Shaw <joeshaw novell com> wrote:

Hi,

On Tue, 2005-09-20 at 20:07 +0200, Nils Erik Svangård wrote:
> I think I get it. If the fileformat for the licq logs is some strange
> format I probably have to write a filter for that. And then make a
> backen who uses that filter?

Essentially, yes.  If I were you, though, I would start with the
backend, since that's where the data will be coming from to start.

I don't know anything about licq, so I'll make up a hypothetical
situation.  Let's say that licq stores its logs in ~/.licq/logs/<icq
number> and that there is a icq number to nickname mapping in
~/.licq/buddy.xml.

Your backend would look in ~/.licq/logs for new/updated logs, and then
look in ~/.licq/buddy.xml to map them to a human readable name.  You
then create an Indexable object and set certain properties on it,
including the nickname of the buddy who sent it to you.

>From there, the filter would take over and parse the logs for more
general information if it's there.  Now, if the licq logs are very
specialized (ie, not plain text, but not HTML or some other standard
format) you can skip the filtering process altogether and just do all of
the work in the backend.

Joe