phobos commit, revision 2309


user: walter

msg:
fix screwed up formatting

http://www.dsource.org/projects/phobos/changeset/2309

paths changed:
U   trunk/phobos/std/zlib.d

On 1/11/2011 12:32 AM, dsource.org wrote:
> phobos commit, revision 2309
> 
> 
> user: walter
> 
> msg:
> fix screwed up formatting
> 
> http://www.dsource.org/projects/phobos/changeset/2309
> 
> paths changed:
> U   trunk/phobos/std/zlib.d

A quick test build would save no small amount of failed builds and extra submits ya know.  The compiler is pretty good at catching these sorts of problems. :)

std/zlib.d(169): Error: cannot implicitly convert expression (destlen) of type
ulong to uint

Brad Roberts wrote:
> On 1/11/2011 12:32 AM, dsource.org wrote:
> 
>> phobos commit, revision 2309
>>
>>
>> user: walter
>>
>> msg:
>> fix screwed up formatting
>>
>> http://www.dsource.org/projects/phobos/changeset/2309
>>
>> paths changed:
>> U   trunk/phobos/std/zlib.d
>> 
>
> A quick test build would save no small amount of failed builds and extra submits ya know.  The compiler is pretty good at catching these sorts of problems. :)
>
> std/zlib.d(169): Error: cannot implicitly convert expression (destlen) of type
> ulong to uint
>
> 

Yah, well, building 64 bit D2 is not part of my build test script yet. It did work on D1, which allows such implicit casts.

I've set up linux.mak on Phobos1 to build:

    lib32/libphobos.a
    lib64/libphobos.a

so that both can be built without stepping on each other. Can you (Brad, Andrei) do the same for Phobos2?

On 1/11/2011 1:19 AM, Walter Bright wrote:
> 
> 
> Brad Roberts wrote:
>> On 1/11/2011 12:32 AM, dsource.org wrote:
>> 
>>> phobos commit, revision 2309
>>>
>>>
>>> user: walter
>>>
>>> msg:
>>> fix screwed up formatting
>>>
>>> http://www.dsource.org/projects/phobos/changeset/2309
>>>
>>> paths changed:
>>> U   trunk/phobos/std/zlib.d
>>> 
>>
>> A quick test build would save no small amount of failed builds and extra submits ya know.  The compiler is pretty good at catching these sorts of problems. :)
>>
>> std/zlib.d(169): Error: cannot implicitly convert expression (destlen) of type
>> ulong to uint
>>
>> 
> 
> Yah, well, building 64 bit D2 is not part of my build test script yet. It did work on D1, which allows such implicit casts.

So invest the few minutes and update your scripts?  I don't understand your reluctance and I'm not particularly eager to facilitate your avoiding it due to it costing me time when you check in broken changes blindly.

Seriously.. catch up, it'll benefit us all.

Also, it's still broken.. so even if you don't want to automate it right now, would you clean up the mess you left?

I'm probably a little grumpier at this than I should be, but the number of things that have been breaking the build for the last several days is approaching ridiculous proportions.  It's really frustrating that I've been the one pointing them out.  This rant isn't pointed at any one individual, but almost everyone.

Frustrated,
Brad

Brad Roberts wrote:
>
> So invest the few minutes and update your scripts?  I don't understand your reluctance and I'm not particularly eager to facilitate your avoiding it due to it costing me time when you check in broken changes blindly.
>
> Seriously.. catch up, it'll benefit us all.
>
> Also, it's still broken.. so even if you don't want to automate it right now, would you clean up the mess you left?
>
> I'm probably a little grumpier at this than I should be, but the number of things that have been breaking the build for the last several days is approaching ridiculous proportions.  It's really frustrating that I've been the one pointing them out.  This rant isn't pointed at any one individual, but almost everyone.
>
> 

You're right.

Although I was waiting on updating my D2 scripts until posix.mak got fixed to build both libraries without stepping all over itself.

On Tue, 11 Jan 2011, Walter Bright wrote:

> Brad Roberts wrote:
> > 
> > So invest the few minutes and update your scripts?  I don't understand your
> > reluctance and I'm not particularly eager to facilitate your avoiding it due
> > to
> > it costing me time when you check in broken changes blindly.
> > 
> > Seriously.. catch up, it'll benefit us all.
> > 
> > Also, it's still broken.. so even if you don't want to automate it right
> > now,
> > would you clean up the mess you left?
> > 
> > I'm probably a little grumpier at this than I should be, but the number of
> > things that have been breaking the build for the last several days is
> > approaching ridiculous proportions.  It's really frustrating that I've been
> > the
> > one pointing them out.  This rant isn't pointed at any one individual, but
> > almost everyone.
> > 
> > 
> 
> You're right.
> 
> Although I was waiting on updating my D2 scripts until posix.mak got fixed to build both libraries without stepping all over itself.

Thanks.. sorry for being abrupt.

Also, it looks like the alignment changes to dmd have broken at least one of the tests (just judging from the failure output):

core.exception.AssertError at template1(1480): Assertion failure

I'm pretty sure it won't be the only failure.

Later,
Brad

Brad Roberts wrote:
>
> Thanks.. sorry for being abrupt.
> 

No prob. I have a very thick skin!

> Also, it looks like the alignment changes to dmd have broken at least one of the tests (just judging from the failure output):
>
> core.exception.AssertError at template1(1480): Assertion failure
>
> I'm pretty sure it won't be the only failure.
>
> 

The previous alignment clearly was broken. zlib wouldn't interact with its C implementation with it.

On Tue, 11 Jan 2011, Walter Bright wrote:

> Brad Roberts wrote:
> > 
> > Thanks.. sorry for being abrupt.
> > 
> 
> No prob. I have a very thick skin!
> 
> > Also, it looks like the alignment changes to dmd have broken at least one of the tests (just judging from the failure output):
> > 
> > core.exception.AssertError at template1(1480): Assertion failure
> > 
> > I'm pretty sure it won't be the only failure.
> > 
> > 
> 
> The previous alignment clearly was broken. zlib wouldn't interact with its C implementation with it.

I understand _why_ it was changed and _why_ the tests are failing.  That's not the point.  The point is that the tests are failing and need to be fixed as a part of fixing the bug.  Leaving the tests broken means you've left half the job incomplete.

Grumpy still,
Brad

Brad Roberts wrote:
>
> I understand _why_ it was changed and _why_ the tests are failing.  That's
> not the point.  The point is that the tests are failing and need to be
> fixed as a part of fixing the bug.  Leaving the tests broken means you've
> left half the job incomplete.
> 

I know I've made some very poor decisions recently, but I can give you my complete assurance that my work will be back to normal. I've still got the greatest enthusiasm and confidence in the mission. And I want to help you.

> Grumpy still,
> 

Look Brad, I can see you're really upset about this. I honestly think you ought to sit down calmly, take a stress pill, and think things over.