researchcyclematrix

managing the research cycle matrix

related google drive sheets

all data is stored on google drive. There's two sheets the package interacts with:

• the research cycle matrix
• the submitted validations sheet

downloading

There are three essential objects, each pulled from google drive:

• rcm_download(): downloads the researchcyclematrix and "standardises" it, which means:

  • keeping only relevant columns
  • simplified column names
  • converting date formats etc.
  • standardise the status column values
  • few other things, see rcm_download() code
  • if you need the rows to match the actual google sheet, and/or the exact columns as they are online, use rcm_download(raw = T)

• subs_download() downloads the submissions sheet as is (converts dates and a few things)

• todo_download() downloads both sheets and creates a "todo" class data frame

  • filters only items "with HQ" in the RCM
  • keeps/creates relevant columns from both sheets such as "date.hqsubmission.actual", "submitter.emergency" (was emergency ticked?), "in.country.deadline"
  • "todo" objects are printed with the todo_next() function (in the code you'll see below the todo_next(), print.todo <- todo_next which does the trick)
  • the todo data frame only contains stuff that has been moved to the research cycle matrix with rcm_update_from_subs(). It also does not contain any items where the 'new.file.id' has not been changed to a real file id and moved to the rcm.

checking for inconsistencies

- obvious; see `rcm_check()` for details

changes to the research cycle matrix / google docs

• the submissions sheet is never changed from this package. By design that sheet is meant is "incoming" only (but maybe should break that principle eventually to add stuff to deal with new ids)
• the reserach cycle matrix status can be changed with rcm_set_to_validated(), rcm_set_to_withHQ() rcm_set_to_withField()
  • internally: these also set the date with rcm_set_validation_date(), rcm_set_withfield_date(), rcm_set_withHQ_date()
  • internally: rcm_set_to_validated uses rcm_set_hours_worked() as well. This writes to column BB just the number you put.
• the reserach cycle matrix comment with rcm_comment(). adds to the existing comment, and also appends a timestamp
• for status update, I mostly use todo_validate_next() which will set the status of the "top" item in a todo data frame to validated, remove that item from the todo data frame and return the updated todo item.. See the validation repository for details internally all these functions use the more generic rcm_change_value() function that takes a file id, column, and value. It will change the row of the first instance of the file.id so if they're not unique we gonna have trouble. If the column order changes we're also in trouble.

rcm dashboard

• rcm_dashboard() renders an rmarkdown document stored in ./inst/md_templates/

visualisations

• unstable & under construction, but in rcm_plots.R there's some approaches to start visualising the rcm