ijlyttle / shinypod Goto Github PK

Hadley has moved the tbl_df class out of dplyr - which will be a short-term pain-in-the-neck, but is a great idea in the long run.

I think the only parts of dplyr we are using in shinypod are tbl_df formatting.

Along the lines of how devtools has all these use_foo() functions to generate templates.

Maybe one for the functional code, and another for the sidebar code?

proposed new function that would wrap an input

• add a (right-justified?) icon-link to the label
• link would bind an html fragment (built in the ui_misc list) to a modal window

note that buttons would have to be dealt-wtih differently

Update vignettes to reflect the fact that the shinyjs hide/show should be in the presentation layer

lubridate has with_tz()

As I wrote on email earlier, I think it would be cool if shinypod could make an initial guess of how the flat file is formatted. I've now put together a package that does that

https://github.com/lukas-rokka/readrGuess.

It simple runs trough different options of formatting the delimiter, decimal mark and grouping mark and chooses the combinations that creates a data frame that make most sense.

we want to allow the user to parse different formats

can we leave it out of the presentation layer, and put it into observers in the server?