Option to disable caching

[lpiep]

For data retrieval functions such as lcd, isd, etc, it would be helpful to include an option not to cache a cleaned version of the data, rather than having to delete the cached files manually with lcd_cache$delete and similar functions. The isd function has a cleanup option already, although this only removes the raw .gz files.

My use case is that I'm using rnoaa to periodically download and process LCD data, and I want to always be getting the data as it appears on NOAA's servers.

Let me know if this seems like a reasonable feature request. Happy to help with implementation.

[sckott]

have you seen https://docs.ropensci.org/rnoaa/reference/rnoaa_caching.html ?

[lpiep]

Thanks @sckott, I did read through that. I see I can clear the cache when I'm done (e.g. with lcd_cache$delete_all(), but it's not obvious to me how to disable caching entirely. That is, prevent cache files from being stored in the home directory at all.

[sckott]

force=TRUE ?

[lpiep]

For lcd, at least, there isn't a force option documented, and using it doesn't appear to prevent the cache file from being used.

> rnoaa::lcd_cache$delete_all()
> x <- rnoaa::lcd('72027763843', 2022)
> x2 <- rnoaa::lcd('72027763843', 2022, force = TRUE)
using cached file: ~/.cache/R/noaa_lcd/2022_72027763843.csv
date created (size, mb): 2022-02-11 04:59:29 (1.705)

Even on isd, which does have aforce argument, it doesn't stop the rds file from being saved to the cache, it only prevents it from being used.

> rnoaa::isd_cache$delete_all()
> x <- rnoaa::isd('720277', '63843', 2022, force = TRUE)
<path>~/.cache/R/noaa_isd/720277-63843-2022.gz

> list.files(rnoaa::isd_cache$cache_path_get())
[1] "720277-63843-2022.rds"

I suppose that in either case, explicitly clearing the cache after downloading is sufficient for my purposes. I thought though that it would be nice to have finer control over what rnoaa saves into the home directory, especially when the user specifies not to use those files.

[sckott]

@djhocking i'll let you take it from here, just tried to chime in while i had a few min (@lpiep i'm the former maintainer)