Radiant fails on loading R Data (".rda") file about radiant

[vnijs]

You should pass-in one dataframe in an rda-file. Your file has multiple and you would need to merge them together into one data.frame to load it into Radiant. Alternatively, you could create a separate file for each data-frame, load them all into Radiant, and merge them together.

The fact that only one data.frame in an rda file can be loaded is not explicitly mentioned in the help-file. I will add a comment about this soon and have the code check the structure of the .rda file to ensure it is compatible with Radiant.

[abnova]

Thank you for very fast reply. Now the Radiant's specifics about loading data is clear. But, if you support the scenario of loading data frames into Radiant separately and then merging them in the application, wouldn't it make sense for Radiant to support R data files with multiple data frames? It would be nice, if Radiant would detect such cases and inform user about the possibility of merging (especially, since the functionality is already supported by Radiant), requesting merge parameters and merge the R data file's data frames, based on the user's input. Just a suggestion of a nice-to-have feature.

One more issue, which I believe is still valid. It seems to me that Radiant doesn't "like" multi-word column names - it just produced the following message (when I tried to use Visualize tab with a single-data-frame dataset and a column named Project Age):

Error: <text>: 1:9: unexpected symbol
1: Project Age
________^

[vnijs]

The fact that R-data files allow the user to put multiple object into one file complicates loading the data appropriately. This is why Hadley (of ggplot2, dplyr, etc.) prefer to use RDS file (http://www.fromthebottomoftheheap.net/2012/04/01/saving-and-loading-r-objects/). You can easily load multiple files into Radiant so, honestly, I doubt I will add this feature. However, I will make sure that a user will get a clear message if a data file contains > 1 data.frame. Thanks for bringing this to my attention.

[abnova]

You're welcome! No problem, a user can definitely merge multiple data frames appropriately into a summary data frame, in order to use it for analysis in Radiant. Still hope to hear from you about Radiant's handling of multi-word column names, as I mentioned in my previous comment. Please let me know, if you want me to create a new issue for this.