On Thu, Jan 14, 2010 at 2:15 PM, Rex C. Eastbourne <rex.eastbou...@gmail.com
> wrote:

> I am encountering problems using the png() function to save around 20
> charts to separate files.
>
> My script is conceptually structured as follows:
>
> ### <script>
> png("Image %03d.png")
>
> # the following are calls to user-defined functions I wrote that call plot,
> barchart, etc. with special arguments.
>  myPlot("a")
> myPlot("b")
> myPlot("c")
>
> myBarchart("a")
> myBarchart("b")
> myBarchart("c")
>
>  myBarchart2("a")
> myBarchart2("b")
> myBarchart2("c")
>
> dev.off()
> ### </script>
>
> When I run the code a few lines at a time it generates the images
> correctly. However, when I execute all lines of the script at once (Ctrl+A,
> Ctrl+R in the Rgui editor), the output doesn't get generated right. Most
> notably:
>
> - Only one plot created by myBarchart (or myBarchart2) gets saved, instead
> of all 3
> - Some images are blank
>
> I'm having trouble isolating the problem to one line of code.
>
> I suspect there is some fundamental paradigm I'm not aware of about how R
> devices and images work, perhaps relating to the importance of the order in
> which the calls to png, dev.off(), etc. are executed, the required state of
> my workspace before I run the code (e.g. whether the previous plot window
> should be open or closed), or maybe even time dependency (do I need to
> insert timed pauses so that my diagram has time to render before saving to a
> file?).
>
> Does this ring a bell to anyone? I read the R documentation on devices and
> png(), and searched the mailing list archives and the web, but haven't
> figured out how to get this working. Any help would be appreciated.
>
> Thanks,
>
> Rex
>

I'm on Windows 7 and R 2.9.2, BTW.

        [[alternative HTML version deleted]]

______________________________________________
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
and provide commented, minimal, self-contained, reproducible code.

Reply via email to