phobos commit, revision 1919


user: SHOO

msg:
Added stopwatch module

Added StopWatch, Ticks, systime, apptime, comparingBenchmark, measureTime.

http://www.dsource.org/projects/phobos/changeset/1919

Still unhappy about the addition of a module with a name narrow enough to preclude any further extensions.

Andrei

On 08/22/2010 10:11 AM, dsource.org wrote:
> phobos commit, revision 1919
>
>
> user: SHOO
>
> msg:
> Added stopwatch module
>
> Added StopWatch, Ticks, systime, apptime, comparingBenchmark, measureTime.
>
> http://www.dsource.org/projects/phobos/changeset/1919
>
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

Me too.  IMHO, it should either be part of a new std.datetime module together with the std.gregorian stuff, or std.stopwatch should be renamed to std.timer, std.timing, or something like that.

-Lars



On Sun, 2010-08-22 at 23:00 -0500, Andrei Alexandrescu wrote:
> Still unhappy about the addition of a module with a name narrow enough to preclude any further extensions.
> 
> Andrei
> 
> On 08/22/2010 10:11 AM, dsource.org wrote:
> > phobos commit, revision 1919
> >
> >
> > user: SHOO
> >
> > msg:
> > Added stopwatch module
> >
> > Added StopWatch, Ticks, systime, apptime, comparingBenchmark, measureTime.
> >
> > http://www.dsource.org/projects/phobos/changeset/1919
> >
> > _______________________________________________
> > phobos mailing list
> > phobos at puremagic.com
> > http://lists.puremagic.com/mailman/listinfo/phobos
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

Suggestions:

1. Put it inside std.date, though I agree that the name isn't suggestive (it does include time-related stuff as of today).

2. Create std.datetime from std.gregorian and stopwatch, and when we're done with the calendar stuff (did you guys see the recent related posts?) we deprecate std.date.

3. Create std.time, put stopwatch in it, and later move std.gregorian in std.date.

Opinions? Shoo, what's your preference?


Andrei

On 08/23/2010 04:12 AM, Lars Tandle Kyllingstad wrote:
> Me too.  IMHO, it should either be part of a new std.datetime module together with the std.gregorian stuff, or std.stopwatch should be renamed to std.timer, std.timing, or something like that.
>
> -Lars
>
>
>
> On Sun, 2010-08-22 at 23:00 -0500, Andrei Alexandrescu wrote:
>> Still unhappy about the addition of a module with a name narrow enough to preclude any further extensions.
>>
>> Andrei
>>
>> On 08/22/2010 10:11 AM, dsource.org wrote:
>>> phobos commit, revision 1919
>>>
>>>
>>> user: SHOO
>>>
>>> msg:
>>> Added stopwatch module
>>>
>>> Added StopWatch, Ticks, systime, apptime, comparingBenchmark, measureTime.
>>>
>>> http://www.dsource.org/projects/phobos/changeset/1919
>>>
>>> _______________________________________________
>>> phobos mailing list
>>> phobos at puremagic.com
>>> http://lists.puremagic.com/mailman/listinfo/phobos
>> _______________________________________________
>> phobos mailing list
>> phobos at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/phobos
>
>
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

Deprecate std.perf, too.

Andrei Alexandrescu wrote:
> Suggestions:
>
> 1. Put it inside std.date, though I agree that the name isn't suggestive (it does include time-related stuff as of today).
>
> 2. Create std.datetime from std.gregorian and stopwatch, and when we're done with the calendar stuff (did you guys see the recent related posts?) we deprecate std.date.
>
> 3. Create std.time, put stopwatch in it, and later move std.gregorian in std.date.
>
> Opinions? Shoo, what's your preference?
>
>
> Andrei
>
> On 08/23/2010 04:12 AM, Lars Tandle Kyllingstad wrote:
>> Me too.  IMHO, it should either be part of a new std.datetime module together with the std.gregorian stuff, or std.stopwatch should be renamed to std.timer, std.timing, or something like that.
>>
>> -Lars
>>
>>
>>
>> On Sun, 2010-08-22 at 23:00 -0500, Andrei Alexandrescu wrote:
>>> Still unhappy about the addition of a module with a name narrow enough to preclude any further extensions.
>>>
>>> Andrei
>>>
>>> On 08/22/2010 10:11 AM, dsource.org wrote:
>>>> phobos commit, revision 1919
>>>>
>>>>
>>>> user: SHOO
>>>>
>>>> msg:
>>>> Added stopwatch module
>>>>
>>>> Added StopWatch, Ticks, systime, apptime, comparingBenchmark, measureTime.
>>>>
>>>> http://www.dsource.org/projects/phobos/changeset/1919
>>>>
>>>> _______________________________________________
>>>> phobos mailing list
>>>> phobos at puremagic.com
>>>> http://lists.puremagic.com/mailman/listinfo/phobos
>>> _______________________________________________
>>> phobos mailing list
>>> phobos at puremagic.com
>>> http://lists.puremagic.com/mailman/listinfo/phobos
>>
>>
>> _______________________________________________
>> phobos mailing list
>> phobos at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/phobos
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos
>
>

I'm sorry about what I committed without consultation.

I was troubled with a module name, too.
I think that it is best that std.time includes all the things about
StopWatch and Date, TimeOfDay, Duration, Calender, ... etc.
However, that is not existing.
It is very agreeable to make it, but there is an insufficient point in it.
In particular, std.gregorian seem to be still insufficient about
functions and designs.
I think waiting for other modules to mature then to unify is good.

Just for the record, I don't think that the breaking changes should be
avoided.
...Because, I had been habituated.
It is exciting that old things are removed and cleaned up above all.

(2010/08/23 19:57), Andrei Alexandrescu wrote:
> Suggestions:
>
> 1. Put it inside std.date, though I agree that the name isn't suggestive (it does include time-related stuff as of today).
>
> 2. Create std.datetime from std.gregorian and stopwatch, and when we're done with the calendar stuff (did you guys see the recent related posts?) we deprecate std.date.
>
> 3. Create std.time, put stopwatch in it, and later move std.gregorian in std.date.
>
> Opinions? Shoo, what's your preference?
>
>
> Andrei
>
> On 08/23/2010 04:12 AM, Lars Tandle Kyllingstad wrote:
>> Me too. IMHO, it should either be part of a new std.datetime module together with the std.gregorian stuff, or std.stopwatch should be renamed to std.timer, std.timing, or something like that.
>>
>> -Lars
>>
>>
>>
>> On Sun, 2010-08-22 at 23:00 -0500, Andrei Alexandrescu wrote:
>>> Still unhappy about the addition of a module with a name narrow enough to preclude any further extensions.
>>>
>>> Andrei
>>>
>>> On 08/22/2010 10:11 AM, dsource.org wrote:
>>>> phobos commit, revision 1919
>>>>
>>>>
>>>> user: SHOO
>>>>
>>>> msg:
>>>> Added stopwatch module
>>>>
>>>> Added StopWatch, Ticks, systime, apptime, comparingBenchmark, measureTime.
>>>>
>>>> http://www.dsource.org/projects/phobos/changeset/1919
>>>>
>>>> _______________________________________________
>>>> phobos mailing list
>>>> phobos at puremagic.com
>>>> http://lists.puremagic.com/mailman/listinfo/phobos
>>> _______________________________________________
>>> phobos mailing list
>>> phobos at puremagic.com
>>> http://lists.puremagic.com/mailman/listinfo/phobos
>>
>>
>> _______________________________________________
>> phobos mailing list
>> phobos at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/phobos

(2010/08/24 20:30), SHOO wrote:
> Just for the record, I don't think that the breaking changes should be
> avoided.
> ...Because, I had been habituated.
> It is exciting that old things are removed and cleaned up above all.
>

Just in case, I don't like chaotic breaking change.
Therefore, we should become careful for making of std.time/std.datetime.

(2010/08/24 5:49), Walter Bright wrote:
> Deprecate std.perf, too.
>

How about std.openrj? I think it should mark deprecated, too.

SHOO wrote:
> (2010/08/24 5:49), Walter Bright wrote:
>> Deprecate std.perf, too.
>>
>
> How about std.openrj? I think it should mark deprecated, too.
>

Sure, it's just that std.perf overlaps functionality with your new code.

We should deprecate std.openrj just because the format itself seems de facto deprecated.  Check out this awesome list of Wikipedia hits for it:

http://en.wikipedia.org/wiki/Special:Search?search=openrj&sourceid=Mozilla-search

On Thu, Aug 26, 2010 at 12:56 PM, Walter Bright <walter at digitalmars.com>wrote:

>
>
> SHOO wrote:
>
>> (2010/08/24 5:49), Walter Bright wrote:
>>
>>> Deprecate std.perf, too.
>>>
>>>
>> How about std.openrj? I think it should mark deprecated, too.
>>
>>
> Sure, it's just that std.perf overlaps functionality with your new code.
>
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.puremagic.com/pipermail/phobos/attachments/20100826/372efb60/attachment.html>