Great, so how should we go on? I suggest we wait for you, David, so that you can make the code a little more stable and change the things you mentioned. You said something about two weeks (?) I would say we should then be at a point where we could get rid of de.lanlab.* packages and move the rest to something like org.apache.larm and then put it into the sandbox. We should also check if performance is a problem, especially with those factory methods. Within this time we should also review the docs and adapt the LARM-speak (MessageHandler? MessageListener? MessageProcessor? Stage? Storage?)
After this time I would like to concentrate on two things: The next thing I would like to do is to break up FetcherTask into at least two pieces (move parsing out) and change Messages such that they contain lists of URLs. This means StoragePipeline becomes a ProcessingPipeline. The other big issue I would like to take care of is the URLVisitedFilter/URLVisitedManager/URLSeenFilter. Its RAM usage must be optimized. I already have some ideas for that. That's only my part; I know that Otis, Kelvin and Peter wanted to work on other parts. May I suggest we all become familiar with David's work and read our docs once again? Btw, I also learned a lot from your code, David. Cheers, Clemens --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
