To run the application:
library(shiny)
runApp("cran_explorer")The data is stored in the cran_explorer/ directory in two files: packages.rds and deps.rds. To update these files, run this from the top-level directory:
source("cran_explorer/utils.R")
json <- download_crandb()
crandb_data <- process_crandb_json(json)
write_rds(crandb_data$packages, "cran_explorer/packages.rds")
write_rds(crandb_data$deps, "cran_explorer/deps.rds")In the application, there is a "Refresh Data" button. It will download new data, but it will not save it over the .rds files. This is to make it easier to demonstrate that the update really occurs -- the saved .rds files should be from an earlier day than the demonstration. (In the future, it might be useful to make it possible to save to disk.)
The JSON data downloaded from crandb is about 140 MB, and is by default not saved to disk -- only the processed data is saved, as tibbles in .rds files. If you want to save the raw JSON for other forms of processing, you can do the following:
# Save the raw JSON
json <- download_crandb(outfile = "all.json")
# Save gzipped JSON
json <- download_crandb(outfile = "all.json.gz")
# Download only the first 1000 packages from crandb
json <- download_crandb(outfile = "all.json.gz", limit = 1000)The process_crandb_json() function can take the name of a JSON (or gzipped JSON) file to process:
crandb_data <- process_crandb_json(filename = "all.json.gz")