April 08, 2005
It's that time of the DMD life cycle again, when the fractional part of the version number happens to be congruent to 0 modulo 10.  So let's see how the peeves are progressing....

http://www.wikiservice.at/wiki4d/wiki.cgi?PendingPeeves

Finally, age-old bugs are getting fixed.  For once, the last of the attribute overloading bugs seems to be fixed.  I still don't know they managed to act like one bug underlying all the attributes when they were actually separate bugs, but at least that's history now.

Structs can now be used as AA keytypes.  However, a bit of a mess was made in implementing this detail:

http://www.digitalmars.com/drn-bin/wwwnews?digitalmars.D.bugs/2642
http://www.digitalmars.com/drn-bin/wwwnews?digitalmars.D.bugs/2643

And we finally have documentation on the abstract attribute.  And for once, abstract classes are really abstract.  But there are still things to be done here.

http://www.digitalmars.com/drn-bin/wwwnews?digitalmars.D.bugs/1939
http://www.digitalmars.com/drn-bin/wwwnews?digitalmars.D.bugs/1940

So, the 'Bugs in DMD' section of the page is slowly working towards being finally cleaned out.

But the other sections of the pending peeves aren't progressing so much.  OK, so a few things have changed.  We now finally have std.file.copy.  At first it copied just the contents of the file, but now on Windows it copies the file properly.  So for Windows it seems the issue's now wrapped up, but there's still work to be done on the Linux implementation.

http://www.digitalmars.com/drn-bin/wwwnews?digitalmars.D.bugs/3082

And a few months ago Walter decided that array operations are to be removed from the 1.0 spec.

http://www.digitalmars.com/drn-bin/wwwnews?digitalmars.D/15803

So it looks like drawing in old Fortraners is going to wait a while. This also raises the question: to what extent is D still really for numerical programmers?  But the real issue left here is that the feature is still waiting to be removed from the spec.  Until this is done, we still have this feature that's in the language but not in the compiler.


So, I guess that while DMD is still by no means ready for 1.0, at least it's getting somewhere.  So, who knows how many more PPPRs there'll be before DMD 1.0 comes out?

Happy programming everyone!

Stewart.

-- 
My e-mail is valid but not my primary mailbox.  Please keep replies on the 'group where everyone may benefit.