Caching requests from the UN Comtrade API

Caching requests from the UN Comtrade API

There is four reasons why we might want to use caching for our requests to the UN Comtrade API:

The package offers a caching option, that is fairly easy to use.

The basics of caching đŸ’Ŋ

When you set the respective argument to TRUE, a directory named comtradr will be set up in your cache directory. This is determined by the tools::R_user_dir(which = 'cache') function. It has the default values for a cache as used by cachem::cache_disk.

See here:

and here:

#### Now
q <- ct_get_data(reporter = "USA", 
               partner = c("DEU", "FRA", "JPN", "MEX"), 
               flow_direction = "import",
               start_date = 2020,
               end_date = 2023,
               cache = TRUE) # <----- set this argument to TRUE

If you want to modify these parameters, you need to set environment variables. You can use the usethis::edit_r_environ(scope = 'project) function to set them only for your current project.

ℹī¸ Restart your R Session after setting the environment variables! 💡

The parameters are:

See here for the details on pruning and other functions of cachem:

Careful with outdated data! ⚠ī¸

The current default for the cache age is set to Inf, hence when you have requested data once, it will never downloaded again. For recent data, this can be a very bad idea, as the United Nations might update their data and you will only work with outdated data.

The recommended way is to set the COMTRADR_CACHE_MAX_AGE environment variable as described above. E.g.: to 60*60*24*365 for one year.

To find your cache and empty it by yourself, you can do:

## to delete all files in your cache
tools::R_user_dir('comtradr', which = 'cache') |> 


## to delete all files in your cache
tools::R_user_dir('comtradr_bulk', which = 'cache') |> 

You will have to restart comtradr to re-create the cache and be able to use the package.