This may be nitpicking, but I noticed that the source files refer to a non-standard license ("JDBM License 1.0"), whereas all documentation indicates license is LGPL. I guess LGPL is now used, and the former license is just a historical leftover?
If so, would it make sense to update source files, or at least the main-level public ones?
A related issue is that there seems to have been bit of cut'n paste copying. Specifically, RecordManagerFactory has comments from RecordManager. It would be nice to update the description (it's pretty easy to see what it does, but for javadocs it'd be good to have accurate description).
Finally, there are a few javadocs warnings that would be easy to fix. Would anyone be interested in diffs if I fix those? (along with updated descriptions perhaps)
-+ Tatu +-
Cees de Groot
Usually Alex is quick to respond but he's probably doing something useful like being on holiday ;-).
I wrote JDBM, at least the initial version, but then left the Java world for greener pastures (Smalltalk), so I've been just keeping an eye on the project to see whether anyone uses it, but nothing really active (all the really hard work, like the Btree stuff, has been done by others).
However, seeing that it is still being used; and seeing that there aren't a lot of bug reports, I think it is time to declare this package stable and get a 1.0 release out.
As it's my 'baby', I could and maybe should do this myself, but I'm totally without a Java development environment so it could take a loooong time before I get around doing this (there's so many stuff on my todo list).
So if someone wants to take the lead here, and the other users who hang around on the forum don't object to that someone taking the lead, I'm more than happy to grant him/her developer access to the project. Even though I'm not using it, it still is my 'baby' and I'd like it to succeed...
Sounds good. I can help with easy stuff (I have diffs for javadocs, and can submit some simple performance improvement fixes), but obviously someone more experienced should drive the release.
Oh, also, thank you for your work: it truly is a useful piece of code. I like its compactness, seems like a nice place to look for the basic DB building blocks.