Walter Bright <newshound2@digitalmars.com> wrote in news:ian2ou$ceg$1 @digitalmars.com:

> Fixed now.

Now all you need to do is comment out line 14 of style.css. :-)

On 29/10/10 18:35, Walter Bright wrote:
> This is primarily a bug fix release.
>
> http://www.digitalmars.com/d/1.0/changelog.html
> http://ftp.digitalmars.com/dmd.1.065.zip
>
> http://www.digitalmars.com/d/2.0/changelog.html
> http://ftp.digitalmars.com/dmd.2.050.zip

This bug was also fixed.

http://d.puremagic.com/issues/show_bug.cgi?id=4441

On Fri, 29 Oct 2010 10:35:27 -0700, Walter Bright wrote:

> This is primarily a bug fix release.
> 
> http://www.digitalmars.com/d/1.0/changelog.html http://ftp.digitalmars.com/dmd.1.065.zip
> 
> http://www.digitalmars.com/d/2.0/changelog.html http://ftp.digitalmars.com/dmd.2.050.zip


Thanks to all contributors for yet another good release.  My personal favourites this time must be the relaxed purity rules (yay, now I can actually use pure), the improvements to Tuple (finally, proper indexing!), and the fixing of bug 4465 (which may seem trivial, but which I've been running into several times a day for a long time).

-Lars

On Fri, 29 Oct 2010 10:35:27 -0700, Walter Bright wrote:

> This is primarily a bug fix release.
> 
> http://www.digitalmars.com/d/1.0/changelog.html http://ftp.digitalmars.com/dmd.1.065.zip
> 
> http://www.digitalmars.com/d/2.0/changelog.html http://ftp.digitalmars.com/dmd.2.050.zip


I would like to point out that the table in the documentation for std.container is still messed up.  Walter, could this be because you are building the docs using win32.mak?  When I build with posix.mak it looks fine.

-Lars

On 03.11.2010 13:29, Lars T. Kyllingstad wrote:
> On Fri, 29 Oct 2010 10:35:27 -0700, Walter Bright wrote:
>
>> This is primarily a bug fix release.
>>
>> http://www.digitalmars.com/d/1.0/changelog.html
>> http://ftp.digitalmars.com/dmd.1.065.zip
>>
>> http://www.digitalmars.com/d/2.0/changelog.html
>> http://ftp.digitalmars.com/dmd.2.050.zip
>
>
> Thanks to all contributors for yet another good release.  My personal
> favourites this time must be the relaxed purity rules (yay, now I can
> actually use pure), the improvements to Tuple (finally, proper
> indexing!), and the fixing of bug 4465 (which may seem trivial, but which
> I've been running into several times a day for a long time).
>
> -Lars

Speaking of fancy pure. When will will the std lib actually start to use it ? I tried to use pure lately but as soon as i used phobos methods i hit a wall. e.g why is std.string.format not pure ? i did not look into it but in my pov it does not change any state and does just return a value depending on the given arguments.

Stephan

On Wednesday, November 03, 2010 06:59:11 Lars T. Kyllingstad wrote:
> On Fri, 29 Oct 2010 10:35:27 -0700, Walter Bright wrote:
> > This is primarily a bug fix release.
> > 
> > http://www.digitalmars.com/d/1.0/changelog.html http://ftp.digitalmars.com/dmd.1.065.zip
> > 
> > http://www.digitalmars.com/d/2.0/changelog.html http://ftp.digitalmars.com/dmd.2.050.zip
> 
> I would like to point out that the table in the documentation for std.container is still messed up.  Walter, could this be because you are building the docs using win32.mak?  When I build with posix.mak it looks fine.
> 
> -Lars

An unfortunate side effect of how ddoc deals with conditional compilation is that any module which wants to produce the same documentation with different versions needs to make sure that it has the some ddoc comments for each version and possibly adding stubs for functions which are platform-specific just so you can have ddoc comments on them. std.container probably hasn't done that properly.

- Jonathan M Davis

Stephan wrote:
> On 03.11.2010 13:29, Lars T. Kyllingstad wrote:
>> On Fri, 29 Oct 2010 10:35:27 -0700, Walter Bright wrote:
>>
>>> This is primarily a bug fix release.
>>>
>>> http://www.digitalmars.com/d/1.0/changelog.html
>>> http://ftp.digitalmars.com/dmd.1.065.zip
>>>
>>> http://www.digitalmars.com/d/2.0/changelog.html
>>> http://ftp.digitalmars.com/dmd.2.050.zip
>>
>>
>> Thanks to all contributors for yet another good release.  My personal
>> favourites this time must be the relaxed purity rules (yay, now I can
>> actually use pure), the improvements to Tuple (finally, proper
>> indexing!), and the fixing of bug 4465 (which may seem trivial, but which
>> I've been running into several times a day for a long time).
>>
>> -Lars
> 
> Speaking of fancy pure. When will will the std lib actually start to use it ? I tried to use pure lately but as soon as i used phobos methods i hit a wall. e.g why is std.string.format not pure ? i did not look into it but in my pov it does not change any state and does just return a value depending on the given arguments.

Most development of Phobos is done with the last released version of DMD, not the version under development. So you'll almost never see Phobos using features from the compiler it is released with.

On 11/03/2010 04:38 PM, Don wrote:
>
> Most development of Phobos is done with the last released version of
> DMD, not the version under development. So you'll almost never see
> Phobos using features from the compiler it is released with.

Doesn't that seem a bit sub-optimal to you?

On Friday, November 05, 2010 17:10:57 Jeff Nowakowski wrote:
> On 11/03/2010 04:38 PM, Don wrote:
> > Most development of Phobos is done with the last released version of DMD, not the version under development. So you'll almost never see Phobos using features from the compiler it is released with.
> 
> Doesn't that seem a bit sub-optimal to you?

Perhaps. But it would be harder to keep Phobos stable if the compiler kept changing under it while it was being worked on.

- Jonathan M Davis

On Wed, 03 Nov 2010 10:19:27 -0700, Jonathan M Davis wrote:

>> I would like to point out that the table in the documentation for std.container is still messed up.  Walter, could this be because you are building the docs using win32.mak?  When I build with posix.mak it looks fine.
>> 
>> -Lars
> 
> An unfortunate side effect of how ddoc deals with conditional compilation is that any module which wants to produce the same documentation with different versions needs to make sure that it has the some ddoc comments for each version and possibly adding stubs for functions which are platform-specific just so you can have ddoc comments on them. std.container probably hasn't done that properly.

Not sure that's the problem, it seems to be only the table cells declared with $(TDNW ...) that are missing.

http://d.puremagic.com/issues/show_bug.cgi?id=5103