Re: A new 'transpose' coreutil

[Ryan]

On Thu, Dec 17, 2015 at 11:38 AM, Pádraig Brady <address@hidden> wrote:
> On 17/12/15 16:25, Assaf Gordon wrote:
>> Hello,
>>
>> On 12/17/2015 06:00 AM, Pádraig Brady wrote:
>>> On 17/12/15 05:06, Ryan wrote:
>>>> Hi everybody,
>>>>
>>>> I looked through the rejected feature requests and did not find this.
>>>> I and others I know have often wished for an efficient 'transpose'
>>>
>>> Interesting.
>>
>> GNU Datamash ( http://www.gnu.org/software/datamash/ ) can transpose files 
>> from the command line:
>>
>>      $ cat 1.txt
>>      Genes   NOX1    DcP     HH
>>      Sample  A1      A2      B3
>>      Counts  514     542     490
>>
>>      $ datamash transpose < 1.txt
>>      Genes   Sample  Counts
>>      NOX1    A1      514
>>      DcP     A2      542
>>      HH      B3      490
>>
>> In addition to performing other various statistical and text-related 
>> operations (in the coming version: also cross-tabulation, a.k.a 
>> pivot-tables).
>
> Oh nice,
> This project is a better match for any such changes.

All right, I'll take a look at GNU Datamash. Thanks for considering it.

>
> BTW I did search for "GNU transpose utility" in various
> combinations and didn't notice this.
> I suggest to mention it at
> http://unix.stackexchange.com/questions/88142/command-to-transpose-swap-rows-and-columns-of-a-text-file
> http://stackoverflow.com/questions/1729824/transpose-a-file-in-bash
> http://unix.stackexchange.com/questions/79642/transposing-rows-and-columns
>
> Also maybe apply some SEO to the function names at:
> http://www.gnu.org/software/datamash/manual/datamash.html
>
> thanks!
> Pádraig