Future ideas

This content was originally moved from our GSoC 2013 project idea page to rather mature the ideas there and continue brainstorming about possible future development here:

• add new input types to rapport like date, color specification, URL etc.
• create new statistical templates to be shipped with the rapport package like:
  • two sample t-test,
  • hierarchical cluster analysis,
  • principal component and factor analysis,
  • multidimensional scaling,
  • three-way crosstables,
  • graphing-only templates with numerous input fields etc.
• extend pander S3 method to support more R objects. A nice list can be found in a recent R-bloggers post.
• let pandoc.table emphasize or format conditionally some cells in a table. See a proposed overview on SO.
• create a helper function to draw significance starts in tables,
• tweak unify images feature in pander for base, lattice and ggplot2 plots e.g. by:
  • splitting too long titles and axis labels in plots,
  • supervise dirty hacks in applying global style to "base" graphics plots,
  • add support for all graphics functions and e.g. plotrix and other packages.
• update exported documents' templates in pander and rapport e.g. with dropping skeleton and custom font family to something with even less footprint to decrease package size
• tweak messages, warning and error reporting in exported documents
• add more support for PHP Markdown Extra beside Pandoc for more seamless knitr integration
• create a knitr hook so that chunks in knitted documents would result in markdown format automatically
• add more tests to pander and rapport
• think about and implement method based reporting based on Jeroen's suggestion. In the long run we would like to have a summary like function that can take any R object as its only one argument and optionally bring up a GUI form (widget) based on input parameters for interactive R object inspection (maybe with shiny)
• tweak documentation and add more use-cases
• revise RAppArmor integration and suggest secured AppArmor profiles
• review and profile pandoc.table for performance issues
• revise and test the automatic caching engine in more details
• add new pander options like in class-based p() function