R help - Feb 2010 - Automate generation of multiple reports using odfWeave

Dear R and odfWeave users,

I am looking for a way to automate generation of many reports using
odfWeave. All reports would use the same input ODT file, the only difference
would be in the name of the dataset which will be analyzed in any particular
report. Right now, the name of the dataset is hardcoded in the first code
chuck in the input file:

<<01 get data, echo=TRUE>>
station = '123'                  # name of the station dataset to be
analyzed in this report
data = get_data(station)    # get data, e.g. from a database
@

This is far from ideal, as it requires a separate input file (ODT) for every
input station, a huge duplication.

Are there ways to streamline this? I am looking for a way to have only one
input ODT file (much easier to maintain one file than many). I am thinking
that the input file could be pre-processed to include the station parameter
and saved as an intermediate ODT, which would then be put through
odfWeave()? Does anyone know of a good way to edit the OO.org ODT file to
put "station = '...'" into the first code chuck?
Is there any other way to do that?

Thank you,
Aleksey

	[[alternative HTML version deleted]]

I tend to do it the other way around.

Hard-code the station into the ODT file as "thisstation".

Then, in R, do something like this:

allstations <- c("station1", "station2",
"station3")

for (i in allstations) {
   thisstation <- i
   odfWeave("inputfile.odt", paste("output-", i,
".odt", sep=""))
}
rm(thisstation)

That way you don't have to have a bunch of files.

Sarah

On Fri, Feb 26, 2010 at 3:43 PM, Aleksey Naumov <anaumov at gmail.com>
wrote:
> Dear R and odfWeave users,
>
> I am looking for a way to automate generation of many reports using
> odfWeave. All reports would use the same input ODT file, the only
difference
> would be in the name of the dataset which will be analyzed in any
particular
> report. Right now, the name of the dataset is hardcoded in the first code
> chuck in the input file:
>
> <<01 get data, echo=TRUE>>
> station = '123' ? ? ? ? ? ? ? ? ?# name of the station dataset to
be
> analyzed in this report
> data = get_data(station) ? ?# get data, e.g. from a database
> @
>
> This is far from ideal, as it requires a separate input file (ODT) for
every
> input station, a huge duplication.
>
> Are there ways to streamline this? I am looking for a way to have only one
> input ODT file (much easier to maintain one file than many). I am thinking
> that the input file could be pre-processed to include the station parameter
> and saved as an intermediate ODT, which would then be put through
> odfWeave()? Does anyone know of a good way to edit the OO.org ODT file to
> put "station = '...'" into the first code chuck?
> Is there any other way to do that?
>
> Thank you,
> Aleksey
-- 
Sarah Goslee
http://www.functionaldiversity.org