In download.R, the importRawTableToDb function is careful to catch errors during data download, but it assumes that once data is downloaded, the DB insertion step will always work.
This doesn't seem to hold for me, and I had a difficult time tracking this down because the resulting errors were very cryptic, like
#27 125.3 Error in `dplyr::bind_rows()`:
#27 125.3 ! Argument 10 must be a data frame or a named atomic vector.
#27 125.3 Backtrace:
#27 125.3 ▆
#27 125.3 1. └─dplyr::bind_rows(...)
#27 125.3 2. └─rlang::abort(glue("Argument {i} must be a data frame or a named atomic vector."))
#27 125.3 Execution halted
I eventually added a call to str(parResultList) after mclapply() runs, which gives me
In
download.R
, the importRawTableToDb function is careful to catch errors during data download, but it assumes that once data is downloaded, the DB insertion step will always work.This doesn't seem to hold for me, and I had a difficult time tracking this down because the resulting errors were very cryptic, like
I eventually added a call to
str(parResultList)
aftermclapply()
runs, which gives meI am still trying to figure out why this is happening, but it would be nice to have this caught and reported more gracefully.