At the moment, I have no idea what it is. There are scattered vague reports.

Can anyone with a verifiable regression from 2.057 please post it here.

On Mon, 13 Feb 2012, Walter Bright wrote:

> At the moment, I have no idea what it is. There are scattered vague reports.
> 
> Can anyone with a verifiable regression from 2.057 please post it here.

Er.. in bugzilla, with a test case.

> On Mon, 13 Feb 2012, Walter Bright wrote:
>
>> At the moment, I have no idea what it is. There are scattered vague reports.
>>
>> Can anyone with a verifiable regression from 2.057 please post it here.
> Er.. in bugzilla, with a test case.

Right.

http://d.puremagic.com/issues/show_bug.cgi?id=7498

On 02/13/2012 02:44 PM, Walter Bright wrote:
> At the moment, I have no idea what it is. There are scattered vague reports.
>
> Can anyone with a verifiable regression from 2.057 please post it here.

On 2/13/2012 4:50 PM, Ellery Newcomer wrote:
> http://d.puremagic.com/issues/show_bug.cgi?id=7498
>
> On 02/13/2012 02:44 PM, Walter Bright wrote:
>> At the moment, I have no idea what it is. There are scattered vague reports.
>>
>> Can anyone with a verifiable regression from 2.057 please post it here.
>

The report does not say or verify it is a regression.

At this stage in the 2.058 release, the only thing that should hold it up are verifiable regressions from 2.057.

On Mon, 13 Feb 2012 21:44:28 +0100, Walter Bright <walter at digitalmars.com> wrote:

> At the moment, I have no idea what it is. There are scattered vague reports.
>
> Can anyone with a verifiable regression from 2.057 please post it here.
> _______________________________________________
> dmd-beta mailing list
> dmd-beta at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-beta

With respect to being a successor of 2.057 the changelog looks impressive.
I'm only aware of one further regression in phobos.
http://d.puremagic.com/issues/show_bug.cgi?id=7476
So this would be a fairly good release for RELEASE (2.057) users.

With respect to being the release of the current development version going
back and forth between the import behavior will cause people to fix code
they
have written since January because of reintroduced compiler bugs.
So this is going to be a disappointing release for HEAD users.

             | 2.058b4 | 2.058b4 + quickfixes | 2.058b4 + real fixing
---------------------------------------------
2.057 users |    :|   |         :|           |         :(

  HEAD users |    :@   |         :(           |         ;)

   dmd       |    :|   |         :(           |         :|



Doing a real fix would take about 2-3 days for implementation/testing and 1-2 weeks beta.

     Import bugs: 313,314,625,1504,2401,2991,3254,5411,5412,6307,6554,7496
Protection bugs: 143,1161,1238,1441,1754,2225,2775,2830,6180

On 2/14/2012 6:22 AM, Martin Nowak wrote:
> On Mon, 13 Feb 2012 21:44:28 +0100, Walter Bright <walter at digitalmars.com> wrote:
>
> With respect to being a successor of 2.057 the changelog looks impressive.
> I'm only aware of one further regression in phobos.
> http://d.puremagic.com/issues/show_bug.cgi?id=7476
> So this would be a fairly good release for RELEASE (2.057) users.
>
> With respect to being the release of the current development version going
> back and forth between the import behavior will cause people to fix code they
> have written since January because of reintroduced compiler bugs.
> So this is going to be a disappointing release for HEAD users.
>
>             | 2.058b4 | 2.058b4 + quickfixes | 2.058b4 + real fixing
> ---------------------------------------------
> 2.057 users |    :|   |         :|           |         :(
>
>  HEAD users |    :@   |         :(           |         ;)
>
>   dmd       |    :|   |         :(           |         :|
>
>
>
> Doing a real fix would take about 2-3 days for implementation/testing and 1-2 weeks beta.
>
>     Import bugs: 313,314,625,1504,2401,2991,3254,5411,5412,6307,6554,7496
> Protection bugs: 143,1161,1238,1441,1754,2225,2775,2830,6180
>

The big problem with the import fixes was the inadequate test suite, which failed to reveal that the fixes fundamentally did not work.

On Tue, 14 Feb 2012, Martin Nowak wrote:

> On Mon, 13 Feb 2012 21:44:28 +0100, Walter Bright <walter at digitalmars.com> wrote:
> 
> > At the moment, I have no idea what it is. There are scattered vague reports.
> > 
> > Can anyone with a verifiable regression from 2.057 please post it here.
> 
> With respect to being a successor of 2.057 the changelog looks impressive.
> I'm only aware of one further regression in phobos.
> http://d.puremagic.com/issues/show_bug.cgi?id=7476
> So this would be a fairly good release for RELEASE (2.057) users.
> 
> With respect to being the release of the current development version going
> back and forth between the import behavior will cause people to fix code they
> have written since January because of reintroduced compiler bugs.
> So this is going to be a disappointing release for HEAD users.
> 
>            | 2.058b4 | 2.058b4 + quickfixes | 2.058b4 + real fixing
> ---------------------------------------------
> 2.057 users |    :|   |         :|           |         :(
> 
> HEAD users |    :@   |         :(           |         ;)
> 
>  dmd       |    :|   |         :(           |         :|
> 
> 
> 
> Doing a real fix would take about 2-3 days for implementation/testing and 1-2 weeks beta.
> 
>    Import bugs: 313,314,625,1504,2401,2991,3254,5411,5412,6307,6554,7496
> Protection bugs: 143,1161,1238,1441,1754,2225,2775,2830,6180

While it's unfortunate that we've had to pull some fixes back due to bugs with the implementation, the correct way to view progress is from release to release.  The interum state is much less relevant to many more people.

It's a little disappointing that so much time passed between committing the first round of changes and reporting of the regressions or we might have been able to fix them during the release cycle, but I have no regrets about backing off and holding them for the next cycle.

Let's finish getting this release out the door so that the next cycle can start.

On 2/14/2012 1:18 PM, Brad Roberts wrote:
>
> While it's unfortunate that we've had to pull some fixes back due to bugs with the implementation, the correct way to view progress is from release to release.  The interum state is much less relevant to many more people.
>
> It's a little disappointing that so much time passed between committing the first round of changes and reporting of the regressions or we might have been able to fix them during the release cycle, but I have no regrets about backing off and holding them for the next cycle.
>
> Let's finish getting this release out the door so that the next cycle can start.
>

I fully agree.

Added two trello cards for these issues to the 2.059 board, "Fix import bugs" and "Fix protection bugs"? We could add Martin's list of bugs to those, but I'm not intimately familiar with the details, I'll leave that to someone else.

-Steve

>________________________________
> From: Brad Roberts <braddr at puremagic.com>
> 
>On Tue, 14 Feb 2012, Martin Nowak wrote:
>
>> On Mon, 13 Feb 2012 21:44:28 +0100, Walter Bright <walter at digitalmars.com> wrote:
>> 
>> > At the moment, I have no idea what it is. There are scattered vague reports.
>> > 
>> > Can anyone with a verifiable regression from 2.057 please post it here.
>> 
>> With respect to being a successor of 2.057 the changelog looks impressive.
>> I'm only aware of one further regression in phobos.
>> http://d.puremagic.com/issues/show_bug.cgi?id=7476
>> So this would be a fairly good release for RELEASE (2.057) users.
>> 
>> With respect to being the release of the current development version going
>> back and forth between the import behavior will cause people to fix code they
>> have written since January because of reintroduced compiler bugs.
>> So this is going to be a disappointing release for HEAD users.
>> 
>>? ? ? ? ? ? | 2.058b4 | 2.058b4 + quickfixes | 2.058b4 + real fixing
>> ---------------------------------------------
>> 2.057 users |? ? :|???|? ? ? ???:|? ? ? ? ???|? ? ? ???:(
>> 
>> HEAD users |? ? :@???|? ? ? ???:(? ? ? ? ???|? ? ? ???;)
>> 
>>? dmd? ? ???|? ? :|???|? ? ? ???:(? ? ? ? ???|? ? ? ???:|
>> 
>> 
>> 
>> Doing a real fix would take about 2-3 days for implementation/testing and 1-2 weeks beta.
>> 
>>? ? Import bugs: 313,314,625,1504,2401,2991,3254,5411,5412,6307,6554,7496
>> Protection bugs: 143,1161,1238,1441,1754,2225,2775,2830,6180
>
>While it's unfortunate that we've had to pull some fixes back due to bugs with the implementation, the correct way to view progress is from release to release.? The interum state is much less relevant to many more people.
>
>It's a little disappointing that so much time passed between committing the first round of changes and reporting of the regressions or we might have been able to fix them during the release cycle, but I have no regrets about backing off and holding them for the next cycle.
>
>Let's finish getting this release out the door so that the next cycle can start.