R help - Dec 2009 - help in merging

Hi All,
   I want to "merge" two datasets by column "ID" and I
don't want the result to
   be sorted by "ID". I am doing the following:
   > z = merge(x, y, by = "ID", sort=F)
   The result is not sorted by "ID". But (as oppose to what I
expected) it is
   not even in the original order of either "x" or "y".
   Can somebody tell what to do if I wanted it to be in the original order of
   x.

   P.S.: As my dataset is very huge and I couldn't find the right subset of
the
   data which explains the above problem, so I can't attach it at the
moment.
   If anybody knows the answer, please reply; or else I will try to get the
   right subset.

   Thanks in advance

   Utkarsh

Hi there,

You can add a order column on x or y and after use this field to order z.
Like

z<-z[order(z$orderfield),]

To generate a order on x or y you can do something like

x$xorder<-1:nrow(x)

cheers

milton
On Thu, Dec 24, 2009 at 2:26 PM,
<utkarsh.singhal@global-analytics.com>wrote:
>
>   Hi All,
>   I want to "merge" two datasets by column "ID" and I
don't want the result
> to
>   be sorted by "ID". I am doing the following:
>   > z = merge(x, y, by = "ID", sort=F)
>   The result is not sorted by "ID". But (as oppose to what I
expected) it
> is
>   not even in the original order of either "x" or "y".
>   Can somebody tell what to do if I wanted it to be in the original order
> of
>   x.
>
>   P.S.: As my dataset is very huge and I couldn't find the right subset
of
> the
>   data which explains the above problem, so I can't attach it at the
> moment.
>   If anybody knows the answer, please reply; or else I will try to get the
>   right subset.
>
>   Thanks in advance
>
>   Utkarsh
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
>
http://www.R-project.org/posting-guide.html<http://www.r-project.org/posting-guide.html>
> and provide commented, minimal, self-contained, reproducible code.
>
	[[alternative HTML version deleted]]

Thanks, but I kind of new this way already and it doesn't seem an optimal
   thing to do.



   What  I  was  looking  for  is to pass some argument in 'merge'
itself
   which doesn't  change  the  ordering of  'x'. Or, more than that,
I am
   interested in knowing that why is it changing the order or what order is it
   taking when I specified sort=F.



   Regards

   Utkarsh





   -------- Original Message --------
   Subject: Re: [R] help in merging
   From: milton ruser <milton.ruser at gmail.com>
   Date: Fri, December 25, 2009 1:38 am
   To: utkarsh.singhal at global-analytics.com
   Cc: r-help at r-project.org
   Hi there,

   You can add a order column on x or y and after use this field to order z.
   Like

   z<-z[order(z$orderfield),]

   To generate a order on x or y you can do something like

   x$xorder<-1:nrow(x)

   cheers

   milton
   On Thu, Dec 24, 2009 at 2:26 PM, <[1]utkarsh.singhal at
global-analytics.com>
   wrote:

       Hi All,
       I want to "merge" two datasets by column "ID" and I
don't want the
     result to
       be sorted by "ID". I am doing the following:
       > z = merge(x, y, by = "ID", sort=F)
       The result is not sorted by "ID". But (as oppose to what I
expected) it
     is
       not even in the original order of either "x" or "y".
       Can somebody tell what to do if I wanted it to be in the original order
     of
       x.
       P.S.: As my dataset is very huge and I couldn't find the right subset
of
     the
       data which explains the above problem, so I can't attach it at the
     moment.
       If anybody knows the answer, please reply; or else I will try to get the
       right subset.
       Thanks in advance
       Utkarsh
     ______________________________________________
     [2]R-help at r-project.org mailing list
     [3]https://stat.ethz.ch/mailman/listinfo/r-help
     PLEASE do read the posting guide
     [4]http://www.R-project.org/posting-guide.html
     and provide commented, minimal, self-contained, reproducible code.

References

   1. mailto:utkarsh.singhal at global-analytics.com
   2. mailto:R-help at r-project.org
   3. https://stat.ethz.ch/mailman/listinfo/r-help
   4. http://www.r-project.org/posting-guide.html

Not sure if its guaranteed but this sqlite join does seem to preserve
the order of the first data frame.
> library(sqldf)
> BOD
  Time demand
1    1    8.3
2    2   10.3
3    3   19.0
4    4   16.0
5    5   15.6
6    7   19.8
> BODrev <- BOD[6:1,]; BODrev
  Time demand
6    7   19.8
5    5   15.6
4    4   16.0
3    3   19.0
2    2   10.3
1    1    8.3
> sqldf("select * from BODrev, BOD using(Time)")
  Time demand demand
1    7   19.8   19.8
2    5   15.6   15.6
3    4   16.0   16.0
4    3   19.0   19.0
5    2   10.3   10.3
6    1    8.3    8.3

See home page at: http://sqldf.googlecode.com


On Thu, Dec 24, 2009 at 2:26 PM,  <utkarsh.singhal at
global-analytics.com> wrote:
>
> ? Hi All,
> ? I want to "merge" two datasets by column "ID" and I
don't want the result to
> ? be sorted by "ID". I am doing the following:
> ? > z = merge(x, y, by = "ID", sort=F)
> ? The result is not sorted by "ID". But (as oppose to what I
expected) it is
> ? not even in the original order of either "x" or "y".
> ? Can somebody tell what to do if I wanted it to be in the original order
of
> ? x.
>
> ? P.S.: As my dataset is very huge and I couldn't find the right subset
of the
> ? data which explains the above problem, so I can't attach it at the
moment.
> ? If anybody knows the answer, please reply; or else I will try to get the
> ? right subset.
>
> ? Thanks in advance
>
> ? Utkarsh
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>