Export to R Plug-in Shell/Proof of Concept
Overview
ZSL, in coordination with the SMART Partnership, wishes to engage in development of a plug-in for SMART that will provide an interface with R, as well as expand community developers’ ability to create SMART plug-ins.
Goals
There are a number of goals for this project:
1. Build a useful interface to R that users can leverage to build more complex analyses and reports in SMART.
2. The plug-in should be sustainable; i.e. easily maintainable for future SMART versions and continue to be useful as SMART evolves and grows.
3. Test and prove the process of community developers building a SMART plug-in.
4. Provide a bridge between Refractions and the community developers to smooth the learning curve of developing a SMART plug-in.
Plan
As discussed during our meeting Nov 15th, 2016, Refractions will:
1. Build the shell of a new SMART plug-in. This would contain all the files necessary to interact with SMART and other plug-ins. The files would be “stubs” - mostly empty files that have the mandatory function descriptions and definitions, but not any code to process anything.
As well, it would contain basic empty database upgrade scripts that would then have to be filled out with the database design and updated as the development was continued.
2. Create the general outline of the database design for the key features of the plug-in. As the GUI and full features are developed, there may be some additional changes for the community developers to make; but the key components that interact with other plug-ins, Refractions would help design.
3. Provide a set number of hours of development support to be requested whenever the community developers have questions or are looking for advice.
ZSL, in coordination with the SMART Partnership, wishes to engage in development of a plug-in for SMART that will provide an interface with R, as well as expand community developers’ ability to create SMART plug-ins.
Goals
There are a number of goals for this project:
1. Build a useful interface to R that users can leverage to build more complex analyses and reports in SMART.
2. The plug-in should be sustainable; i.e. easily maintainable for future SMART versions and continue to be useful as SMART evolves and grows.
3. Test and prove the process of community developers building a SMART plug-in.
4. Provide a bridge between Refractions and the community developers to smooth the learning curve of developing a SMART plug-in.
Plan
As discussed during our meeting Nov 15th, 2016, Refractions will:
1. Build the shell of a new SMART plug-in. This would contain all the files necessary to interact with SMART and other plug-ins. The files would be “stubs” - mostly empty files that have the mandatory function descriptions and definitions, but not any code to process anything.
As well, it would contain basic empty database upgrade scripts that would then have to be filled out with the database design and updated as the development was continued.
2. Create the general outline of the database design for the key features of the plug-in. As the GUI and full features are developed, there may be some additional changes for the community developers to make; but the key components that interact with other plug-ins, Refractions would help design.
3. Provide a set number of hours of development support to be requested whenever the community developers have questions or are looking for advice.
Leave a comment
The eventually goal allows the user to:
1. Add new R-Script into SMART, pick the file from hard Drive, give it a name, organize them in folders in SMART.
2. For any saved SMART query, link a particular R-script to be run with that Query results format/columns
3. A button or export option for Queries, "Run R-Analysis".
4. That button would export the csv query results to a file and run R with a command line saying run the linked script X with the data we just saved.
5. Ideally, then somehow load the graphs or table outputs of R back into SMART and allow adding them to Reports somehow.
1. Add new R-Script into SMART, pick the file from hard Drive, give it a name, organize them in folders in SMART.
2. For any saved SMART query, link a particular R-script to be run with that Query results format/columns
3. A button or export option for Queries, "Run R-Analysis".
4. That button would export the csv query results to a file and run R with a command line saying run the linked script X with the data we just saved.
5. Ideally, then somehow load the graphs or table outputs of R back into SMART and allow adding them to Reports somehow.