Back on CRAN?

[hrbrmstr]

Welcome @noelle-gray!

I honestly do not remember this: https://cran-archive.r-project.org/web/checks/2021/2021-06-28_check_results_wayback.html

It is likely due to that being a time when burnout fully kicked in.

It shld be straightforward to debug. When you tap the collab "accept", drop a note here and we can make a plan.

[noelle-gray]

Hi! I've done some basic debugging on this for the Linux build and I think that this build failed in CRAN for:

Result: WARN
    Missing or unexported objects:
     'checkpoint::getValidSnapshots' 'checkpoint::mranUrl'

The R CMD check revealed no serious issues, so long as we have curl, libcurl4-openssl-dev, and libxml2-dev installed as dependencies.

It's possible these builds failed previously because necessary scrape/vignette code running in R CMD check:

> nasa <- ia_scrape("collection:nasa", count=100L)
Error in curl::curl_fetch_memory(url, handle = handle) :
  Timeout was reached: [archive.org] Connection timeout after 11222ms.

Pretty sure this can be fixed inside the curl::curl_fetch_memory if there is a greater timeout, which then builds the LaTeX doc that causes wayback-Ex.pdf to fail. But, no code critically fails.

Internet Archive changed their timeout on the Python package ia to 24 seconds in as documented here.

My suggestion is to fix the docs in this section:

  httr::GET(
    url = "https://archive.org/services/search/v1/scrape",
    httr::accept_json(),
    httr::user_agent("R wayback package; <https://gitlab.com/hrbrmstr/wayback>"),
    query = params,
    httr::timeout(24)
  ) -> res

Or similarly so to avoid build errors and match the matching Python function?