R help - Feb 2006 - Adding header lines to a dataframe that is exported using write.csv

I would like to export a dataframe to a .csv using:
>write.csv(dataframe,"dataframe.csv")
but I need to add four "header" lines to the csv that are not part of
the dataframe (which itself has a line of column headers).

The difficulty (for me, at least!) lies in the requirement that
certain elements of the header (X, Y and the number of "Q"s - please
see example below) must be defined based on the number of rows and
columns in the dataframe, which vary depending on the input file.

Here's what the 3 .csv header lines should look like, followed by a
number of dataframe rows (i.e., these lines are not R code, but are
what R will produce).

X, plots ,,,, #where X=number of rows in the dataframe
Y, species,,,, #where Y=number of columns in the dataframe
,Q,Q,Q,Q,Q #where the number of Qs=the number of columns in the dataframe

Those 3 .csv header lines would be followed by dataframe, which
consists of one row containing column headers and X "data" rows:

,spec1,spec2,spec3,sp3c4,spec5 #these are the dataframe's column headers
plot1,15.84,0,0,792,7 #this is an example "data" row

In case the above is unclear, I have also attached a small .csv as an
example of what the output should look like.

Thank you. Mark

On 2/25/2006 1:20 PM, Mark wrote:
> I would like to export a dataframe to a .csv using:
> 
>> write.csv(dataframe,"dataframe.csv")
> 
> but I need to add four "header" lines to the csv that are not
part of
> the dataframe (which itself has a line of column headers).
Open a connection and write the header to it first, then write the
dataframe.  For example,

df <- data.frame(a=1:5,b=6:10)

f <- file("dataframe.csv", "w")

writeLines(paste(c(nrow(df), ncol(df)), c("plots",
"species"), ",,,",
sep=","),f)
writeLines(paste(rep(",Q", ncol(df)), collapse=""),f)
write.csv(df, f)

close(f)

Duncan Murdoch
> 
> The difficulty (for me, at least!) lies in the requirement that
> certain elements of the header (X, Y and the number of "Q"s -
please
> see example below) must be defined based on the number of rows and
> columns in the dataframe, which vary depending on the input file.
> 
> Here's what the 3 .csv header lines should look like, followed by a
> number of dataframe rows (i.e., these lines are not R code, but are
> what R will produce).
> 
> X, plots ,,,, #where X=number of rows in the dataframe
> Y, species,,,, #where Y=number of columns in the dataframe
> ,Q,Q,Q,Q,Q #where the number of Qs=the number of columns in the dataframe
> 
> Those 3 .csv header lines would be followed by dataframe, which
> consists of one row containing column headers and X "data" rows:
> 
> ,spec1,spec2,spec3,sp3c4,spec5 #these are the dataframe's column
headers
> plot1,15.84,0,0,792,7 #this is an example "data" row
> 
> In case the above is unclear, I have also attached a small .csv as an
> example of what the output should look like.
> 
> Thank you. Mark
> 
> 
> ------------------------------------------------------------------------
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html

Just a tips: When you add headers to tabulate files like yours, it is
convenient to start each header line with a '#' (like an R comment),
because then read.table() will not complain about the header lines. 
It is easy to strip the '#' off the header lines, i.e.
grep("^#", "",
hlines) before further parsing.

/Henrik

On 2/25/06, Mark <mtb954 at gmail.com> wrote:
> I would like to export a dataframe to a .csv using:
>
> >write.csv(dataframe,"dataframe.csv")
>
> but I need to add four "header" lines to the csv that are not
part of
> the dataframe (which itself has a line of column headers).
>
> The difficulty (for me, at least!) lies in the requirement that
> certain elements of the header (X, Y and the number of "Q"s -
please
> see example below) must be defined based on the number of rows and
> columns in the dataframe, which vary depending on the input file.
>
> Here's what the 3 .csv header lines should look like, followed by a
> number of dataframe rows (i.e., these lines are not R code, but are
> what R will produce).
>
> X, plots ,,,, #where X=number of rows in the dataframe
> Y, species,,,, #where Y=number of columns in the dataframe
> ,Q,Q,Q,Q,Q #where the number of Qs=the number of columns in the dataframe
>
> Those 3 .csv header lines would be followed by dataframe, which
> consists of one row containing column headers and X "data" rows:
>
> ,spec1,spec2,spec3,sp3c4,spec5 #these are the dataframe's column
headers
> plot1,15.84,0,0,792,7 #this is an example "data" row
>
> In case the above is unclear, I have also attached a small .csv as an
> example of what the output should look like.
>
> Thank you. Mark
>
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
>
>

--
Henrik Bengtsson
Mobile: +46 708 909208 (+1h UTC)