I just did some quick scanning of my dmd-internals folder for bug lists (primarily don's, but some from me and others) to take a look for ones that have been discussed as highly desired at various points in time and were still open (though I suppose some might have been dupes of others that were resolved). Some have patches, some have at least pointers at the code involved, etc.

It's far from an exhaustive list, but here's the results (in numerical order for lack of any time spent trying to categorize them):

http://d.puremagic.com/issues/show_bug.cgi?id=314
[module] Static, renamed, and selective imports are always public

http://d.puremagic.com/issues/show_bug.cgi?id=1513 try/catch/finally misbehavior on windows

http://d.puremagic.com/issues/show_bug.cgi?id=2451
Adding structs that use opAssign or postblit to an AA is broken

http://d.puremagic.com/issues/show_bug.cgi?id=2716 Confusion of auto and scope as the class attribute

http://d.puremagic.com/issues/show_bug.cgi?id=3150
cast from dynamic array to ulong is allowed

http://d.puremagic.com/issues/show_bug.cgi?id=3244
with -inline dmd sometime instantiates nested functions that then cannot be accessed

http://d.puremagic.com/issues/show_bug.cgi?id=3516 Destructor not called on temporaries

http://d.puremagic.com/issues/show_bug.cgi?id=3554
Ddoc generates invalid output for documentation comments with non paired parenthesis

http://d.puremagic.com/issues/show_bug.cgi?id=3996
ICE Regression(2.041): Passing struct as AA template parameter (Algebraic with
struct)

http://d.puremagic.com/issues/show_bug.cgi?id=4149
refs displayed as pointers in gdb

http://d.puremagic.com/issues/show_bug.cgi?id=4269
Regression(2.031): invalid type accepted if evaluated while errors are gagged

http://d.puremagic.com/issues/show_bug.cgi?id=4278
undo limitations of bug3500's fix

http://d.puremagic.com/issues/show_bug.cgi?id=4298
Constant array translated to unnecessary array literal creation

http://d.puremagic.com/issues/show_bug.cgi?id=4397 const/CTFE does not work

> http://d.puremagic.com/issues/show_bug.cgi?id=314
> [module] Static, renamed, and selective imports are always public

That's the bug with the most votes in bugzilla.
Also has a patch attached.


To me, forward reference errors are the real PITA.
I recently added 2 very simple cases that are pretty likely to come up in
D code and block my development.


http://d.puremagic.com/issues/show_bug.cgi?id=4503
forward reference to aliased template instance

http://d.puremagic.com/issues/show_bug.cgi?id=4267
forward reference error when 2-fold aliasing a template instance

> http://d.puremagic.com/issues/show_bug.cgi?id=4503
> forward reference to aliased template instance
>
> http://d.puremagic.com/issues/show_bug.cgi?id=4267
> forward reference error when 2-fold aliasing a template instance

4503 now contains a 2-line patch that seems to solve 4267 as well.

 On 25.07.2010 05:02, Brad Roberts wrote:
> http://d.puremagic.com/issues/show_bug.cgi?id=3554
> Ddoc generates invalid output for documentation comments with non paired parenthesis
>
I just posted a patch for this one. Would be great to get this issue fixed.

-- 
Johannes Pfau