[TxMt] Re: Send variable to R app

[Tim Bates]

If you wrote it as

summary(lm(mpg ~ drat  + hp, data = mtcars))

then I think it is more doable.

the command could then run with just 1 carat specifying the variable, with the content of the line allowing deduction of mtcars as the data)
Return data$var to the R console is then straightforward


Lets say I have this code:

>> summary(lm(mpg ~ drat  + hp, mtcars))
>> 
>> How can select 'mtcars’ and ‘hp’ simultaneously, press a keyboard shortcut and run this code in R app:
>> 
>> mtcars$hp
> 
> unfortunately up to now it would be a stony way to achieve this - due to TM's "alpha" status I think. The point here is that if you have more than one selection the TM variable "TM_SELECTION" holds "only" the information about the line number-start of selection:length of selection. In addition to that - even if all selections are in the SAME line - TM 2.0 doesn't set the TM variable "TM_CURRENT_LINE". This would mean if you want to get the actual content of the selections you have to calculate the contents by yourself. This could be done maybe by:
> - set the tmCommand's "Input:" to "Document"
> - get the entire document content
> - iterate through each selection in "TM_SELECTION" by
> - - get line number -> get the content of the line based on the document's content
> - - get start and length of your selection
> - - get the substring (content) for each selection
> - concatenate the selections by "$" to cmd
> - send cmd to R.app