+1 I think that we need to enforce some sort of coding standards, but I don't think we should be blindly enforcing conventions that are out of date. Take a look at the 'official' Sun code conventions [1] and note that they haven't been updated since April 20, 1999. I particularly like the reason for the 80-column line [2]... 'Avoid lines longer than 80 characters, since they're not handled well by many terminals and tools.'
[1] http://java.sun.com/docs/codeconv/ [2] http://java.sun.com/docs/codeconv/html/CodeConventions.doc3.html#313 Kevin Sutter wrote: > > A few interesting points with this discussion... > > 1) Under what conditions should we be checking in these generated > metamodel > classes? We don't want the metamodel classes generated for all of our > Entities, do we? Seems like a waste of build time. So, how are we > determining which Entities should have generated metamodel classes? > > 2) Our processing in this arena can help define a "best practice" for our > eventual JPA 2.0 users. That is, how and when to do the metamodel > generation. How to package these generated classes for the application. > And, whether these should be checked into SVN or not. > > 3) And, why do we continue to have the 80 column limit for our source > files? :-) It seems so archaic. > > Kevin > -- View this message in context: http://n2.nabble.com/80-column-line-width---auto-generated-model-classes-tp3025746p3030641.html Sent from the OpenJPA Developers mailing list archive at Nabble.com.
