[datatable-help] by=.EACHI and related - please check ok

Frank Erickson FErickson at psu.edu
Mon Mar 17 23:30:05 CET 2014


Cool stuff! I'd like to see eddi's second extension, too. It would be nice
if ".EACHI" were shorter to type, though, like ".i" or ".is" or something.
--Frank

p.s. Hmm, got stopped by the mailer daemon the first time I sent this:


On Mon, Mar 17, 2014 at 11:26 AM, Frank Erickson <FErickson at psu.edu> wrote:

> Cool stuff! I'd like to see eddi's second extension, too. It would be nice
> if ".EACHI" were shorter to type, though, like ".i" or ".is" or something.
> --Frank
>
>
> On Thu, Mar 13, 2014 at 10:22 AM, Eduard Antonyan <
> eduard.antonyan at gmail.com> wrote:
>
>> Looks great!!
>>
>> I think there were two extensions for this floating around, which are now
>> possible to have - one (I think suggested by Gabor?) was to have by=.EACHI
>> work with other types of i-expressions - simplest example would be d[, j,
>> by=.EACHI] doing the same as d[, j, by = 1:nrow(d)], and another one was to
>> be able to combine .EACHI with other by's, i.e. have smth like d[i, j, by =
>> list(.EACHI, somecol)] (this one is somewhat more exotic and doesn't have
>> an existing analogue afaik). Not sure if an FR for these exists.
>>
>>
>> On Thu, Mar 13, 2014 at 7:45 AM, Matt Dowle <mdowle at mdowle.plus.com>wrote:
>>
>>>
>>> Not yet,  but yes NEWS says :
>>>
>>>
>>>     A "classic" option to restore the previous default behaviour is to be discussed and confirmed.
>>>
>>> Maybe :
>>>
>>>   options(datatable.bywithoutby = TRUE)
>>>
>>> is all that's needed.  Will do.
>>>
>>> Matt
>>>
>>>
>>>
>>> On 13/03/14 12:21, carrieromichele wrote:
>>>
>>> Cool! But I still remain a fan of the "by-without-by" :-)
>>>
>>> Is there a option to make the future stable version 1.9.4 to behave
>>> exactly as the 1.8.10? (same script same results)
>>>
>>> Thanks,
>>>
>>> Michele
>>>  On 13 Mar 2014 11:36, "Matt Dowle" <mdowle at mdowle.plus.com> wrote:
>>>
>>>>
>>>> Dear all,
>>>>
>>>> by=.EACHI is now implemented and available in v1.9.3 from R-Forge.
>>>>
>>>> Please take a look at NEWS and see what you think :
>>>>
>>>> https://r-forge.r-project.org/scm/viewvc.php/pkg/NEWS?view=markup&root=datatable
>>>>
>>>> Quite a few related bugs and feature requests get resolved by this,
>>>> still going through them updating NEWS and adding tests.  Thanks to all.
>>>>
>>>> The changes are very much up for debate and change at this stage.
>>>>
>>>> Matt
>>>>
>>>>
>>>> _______________________________________________
>>>> datatable-help mailing list
>>>> datatable-help at lists.r-forge.r-project.org
>>>>
>>>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/datatable-help
>>>>
>>>
>>>
>>> _______________________________________________
>>> datatable-help mailing list
>>> datatable-help at lists.r-forge.r-project.org
>>>
>>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/datatable-help
>>>
>>
>>
>> _______________________________________________
>> datatable-help mailing list
>> datatable-help at lists.r-forge.r-project.org
>>
>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/datatable-help
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/datatable-help/attachments/20140317/1875c2ba/attachment.html>


More information about the datatable-help mailing list