[R] Set window title for plot on any OS
Simon Zehnder
szehnder at uni-bonn.de
Sat Jul 13 15:45:58 CEST 2013
Hi Duncan,
thank you very much for your advice! That makes it all work.
I check in addition for a "title" argument in the device via
if (any(names(getOption("device")) == TRUE)) {
dev.new(title = "title")
}
Thanks again!
Simon
On Jul 13, 2013, at 2:40 PM, Duncan Murdoch <murdoch.duncan at gmail.com> wrote:
> On 13-07-13 1:33 PM, Simon Zehnder wrote:
>> Dear R-Users,
>>
>> I am writing a package using S4 classes. In the generic method "plot" I want to set the title for the plotting window as I will have several windows and window titles help the users to distinguish the graphics without putting a title into the plot itself (this can be done by users whenever they want)
>>
>> So I created a helper function .setDeviceTitle which I called after the plot has been done:
>>
>> ".setDeviceTitle" <- function(title = "title", dev = dev.cur()) {
>> dev <- names(dev)[1]
>>
>> ## check for OS ##
>> if (dev == "windows") {
>> windows(title = title)
>> } else if (dev == "X11") {
>> X11(title = title)
>> } else {
>> quartz(title = title)
>> }
>> }
>>
>> The result is a new device with the title in addition to the old. Is it possible to give a window a title after the plot has been done? If not: Before I plot the device I cannot know what device it will be, so I thought about a check via capabilities():
>>
>> if (any(names(capabilities()) == "X11")) {
>> X11(title = title)
>> }
>> else if (any(names(capabilities)) == "windows") {
>> windows(title = title)
>> } else {
>> quartz(title = title)
>> }
>>
>> I want to have a safe method, which works on each OS R can run. How would you solve the problem?
>
> Use dev.new() rather than picking a particular device. If all the possible devices support the "title" argument, then
>
> dev.new(title=title)
>
> will be fine. If you might need more customization (or want to protect against a user who chooses a device that doesn't have "title" as an argument), use getOption("device") to examine the call that will be used.
>
> Duncan Murdoch
>
More information about the R-help
mailing list