I'm hoping to target the weekend of the 20th for a freeze and vote, so you might just be able to squeeze it in.
At the very least, some javadoc improvements to the existing MultiMap should be made. Stephen PS. I suggest that when you are ready to discuss actual coding changes, you use the commons-dev mailing list. > from: "Ted M. Young" <[EMAIL PROTECTED]> > What would the deadline be for getting at least some fixes for MultiMap into > 3.0? I don't think the interface is horrible, it certainly suited my exact > purpose and I'd be happy to fix up the JavaDoc and improve the unit tests, > however I'll be on vacation starting tomorrow through 9/14 so I wouldn't be > able to contribute anything until after 9/14. And I could certainly take up > the discussion on design and then implement it. > > If that's too late for 3.0, I'll still look to do it for the next release. > > I think MultipleMap is too closely named to MultiMap, but I'm also open to > suggestions (I'll look at the bugzilla entries and list archives for ideas). > > ;ted > > ----- Original Message ----- > From: "Stephen Colebourne" <[EMAIL PROTECTED]> > To: "Jakarta Commons Users List" <[EMAIL PROTECTED]> > Sent: Thursday, September 04, 2003 5:25 PM > Subject: Re: [collections] MultiMap > > > > Yes there are two bugzilla entries for MultiMap ideas and implementations. > > It was going to miss the 3.0 release, however if you are volunteering to > > help then it could make it in. > > > > I believe that the MultiMap interface itself is weak and needs to be > > replaced (we can't change it as its an interface). MultipleMap is the best > > name I could think of for the replacement, but I'm open to offers. > > > > One or two new implementations of the new interface would then need writing, > > based on the bugzilla entries, together with a brand new test class (ideally > > designed like the other test classes in the framework). > > > > It basically needs someone to commit to discuss this and implement it. If > > thats you, then great!!! > > Thanks > > Stephen > > > > ----- Original Message ----- > > From: "Ted M. Young" <[EMAIL PROTECTED]> > > > I see that there's still an open enhancement issue for the ChainedHashMap > > > implementation (16465, opened Jan 27 2003 and status is still NEW). Will > > this > > > get into the 3.0 release? > > > > > > I'd also love to see some better JavaDocs for MultiMap about order of the > > > collection retrieved via a get(). The MultiMap interface incorrectly > > states > > > that "This implementation uses an ArrayList as the collection." which is > > > impossible to say in an interface and implies a specific ordering of > > values per > > > key. In addition, the MultiHashMap implementation states the same thing, > > but > > > it uses Values, which is an inner class that extends AbstractCollection > > and > > > doesn't say anything (one way or the other) about ordering. > > > > > > Finally, I find the JUnit tests for MultiHashMap to be somewhat obscure. > > I was > > > looking for basic tests on adding values, removing values, removing an > > entire > > > key, etc. Instead I found random strings, etc., that don't help someone > > new to > > > the MultiMap interface. > > > > > > I'd be happy to help out with these things, btw. > > > > > > ;ted > > > > > > -- > > > Internal Data Tools Staff Engineer, > > > eBay, Inc. > > > > > > > > > --------------------------------------------------------------------- > > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > For additional commands, e-mail: [EMAIL PROTECTED] > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
